OmniRoute/docs/i18n/th/CHANGELOG.md

Changelog (ไทย)

🌐 Languages: 🇺🇸 English · 🇸🇦 ar · 🇧🇬 bg · 🇧🇩 bn · 🇨🇿 cs · 🇩🇰 da · 🇩🇪 de · 🇪🇸 es · 🇮🇷 fa · 🇫🇮 fi · 🇫🇷 fr · 🇮🇳 gu · 🇮🇱 he · 🇮🇳 hi · 🇭🇺 hu · 🇮🇩 id · 🇮🇹 it · 🇯🇵 ja · 🇰🇷 ko · 🇮🇳 mr · 🇲🇾 ms · 🇳🇱 nl · 🇳🇴 no · 🇵🇭 phi · 🇵🇱 pl · 🇵🇹 pt · 🇧🇷 pt-BR · 🇷🇴 ro · 🇷🇺 ru · 🇸🇰 sk · 🇸🇪 sv · 🇰🇪 sw · 🇮🇳 ta · 🇮🇳 te · 🇹🇭 th · 🇹🇷 tr · 🇺🇦 uk-UA · 🇵🇰 ur · 🇻🇳 vi · 🇨🇳 zh-CN

---

[3.8.31] — 2026-06-20

[3.8.49] — TBD

---

[3.8.48] — 2026-07-13

⚠️ Hotfix release. The published npm package for 3.8.47 crashed on every boot (#7065) and was deprecated — 3.8.48 is the first installable release of the v3.8.47 cycle, so everything listed under [3.8.47] below ships here.

🐛 Bug Fixes

• fix(build): ship dist/head-response-guard.cjs in the npm tarball — the prepublish prune allowlist lacked it, so every omniroute boot of the published 3.8.47 crashed with ERR_MODULE_NOT_FOUND (3rd occurrence of this class after tls-options/3.8.41); now allowlisted, enforced by check:pack-artifact, and guarded by a closure test that derives every server-ws.mjs sibling import (#7065, #7040)
• fix(build): Electron Windows packaging — the better-sqlite3 Electron-ABI rebuild now spawns npx.cmd through a shell (Node's CVE-2024-27980 hardening made the shell-less spawn fail with status null on Windows runners, breaking the v3.8.47 desktop build)
• fix(ci): Sonar quality gate zeroed on new code — the coverage lcov now reaches the scanner at coverage/lcov.info (it read 0% on every scan), the async isCloudEnabled() gate in the Kiro auto-import route is awaited (cloud sync ran even when disabled), the dead structuredClone fallback in the reasoning-split clone is a real JSON fallback, the codex executor handles the async reader.cancel() rejection, deterministic localeCompare sorts, a path-traversal guard in classify-pr-changes.mjs, and the Docker better-sqlite3 rebuild uses npm's bundled node-gyp instead of npx --yes
• chore(ci): the Sonar quality gate is informational (sonar.qualitygate.wait=false) while the org's SonarCloud plan cannot associate the tuned "OmniRoute way" gate (coverage ≥60 aligned with the repo floor)

---

[3.8.47] — TBD

Living section — bullets land here as PRs merge into release/v3.8.47 (parallel-cycle model; cycle opened at the v3.8.46 release freeze). Finalized at the v3.8.47 release.

📝 Maintenance

• chore(quality): validate-release-green --full-ci reproduces the full ci.yml static gate set locally — the pre-flight now reads ci.yml itself and runs every npm run check:* from the lint / quality-gate / quality-extended / docs-sync-strict / pr-test-policy jobs (-- ratchet flags preserved, test-masking against GITHUB_BASE_REF=main), skipping only the non-local pr-evidence/codeql-ratchet. Closes the gap where 11 static base-reds leaked to the v3.8.46 release PR in ~2h of layered CI. Also wired into nightly-release-green so a static base-red opens a tracking issue the night it lands. Regression guard: tests/unit/validate-release-green.test.ts (+5 extractCiGates cases).

---

[3.8.46] — 2026-07-07

✨ New Features

• feat(sse): hide paid-only models from auto/* routing when hidePaidModels is on (#6512) — follow-up to #6328/#6495. PR #6495 hid paid-only models from the GET /v1/models listing, but auto/* combos (auto/best-coding, auto/glm, …) could still pick a paid-only backend into their candidate pool → a 402/403 at request time. createVirtualAutoCombo now filters the candidate pool through the new pure open-sse/services/autoCombo/paidModelFilter.ts (filterPaidOnlyCandidates), applying the same free-model predicate #6495 uses in catalog.ts (providerHasFreeModels(provider) && isFreeModel(provider, {id})) whenever settings.hidePaidModels === true. Applied before the category/tier/family narrowing, so it covers every auto/* combo; an all-paid pool degrades to the existing graceful empty-pool path. Opt-in — default OFF leaves the pool unchanged (identity). Regression guard: tests/unit/autoCombo/paid-model-filter-6512.test.ts (4, incl. the default-off identity guard).
• feat(sse): provider-family auto combos — auto/glm, auto/minimax, auto/mimo, auto/zai, auto/gemma, auto/llama, auto/gemini (#6453) — new routable ids that materialize an on-demand virtual combo spanning whatever installed backends currently expose that model family, degrading gracefully as backends rotate. A new pure open-sse/services/autoCombo/modelFamily.ts (detectModelFamily) classifies by model-id prefix for six families; zai is instead resolved by provider id (z.ai's hosted API serves the same glm-* model ids as every other GLM backend, so auto/zai means "route to my z.ai backend specifically" vs auto/glm's "any connected GLM backend"). Reuses the existing createVirtualAutoCombo on-demand materialization path (no DB writes) and the /v1/models catalog advertising loop. Regression guard: tests/unit/autoCombo/provider-family-combos.test.ts (11).
• feat(proxy): native proxy-pool round-robin / egress IP rotation (#6365) — a scope (global / provider / account) can now hold multiple proxies as a pool with a rotation strategy, so outbound requests cycle their egress IP instead of pinning one proxy per scope. Migration 117_proxy_pool_rotation.sql lifts the UNIQUE(scope, scope_id) constraint (rebuild via the canonical rename/copy/drop; existing single assignments become 1-element pools) and adds a proxy_scope_rotation companion table holding the per-scope strategy + a persisted monotonic round-robin cursor. Strategies: round-robin (default, monotonic cursor — never Math.random), random, and sticky-per-N-min. Resolution (resolveProxyForScopeFromRegistry / resolveProxyForConnectionFromRegistry) now fetches the alive, position-ordered candidate set (unchanged PROXY_ALIVE_PREDICATE) and applies the strategy; an empty / all-dead pool still returns null — the #6246 fail-closed guard is untouched (never falls through to direct egress). Backend + DB only; dashboard pool-builder UI is a follow-up. Regression guard: tests/unit/proxy-pool-rotation-6365.test.ts (8, incl. fail-closed + backward-compat).
• feat(providers): end-to-end tool/function calling on the native Gemini /v1beta endpoint (#6222) — both directions of the Gemini↔OpenAI conversion now preserve tool data (previously silently dropped). Request side: convertGeminiToInternal (extracted to its own testable module) maps tools[].functionDeclarations → OpenAI tools, prior functionCall parts → assistant tool_calls, and functionResponse parts → tool-role messages. Response side: convertOpenAIResponseToGemini emits parts[].functionCall {name,args} from message.tool_calls, and the streaming openAIChunkToGeminiChunk accumulates fragmented tool_calls deltas by index into complete functionCall parts. The non-Gemini client paths (Claude, OpenAI-Responses) already preserved tool calls — this closes the gap specific to the native Gemini surface. Regression guard: tests/unit/v1beta-gemini-tool-calling-6222.test.ts (6, incl. a streaming SSE round-trip).
• feat(providers): copilot-m365-web enterprise / work tier support (#6334) — mirrors the EDU-tier pattern (#6210): M365ConnectionParams gains an agent field, a new opt-in M365_ENTERPRISE_OVERRIDES preset (agent=work, scenario=officeweb, licenseType=Premium) applies via providerSpecificData.tier="enterprise" (alias "work"), and agent is also overridable directly via providerSpecificData.agent. buildWsUrl was hardcoding agent="web" (the one enterprise-distinguishing param with no override path), so a Premium work account handshook then returned an empty stream. The individual and EDU paths are untouched. Kilo's dup flag vs #6210 (EDU tier) was a false positive — different tier. Regression guard: tests/unit/copilot-m365-enterprise-6334.test.ts (7). End-to-end confirmation on a real Premium work account is a live-VPS validation follow-up (Hard Rule #18). (thanks @Forcerecon)
• feat(api): standardized, provider-agnostic effort + thinking request params (#6241) — a thin standardization layer over the existing mature per-provider reasoning plumbing (no provider mapper touched). providerChatCompletionSchema gains a canonical effort (reusing the shared none/low/medium/high/xhigh vocabulary — the UI tiers extra/max collapse onto xhigh) and a boolean thinking. A pure normalizeReasoningRequest (wired once in src/sse/handlers/chat.ts, before any reasoning field is read) folds them onto the fields the translators already consume (reasoning_effort / reasoning.effort / thinking), so they fan out to Anthropic / Gemini / xAI / Responses — an explicit client reasoning_effort / object-shaped thinking always wins (backward-compatible). /models additively exposes supportsThinking + effort_tiers so the frontend can render the toggles (UI component is a follow-up). Regression guard: tests/unit/effort-thinking-standardization-6241.test.ts (12). (thanks @Iammilansoni, @shabeer)
• feat(combo): new pipeline (sequential) combo strategy (#6297) — the 18th routing strategy runs targets in order, threading each step's output into the next step's input, with an optional per-step prompt (system instruction); only the final step's response is returned. Distinct from fusion (parallel fan-out + judge). Implemented as a self-contained open-sse/services/pipeline.ts (sibling to fusion.ts), dispatched from combo.ts; the step list reuses combo.models order and reads an optional prompt off each target (backward-compatible — ignored by every other strategy). Intermediate steps run non-streaming with tools stripped (complete prose to thread forward); the final step keeps the client's stream flag + tools. A failing/empty/unparseable intermediate step fails the whole pipeline explicitly via a sanitized error (never silently swallowed). Kilo's dup flag vs #563 was a false positive (that's model→chain selection; this is a sequential chain). Regression guard: tests/unit/combo-pipeline-strategy.test.ts (5). (thanks @ofekbetzalel)
• feat(ci): check:test-masking now flags inline-reimplemented prod conditions (#6348) — a new report-only subcheck (v2, 6A.10 family) catches the wrong-shape contract test: a test that recomputes the condition under test inline instead of importing/exercising the real function (the #6216 class, where === 500 → >= 500 stayed green because the test re-implemented the branch). For each added/modified test file it warns when the file textually duplicates a ≥3-token conditional from a production file touched in the same PR and does not import the symbol/module owning it, via a pure, fixture-tested findReimplementedConditions() with an allowlist mirroring assertReductionAllowlist. Report-only for now (does not fail the gate) — to be promoted to blocking after a triage cycle. Regression guard: tests/unit/check-test-masking.test.ts (45).
• feat(sse): per-connection routing override (native vs CLIProxyAPI) (#6339) — the previously-dead isCliproxyapiDeepModeEnabled helper is now wired into resolveExecutorWithProxy: a single connection can opt itself into the CLIProxyAPI passthrough executor via providerSpecificData.cliproxyapiMode="claude-native", with precedence connection override > provider upstream_proxy_config mode > default. resolveExecutorWithProxy now receives the resolved connection's providerSpecificData (threaded from chatCore.ts), so one connection can deep-route while the provider's other connections stay native — no DB schema change (the toggle rides in providerSpecificData). Also resolves the same-provider mixing ask in #6340. Regression guard: tests/unit/chatcore-executor-proxy.test.ts (9). (thanks @RaviTharuma)
• feat(dashboard): "Add session cookie" modal now shows a prominent "Open ‹host› →" link to the provider's own site (#6268) — every -web cookie-session provider (chatgpt-web, claude-web, gemini-web, kimi-web, lmarena, qwen-web, m365-copilot-web, …) renders a one-click external link (opening the provider's login/home page in a new tab) so operators no longer tab away to retype the URL mid-setup. The host resolves from a pure, unit-tested resolveWebProviderHost() (prefers WEB_COOKIE_PROVIDERS[id].website, falls back to the registry baseUrl origin); non-web providers render exactly as before. Kilo's dup flag vs #6265 (modal-too-small-on-1080p) was a false positive — distinct concern. Regression guard: tests/unit/resolve-web-provider-host.test.ts (5). (thanks @chirag127)
• feat(providers): add DigitalOcean AI (serverless inference) as an OpenAI-compatible API-key provider (#6373) — base https://inference.do-ai.run/v1, wired through the shared OpenAI-compatible registry with full model passthrough (open-sse/config/providers/registry/digitalocean/, src/shared/constants/providers/apikey/inference-hosts.ts). Regression guard: tests/unit/digitalocean-provider.test.ts. (thanks @newnol)
• feat(providers): add Huancheng Public API (hcnsec) as an OpenAI-compatible regional provider (#6410) — Xinjiang Huancheng Cybersecurity's public LLM platform (base https://api.hcnsec.cn/v1, free credits via daily check-ins), wired through the shared OpenAI-compatible registry with full model passthrough (open-sse/config/providers/registry/hcnsec/, src/shared/constants/providers/apikey/regional.ts). Regression guard: tests/unit/hcnsec-provider.test.ts. (thanks @UnrealAryan)
• feat(dashboard): the web-session credential guide now shows an "Open {host}" link (#6316) to the provider's sign-in site (derived from the provider website via getProviderWebsiteHost), so you can jump straight to the page where the cookie/session must be captured. Regression guard: tests/unit/web-session-provider-link-6316.test.ts. (thanks @jordansilly77-stack)
• feat(cerebras): add the Gemma 4 31B model (gemma-4-31b) to the Cerebras registry + pricing table (#6331). Regression guard extends tests/unit/t28-model-catalog-updates.test.ts. (thanks @backryun)
• feat(providers): add Yuanbao (web) as a cookie-session provider (#6196) — yuanbao-web (Tencent Yuanbao, yuanbao.tencent.com) with cookie-only auth (hy_user/hy_token + public agent id), SSE→OpenAI translation incl. reasoning_content, exposing DeepSeek V3/R1 + Hunyuan / Hunyuan-T1. Regression guard: tests/unit/providers-yuanbao-web.test.ts. together-web was deferred (no verifiable web-session endpoint — needs a captured request) and huggingchat-web dropped (the existing huggingchat already is a web-cookie provider). (thanks @chirag127)
• feat(providers): route the built-in agentrouter through the dynamic Claude-Code wire image (#6056) — a small static allow-set (CC_WIRE_IMAGE_BUILTINS in open-sse/services/ccWireImageBuiltins.ts), consulted by isClaudeCodeCompatible / isClaudeCodeCompatibleProvider / applyFingerprint, makes agentrouter adopt the CC wire-image headers + fingerprint while guarding the CC baseUrl/auth branches so it keeps its own registry baseUrl and x-api-key auth. Regression guard: tests/unit/agentrouter-cc-wire-image.test.ts (asserts the wire image is applied AND agentrouter's baseUrl/auth are preserved). Live WAF-acceptance against agentrouter.org is a VPS validation follow-up (Hard Rule #18).
• feat(providers): bulk-add API keys for Cloudflare Workers AI (#6174) — cloudflare-ai is removed from the bulk-add exclusion list and the bulk parser gains a 3-field name|accountId|apiKey mode; the bulk route now builds a per-entry providerSpecificData so each key carries its own accountId (fixing the previous shared-object reuse), and both the create + key-validation paths receive it. Regression guard: tests/unit/bulk-api-key-parser-cloudflare.test.ts. (thanks @muflifadla38)
• feat(dashboard): routing/settings UX clarity (#6147) — (1) weighted combos show the effective routing share % next to each weight when weights don't sum to 100 (WeightTotalBar.tsx); (2) the status widget's user-facing "Cloud Sync" label is renamed to "Remote Settings Sync" (CloudSyncStatus.tsx; internal ids/state untouched); (3) built-in providers gain an opt-in advanced base-URL override (isBaseUrlOverrideEligibleProvider, hidden behind an "Advanced" toggle, reusing the existing providerSpecificData.baseUrl persistence — not globally widened). Regression guard: tests/unit/routing-settings-ux-6147.test.ts.
• feat(combo): add an option to disable session stickiness, per-combo or globally — round-robin / random combos can rotate to a different connection on every request instead of pinning a whole conversation to one connection by its first-message hash. Resolution precedence per-combo config.disableSessionStickiness → global settings.disableSessionStickiness → default false (preserves the #3825 prompt-cache/504 fix); gates both stickiness call sites in open-sse/services/combo.ts. Exposed as a global toggle (Combo Defaults) and a per-combo Inherit/on/off control. (#6168) Regression guard: tests/unit/combo-disable-session-stickiness.test.ts. (thanks @RCrushMe)
• feat(docker): add the OMNIROUTE_NO_SUDO env flag for root-less / user-namespaced deployments — the MITM cert-trust command path (resolveSudoSpawn in src/mitm/systemCommands.ts) now strips the leading sudo when the flag is truthy, in addition to the existing root / sudo-missing cases, so the Proxy Agent runs without sudo (the operator trusts the CA manually, e.g. via NODE_EXTRA_CA_CERTS). Argv-array spawn preserved — no shell interpolation (Hard Rule #13). (#6122) Regression guard: tests/unit/mitm-systemCommands-no-sudo.test.ts. (thanks @powellnorma)
• feat(providers): add Requesty as an OpenAI-compatible gateway provider (BYOK, base https://router.requesty.ai/v1, ~200 free requests/day) — wired through the shared OpenAI-compatible registry with full model passthrough (open-sse/config/providers/registry/requesty/, src/shared/constants/providers/apikey/gateways.ts). (#6120) Regression guard: tests/unit/requesty-provider.test.ts. (thanks @chirag127)
• feat(dashboard): add configured-only / available-only filters to the Free Provider Rankings page (#6150) — hide providers you haven't configured, or whose connections are all rate-limited / out of quota, via server-side query params (?configuredOnly / ?availableOnly on GET /api/free-provider-rankings) backed by a testable lib helper reusing the in-process connection state (no Redis). Both filters default off, so the default view is unchanged; this supersedes the earlier client-side "Configured Only" toggle (#6245) with an available-only dimension and unit-tested logic. Regression guard: tests/unit/freeProviderRankings-filters.test.ts.

🔧 Bug Fixes

• fix(dashboard): adding a second API key connection for the same provider no longer silently overwrites the first — the Add-API-key modal now derives a unique default connection name (main, then main-2, main-3, …) so the backend name-based upsert can't collide (#6499 — thanks @dilneiss).

• fix(compression): the session-dedup engine now also deduplicates a large multi-line block repeated within a single message (intra-message dedup), not just across turns; the compression-preview API surfaces a fallbackReason, and the fusion panel reports how many models were rate-limited vs failed on total-panel failure (#6501 — thanks @chirag127).

• fix(compression): a stacked-pipeline step naming an unregistered engine now surfaces a validationErrors entry instead of silently no-op'ing, so misconfigured pipelines are visible in the preview API (#6506 — thanks @chirag127).

• feat(usage): add a Codex reset-credit redemption flow to the Provider Limits UI (#6361) — a useCodexResetCreditRedemption hook + /api/usage/codex-reset-credit route + codexResetCredits lib let you redeem banked Codex reset credits from the quota card. Regression guard: tests/unit/codex-reset-credits.test.ts. (thanks @JxnLexn)

• feat(glm): add team-plan quota settings for glm-cn connections (#6351) — a dedicated GlmTeamQuotaFields form section (team quota id / limits) threaded through the Add/Edit connection modals, persisted via providerSpecificData, with the GLM usage service reading the team quota. Regression guards: tests/unit/glm-team-quota.test.ts, provider-specific-data-schema.test.ts. (thanks @hao3039032)

• feat(providers): add TinyFish web-fetch/search support (#6349) — a tinyfish-fetch executor + /v1/web/fetch route + MCP web-fetch tool, registered as a specialty-media provider with request-validation and a search-provider catalog entry. Regression guards: tests/unit/executor-tinyfish-fetch.test.ts, web-fetch-handler.test.ts, mcp-web-fetch-tool.test.ts, provider-validation-tinyfish.test.ts. (thanks @dtybnrj)

• fix(cli): omniroute launch-codex now spawns codex.cmd through a shell on Windows (the npm .cmd shim is unresolvable by bare spawn → ENOENT), mirroring the qodercli Windows fix (#6263) (#6312). Regression guard: tests/unit/launch-codex-windows-spawn-6312.test.ts. (thanks @swingtempo)

• fix(codex): isolate the Spark quota from the shared Codex quota and stabilize the quota UI ordering / hydration so per-scope limits render consistently (#6336). Regression guards: tests/unit/codex-quota-selection-hydration.test.ts, provider-limits-ui.test.ts + 3 more. (thanks @xz-dev)

• feat(api): add a hidePaidModels setting that filters paid-only models out of the /v1/models catalog. Regression guard: tests/unit/models-catalog-hide-paid.test.ts. (thanks @chirag127)

• fix(api-manager): the fallback model picker now preserves combos instead of dropping them when a primary model is unavailable (#6443). Regression guard: tests/unit/api-manager-page-static.test.ts. (thanks @jmengit)

• fix(providers): recoverable Antigravity / Cloud-Code (Gemini Code Assist) 403 responses (#6452) are now classified as a retryable project-config error instead of a terminal account ban, so a fixable project/API-disabled 403 no longer forces a ~1-year cooldown / full OAuth reconnect. Regression guard: tests/unit/errorclassifier-antigravity-403.test.ts. (thanks @developerjillur)

• fix(mitm): sanitizeHeaders now redacts Set-Cookie response headers so upstream session cookies never leak into logs / diagnostics (#6451). Regression guard: tests/unit/mitm-sanitize-headers.test.ts. (thanks @developerjillur)

• fix(api): /api/compression/preview now accepts mode: "caveman" and correctly handles stacked / zero-compression previews (#6425). Regression guard: tests/unit/api/compression-preview-caveman-and-stacked-6425.test.ts. (thanks @chirag127)

• feat(providers): add Zed hosted LLM aggregator as a native-app provider (#6118) — OAuth sign-in via the Zed hosted flow, registered through the shared provider registry + executor. Regression guards: tests/unit/zed-oauth-provider.test.ts, zed-import-utils.test.ts, zed-docker-detect.test.ts, mitm-handler-zed.test.ts. VPS-validated via live operator login (Hard Rule #18).

• fix(oauth): the Kiro SSO-cache auto-import now preserves the IDC region — cross-region Amazon Q / Kiro profiles imported from the SSO cache are no longer collapsed to the default region (#6113). Regression guard: tests/unit/kiro-auto-import-idc-2059.test.ts. VPS-validated via live operator login (Hard Rule #18).

• fix(dashboard): passthrough model aliases no longer collide when two namespaced model ids share a last segment (port from 9router#1850, #6431). enx/gpt-5.5 and enx/codebuddy/gpt-5.5 both auto-generated the alias gpt-5.5, so the second model could never be added (the UI just alerted "alias already exists"). Aliases are now disambiguated deterministically — bare last segment when free, then parent-qualified (codebuddy-gpt-5.5), then a numeric suffix — while re-adding the exact same model id is still blocked. Regression guard: tests/unit/passthrough-alias-1850.test.ts. (thanks @arpicato)

• fix(translator): preserve a Gemini functionResponse co-located with other parts (another functionCall, or trailing text) in the same content when translating Gemini → OpenAI (#6376). convertGeminiContent() early-returned the tool message on the first functionResponse part, dropping any co-located parts; such contents are now pre-split (one tool message per functionResponse, emitted first, plus one message for the remaining parts). Regression guard: tests/unit/gemini-to-openai-function-response.test.ts. (thanks @warelik)

• fix(headroom): detect a python interpreter managed by mise / pyenv / asdf / conda (port from 9router#2353, #6382). Headroom's python probe (src/lib/headroom/detect.ts) searched a hardcoded PATH, but version managers expose their interpreters via shim dirs that only join PATH through interactive-shell activation — which the non-interactive server never runs, so a managed python (≥3.10) was invisible and Headroom reported it missing. The search path now prepends the well-known shim/bin dirs (~/.local/share/mise/shims, ~/.pyenv/shims, ~/.asdf/shims, $CONDA_PREFIX/bin, ~/.local/bin, respecting MISE_DATA_DIR/PYENV_ROOT/ASDF_DATA_DIR when set), and a new HEADROOM_PYTHON env override lets operators point straight at their interpreter (mirroring HEADROOM_URL). Still shell-free (execFileSync). Regression guard: tests/unit/headroom-detect.test.ts (5). (thanks @loopyd)

• fix(executors): strip the OpenAI-Codex/Claude-CLI client_metadata passthrough field for NVIDIA requests (port from 9router#1887, #6411). NVIDIA's OpenAI-compatible wrapper rejects it with 400 Unsupported parameter, the same class already handled for cerebras/mistral; nvidia (executor default) was missing from the strip allowlist so Codex/Claude-Code passthrough requests 400'd. Regression guard: tests/unit/executor-default-strip-client-metadata.test.ts (+nvidia case). (thanks @phidinhmanh)

• fix(translator): strip the Claude-style thinking field for NVIDIA z-ai/glm-5.2 (port from 9router#2023, #6413). NVIDIA's OpenAI-compatible wrapper 400s on thinking (a Claude-format client routed here leaves a thinking:{type:"adaptive"}); the existing strip rule only dropped reasoning. Same class already handled for minimax-m2.7. Regression guard: tests/unit/nvidia-minimax-thinking-strip.test.ts (+glm-5.2 case). (thanks @phidinhmanh)

• fix(translator): suppress the streamed </think> close marker for the Antigravity IDE client (port from 9router#1061, #6415). On thinking-only turns Antigravity rendered a bare </think> as the sole visible content, tripping its loop-detection and wasting requests. Antigravity's UA (vscode/<v> (Antigravity/<v>)) is added to the marker-suppress allowlist (alongside OpenCode); Claude Code / Cursor still get the marker, and x-omniroute-thinking-marker: on force-restores it. Regression guard: tests/unit/think-close-marker-suppress-5245.test.ts. (thanks @abdofallah)

• fix(executors): strip nested reasoning_content from messages for Mistral (port from 9router#1649, #6417). Mistral's API returns 422 extra_forbidden when an assistant message carries reasoning_content (replayed thinking from a prior turn, e.g. via the Codex /responses path); the generic top-level 400 field-downgrade retry never covered the nested per-message field. DefaultExecutor now strips it for provider mistral only, so DeepSeek (which requires replayed reasoning_content) is unaffected. Regression guard: tests/unit/mistral-strip-reasoning-content-1649.test.ts. (thanks @xxy9468615)

• fix(executors): strip the client_metadata passthrough field on the OpenCode path (port from 9router#1442, #6418). OpenCode upstreams (e.g. kimi-k2.6 via opencode-go) reject it with 400 "Extra inputs are not permitted, field: 'client_metadata'"; the DefaultExecutor strip only covered cerebras/mistral and OpencodeExecutor extends BaseExecutor directly, so nothing removed it there. Regression guard: tests/unit/opencode-strip-client-metadata-1442.test.ts. (thanks @yanpaing007)

• fix(executors): inject the reasoning_content echo for the native Moonshot Kimi provider (port from 9router#1480, #6419). Kimi (executor default) is a thinking-mode upstream that 400s with "reasoning_content must be passed back" when a prior assistant turn lacks it; the placeholder injection was only wired into the OpenCode meta-provider, so direct multi-turn Kimi conversations failed. Scoped to kimi (gateway-served models matching the thinking-model name pattern are unaffected). Regression guard: tests/unit/kimi-native-reasoning-injected-1480.test.ts. (thanks @2220258345)

• fix(executors): recover from a strict gateway's context_management: Extra inputs are not permitted 400 (port from 9router#1468, #6420). Claude Code always sends a top-level context_management field; strict anthropic-compatible gateways reject it. The dedicated context-editing 400-fallback only fired when OmniRoute's own contextEditing feature was enabled (default off), so a client-sent field passed through untouched and 400'd. context_management is now in the generic reactive field-strip list, so it's stripped-and-retried once regardless of the feature flag (with correct request re-signing for claude-compatible relays). Regression guard: tests/unit/provider-field-strips.test.ts. (thanks @ohahe52-dot)

• fix(network): enable RFC 8305 Happy Eyeballs (autoSelectFamily) on the direct-egress undici dispatcher (port from 9router#1237, #6423). When DNS returns both IPv6 (AAAA) and IPv4 (A) and the IPv6 route is broken (e.g. a NAT64 64:ff9b:: prefix without routing), undici tried IPv6 first and hung until ETIMEDOUT (then a 502 + account lockout), even though curl reached the same host. The direct dispatcher now races both families and uses whichever connects first. Proxy paths pin family via proxyTls and are unaffected. Regression guard: tests/unit/direct-dispatcher-pipelining-4580.test.ts. (thanks @adentdk)

• fix(combo): round-robin now advances the rotation pointer past the model that actually served, not the eagerly-scheduled one (port from 9router#948, #6428). With stickyLimit: 1 (true round-robin), when the scheduled model failed and a different model served via fallback, the counter had already advanced +1 from the scheduled index — so the next request reused the fallback-served model, degrading round-robin into hot-spotting on whichever model was healthy. The pointer now advances to the served index + 1 (mirroring the sticky-limit>1 path). Session-stickiness (#3825) and distribution are preserved. Regression guard: tests/unit/combo-rr-fallback-advance-948.test.ts. (thanks @binsarjr)

• fix(sse): a non-string model field is now rejected with a 400 before the resolver, instead of crashing downstream .toLowerCase()/.split() calls into an empty-body 500 that escapes the error sanitizer (#6407). Regression guard: tests/unit/chat-non-string-model-6407.test.ts. (thanks @chirag127)

• fix(api): unknown /api/* routes now return a JSON 404 (instead of the dashboard HTML shell) and scalar chat params (model/temperature/etc.) are validated before the provider lookup so malformed requests fail fast with a clear 400 (#6424, #6412). Regression guards: tests/unit/api/api-catchall-json-404.test.ts, tests/unit/chat-early-schema-validation-6412.test.ts. (thanks @chirag127)

• fix(api): /v1/chat/completions now rejects a non-JSON Content-Type with a 400 before parsing the body (#6414). Regression guard: tests/unit/v1-chat-completions-content-type-6414.test.ts. (thanks @chirag127)

• fix(api): the X-OmniRoute-Compression response header is now echoed on /v1/chat/completions and /v1/completions (#6422). Regression guard: tests/unit/compression-header-echo-6422.test.ts. (thanks @chirag127)

• fix(api): concurrent GET /v1/models requests are coalesced into a single catalog build (#6408). Regression guard: tests/unit/v1-models-concurrent-6408.test.ts. (thanks @chirag127)

• fix(api): /v1/completions now echoes the requested body.model in its JSON + streamed responses (#6429). Regression guard: tests/unit/completions-body-model-echo.test.ts. (thanks @chirag127)

• fix(api): env-var master keys now see the full /v1/models catalog (#6406). Regression guard: tests/unit/models-catalog-envkey-6406.test.ts. (thanks @chirag127)

• fix(api): non-streaming /v1/completions responses now echo body.model aligned with the X-OmniRoute-Model header (#6426). Regression guard: tests/unit/v1-completions-model-header-match-6426.test.ts. (thanks @chirag127)

• fix(api): unknown /v1/* routes now return a JSON 404 not_found instead of the Next.js dashboard HTML shell (#6405). Regression guard: tests/unit/api/v1-catchall-json-404.test.ts. (thanks @chirag127)

• fix(api): the per-connection provider models route now degrades to the shipped catalog when a provider's /models endpoint answers with a redirect (#6267) — a qwen-web import failed with a raw Redirect blocked … (307) 503. safeOutboundFetch throws REDIRECT_BLOCKED on the 307, getSafeOutboundFetchErrorStatus maps it to 503, and buildDiscoveryErrorFallbackResponse treated every 503 as a hard error — so the non-empty getModelsByProviderId("qwen-web") catalog was never surfaced. A models-endpoint redirect is not a fixable-config error (unlike URL_GUARD_BLOCKED/INVALID_URL, which stay hard errors), so it now falls back to the local/cached catalog before the 503 short-circuit. General fix — covers any config-driven provider that 307s. Regression guard: tests/unit/provider-models-qwen-web-redirect-6267.test.ts. (thanks @chirag127)

• fix(api): the per-connection provider models route (MCP list_models_catalog + the dashboard import view) now merges USER-ADDED custom models into its response (#6247) — custom models live in the key_value namespace customModels, which the live REST /api/v1/models already merges, but src/app/api/providers/[id]/models/route.ts never read getCustomModels, so custom models were dropped on both the discovery-success and local_catalog paths. They are now folded into the returned model list (deduped by id, stamped owned_by: provider), fixing MCP + the dashboard import view in one place. Regression guard: tests/unit/provider-models-custom-merge-6247.test.ts. (thanks @RCrushMe)

• fix(providers): GitLab Duo tool-calling follow-up turns no longer fail upstream with 422 {"detail":"Validation error"} (tokens 0/0, rejected pre-inference). The #6234 tool-result-feedback fix serialized the entire multi-turn conversation into GitLab's single-file code_suggestions (small_file) generation endpoint — folded history that turn-N sent as an oversized current_file.content_above_cursor and duplicated verbatim into user_instruction, tripping the AI-Gateway's small_file validation guard. The executor now bounds that prompt: it keeps system + latest user message + the most-recent tool round (dropping older turns), caps oversized tool results, and stops duplicating the full prompt into user_instruction (which now carries only the short latest user message) — while still feeding the most-recent tool result back so the agent continues (open-sse/executors/gitlab.ts, #6220). The unit test covers the bounding logic; the upstream 422→200 clearing is VPS-only (Hard Rule #18). Regression guard: tests/unit/gitlab-tool-exchange-bounded-6220.test.ts.

• fix(i18n): the provider-detail (/dashboard/providers/[id]) connection-status filter labels no longer render as __MISSING__:All / __MISSING__:Active / __MISSING__:Error / __MISSING__:Banned / __MISSING__:CreditsExhausted in non-English locales (notably pt-BR) (#6290). Root cause was not the namespace mismatch the issue guessed — the providers.filter* keys resolve correctly in en.json; the debt lived in the locale mirrors (src/i18n/messages/*.json), where these five keys carried the __MISSING__: sync sentinel in ~15 locales and were absent entirely in ~26 others, so next-intl found the key and echoed the sentinel verbatim. All 40 non-English/-Chinese mirrors now ship real translations for the five providers.filter* labels. Regression guard: tests/unit/i18n-provider-filter-keys-6290.test.ts. (thanks @diegosouzapw)

• fix(providers): the copilot-m365-web streaming executor now emits debug-level WebSocket diagnostics (#6210) — the outbound WS URL (with the access_token redacted via redactWsUrl()), handshake success/failure, and each received SignalR frame's type/target. Previously the streaming path logged nothing, so an empty content:null response (the M365 Education / Starter tier symptom fixed in #6234) was undiagnosable even at APP_LOG_LEVEL=debug. The change is debug-level and side-effect-free — it does not alter streaming behavior or the frame parser, and the token never reaches the logs. Regression guard: tests/unit/copilot-m365-web-logging-6210.test.ts (thanks @qpeyba)

• fix(resilience): a round-robin combo no longer returns 503 all upstream accounts are unavailable when a compatibility-rejected target is actually healthy (#6238). filterTargetsByRequestCompatibility drops request-incompatible targets (tool/vision/structured-output unsupported, or below the required context window) before any availability check runs, and its compatible.length === 0 safety net only fired when all targets were filtered — not when the kept targets later all turned out runtime-unavailable (circuit-open / cooldown / no credentials). So a combo could 503 while a compat-rejected-but-healthy provider sat unused. handleRoundRobinCombo now keeps the compat-rejected set and, when every compat-kept target was skipped without a single real attempt, probes those rejected targets as a last-resort fallback tier (via the new pure open-sse/services/combo/comboCompatFallback.ts) before crystallizing the 503. Regression guard: tests/unit/combo-roundrobin-compat-fallback-6238.test.ts. (thanks @ThongAccount)

• fix(startup): best-effort self-heal for a corrupted Turbopack dev cache on Windows (#6289). On Windows, pnpm dev can fail at startup when Turbopack mmaps a persistent-cache SST file and the OS refuses the mapping (os error 1455 — "paging file too small"), which Turbopack surfaces as a misleading Module not found: Can't resolve '@/shared/utils/machine'. This is a known upstream Turbopack cache-corruption bug — not our code. The dev launcher (scripts/dev/run-next.mjs) now wraps nextApp.prepare() and, when it rejects with that signature (isTurbopackCacheCorruption in the new scripts/dev/turbopackCacheHeal.mjs), purges .build/next/**/cache/turbopack and retries once with a clear log. Caveat — best-effort only: the corruption often surfaces as a runtime overlay rather than a prepare() rejection, so this cannot always intercept it; the reliable remedy remains manually deleting the Turbopack cache dir. Regression guard: tests/unit/turbopack-cache-heal-6289.test.ts. (thanks @chirag127)

• fix(providers): qodercli PAT auth no longer fails with spawn qodercli ENOENT on Windows (#6263) — spawnQoderCli spawned the bare qodercli name with shell:false and an unenriched env, so the npm .cmd wrapper under %APPDATA%\npm (a user-PATH directory) was never resolved. It now resolves the absolute .cmd/.exe path through the existing getCliRuntimeStatus("qoder") resolver in src/shared/services/cliRuntime.ts (memoized), spawns with shell when the target is a .cmd/.bat, and uses the cliRuntime-enriched env (PATH + PATHEXT + APPDATA); the ENOENT error now lists the searched paths plus the CLI_QODER_BIN override. End-to-end spawn on a real Windows host is host-only (Hard Rule #18); the path-resolution logic is unit-tested. Regression guard: tests/unit/qodercli-windows-resolve-6263.test.ts. (thanks @chirag127)

• fix(sse): the reasoning-token buffer no longer inflates probe-sized max_tokens (#6274) — Claude Code's /model capability check sends max_tokens: 1, but for a thinking-capable model with a large output cap (e.g. glm-5.2) the #3587 headroom heuristic (max(current + 1000, ceil(current * 1.5))) rewrote it to 1001 and forwarded that upstream, wasting tokens on a request that was never a genuine reasoning budget. resolveReasoningBufferedMaxTokens() (open-sse/services/reasoningTokenBuffer.ts) now short-circuits and returns the caller's value verbatim when it is below the new REASONING_BUFFER_MIN_TRIGGER (256) threshold — a tiny explicit limit is a probe, not a reasoning request. Real budgets still receive the #3587 headroom unchanged, and the guard runs after the existing capability checks so unknown / non-reasoning models keep returning null. Regression guard: tests/unit/reasoning-token-buffer-6274.test.ts. (thanks @brightfiscalband)

• fix(cli): omniroute reset-password now works as a real subcommand, and password resets over piped (non-TTY) stdin actually apply (#6261, #6258). Two coupled defects: (1) #6261 — bin/omniroute.mjs routed everything through Commander with only two pre-Commander bypasses (--mcp, reset-encrypted-columns), so omniroute reset-password was rejected as an unknown command; only the separate omniroute-reset-password bin worked, while the docs falsely advertised the subcommand (incl. a bogus "legacy alias still works"). A pre-Commander bypass mirroring reset-encrypted-columns now dynamically imports bin/reset-password.mjs (which self-executes) before Commander parses; the three doc lines were corrected. (2) #6258 — bin/reset-password.mjs issued two sequential rl.question prompts; under piped stdin the second read never settled at EOF, so main() never reached resetManagementPassword and the reset was a silent no-op (both prompts printed, no success, password unchanged). The CLI now detects non-TTY stdin and reads it once (first line = password, second line = confirm if present, else reused), adds a --password-stdin flag (entire stdin is the password, no confirmation), and exits 0 explicitly so the success line always flushes; interactive TTY behavior is unchanged. Regression guard: tests/unit/reset-password-cli-6261-6258.test.ts (3). (thanks @chirag127)

• fix(db): the mass-migration safety abort now tells the operator how to bypass it and stops flooding the log (#6260) — after restoring a backup that wiped the migration tracking table, runMigrations() threw the abort on every downstream ensureDbInitialized(), re-logging the full banner 11+ times, and the message never mentioned the existing OMNIROUTE_MAX_PENDING_MIGRATIONS escape hatch. The abort text now appends a bypass hint (set OMNIROUTE_MAX_PENDING_MIGRATIONS=0 in server.env / DATA_DIR/.env), and a new MigrationSafetyAbortError is memoized so repeated calls in the same process throw the same instance and emit a single concise line instead of the full cascade. Regression guard: tests/unit/migration-safety-abort-6260.test.ts. (thanks @chirag127)

• fix(auth): importing a distinct Codex/ChatGPT OAuth auth.json is no longer falsely rejected as "already exists" when it belongs to a different user in the same workspace (#6301). findExistingCodexConnection (in src/lib/oauth/utils/codexAuthImport.ts) deduped only on providerSpecificData.workspaceId === accountId, where accountId is the shared chatgpt_account_id/tokens.account_id — so two members of the same ChatGPT Team collapsed onto a single connection (409 duplicate_account). The id_token's https://api.openai.com/auth claim carries a per-user chatgpt_user_id alongside the workspace id (the device-flow path already persisted it as chatgptUserId, but the import path did not). Now parseAndValidateCodexAuth extracts userId (chatgpt_user_id → user_id → JWT sub) into ParsedCodexAuth, the create/update paths persist chatgptUserId in providerSpecificData (mirroring codex.ts), and dedup keys on workspaceId AND chatgptUserId — with a backward-compat fallback to legacy accountId-only matching when no stored connection for that workspace records a chatgptUserId, so genuinely-same accounts still dedup. Regression guard: tests/unit/codex-auth-import-userid-dedup-6301.test.ts (4). (thanks @anungma)

• fix(providers): importing models for the venice-web provider no longer fails with a red "Provider venice-web does not support models listing" (#6269). venice-web is a web-cookie provider with an executor but no upstream /v1/models endpoint and no registry models, so the models route fell through to the tail 400. Mirroring the jules/linkup-search/ollama-search fix (#5569), it now ships a static local catalog entry in src/lib/providers/staticModels.ts — seeding the current Venice lineup (venice-uncensored, llama-3.3-70b, qwen3-235b, qwen3-4b, deepseek-r1-671b; Venice rotates its catalog, see docs.venice.ai/models/overview) — so the route returns 200 with source:"local_catalog", intentional:true. Regression guard: tests/unit/static-models-venice-web-6269.test.ts. (thanks @chirag127)

• fix(api): the specialty model catalogs (/v1/embeddings, /v1/images, /v1/music, /v1/videos model lists) are now derived from the unified catalog filtered by a predicate (getSpecialtyModelsResponse) instead of ad-hoc per-route logic, so they consistently respect active-credential visibility and stay in sync with the main catalog (#6303). Regression guard: tests/unit/specialty-model-catalog-routes.test.ts. (thanks @makcimbx)

• fix(api): the agent-bridge server route now resolves the MITM manager via a dynamic import("@/mitm/manager.runtime") so Turbopack does not statically pull the stub (or over-bundle the manager), and the agent-skills generator anchors its output base path with path.join(process.cwd(), …) so Turbopack's static analyzer stops tracing the whole project root (#6329, #6366). Regression guard: tests/unit/agent-bridge-server-route-dynamic-import.test.ts. (thanks @Iammilansoni)

• fix(api): internal probes (combo-test, cloud-sync verify) now pick a management-scoped / allow-all API key instead of naively grabbing getApiKeys()[0] — a restricted self:usage first row made the probe fail with "Model X is not allowed for this API key" even when the combo path was healthy (pickApiKeyForInternalUse in src/lib/db/apiKeys.ts). The API-manager model editor also falls back to /api/models?all=true when /v1/models is catalog-protected (#6372). Regression guard: tests/unit/pick-internal-api-key-6372.test.ts. (thanks @jmengit)

• fix(live-ws): the Live Dashboard WebSocket server now rejects on bind failure (e.g. EADDRINUSE when the API bridge already holds the port) instead of letting the error surface as an unhandled error event that crash-loops the process — the error listener is attached to wss (not server) and releases the EventBus subscription on a failed start (#6324). Regression guard: tests/unit/live-ws-eaddrinuse-6324.test.ts. (thanks @vinayakkulkarni)

• fix(dashboard): the Home provider-topology widget now trusts the live provider-metrics snapshot — it uses topology.errorProvider and live activeRequests directly instead of re-deriving state from a stale lastErrorAt or applying a frontend timeout filter, so the topology reflects real-time provider health (#6322). Regression guard: tests/unit/home-provider-topology-live-state.test.ts. (thanks @xz-dev)

• fix(sse): strip zero-width markers from streamed tool-call arguments — a follow-up to #5857. That PR removed injected zero-width joiners (U+200D) from streamed assistant text/reasoning but deliberately left tool-call argument JSON byte-exact. The request-side obfuscation (open-sse/services/claudeCodeObfuscation.ts) injects ZWJ into agent words — including the temp path inside the Bash tool description — and Claude models copy that verbatim into generated commands, which are delivered as tool-call arguments rather than assistant text. As a result the ZWJ survived and corrupted code blocks (e.g. a temp path rendered with an invisible joiner). Now open-sse/handlers/responseSanitizer.ts strips zero-width code points from tool-call argument strings at every emit site (OpenAI non-stream/stream chat tool_calls + legacy function_call, native Responses function_call items, the OpenAI→Responses conversion, and the native Responses streaming response.function_call_arguments.delta/.done events). Only zero-width code points are removed; JSON structure and all other bytes stay identical (no parse/restringify), so normal arguments remain byte-exact. Regression guard: 6 new cases in tests/unit/response-sanitizer.test.ts (suite 50/50).

• fix(nodejs): the default app log path now resolves under DATA_DIR (~/.omniroute/logs/application/app.log) instead of process.cwd() (#6197) — the globally-installed CLI runs from an arbitrary working directory, so anchoring the default to cwd made file logging silently write to (or no-op under) an unrelated directory, contradicting the documented .env.example default. getAppLogFilePath() now computes the default lazily via the pure resolveDataDir() resolver (honours a per-process DATA_DIR, no directory-creation side effect); an explicit APP_LOG_FILE_PATH still wins. Regression guard: tests/unit/logenv-datadir-path-6197.test.ts (3).

• fix(docker): AgentBridge/startMitm no longer aborts in containers/headless when the Antigravity-default DNS step can't write /etc/hosts (#6127), and the privileged command's stderr now reaches app.log instead of only a bare exit code hitting the toast (#6198). The default DNS step (addDNSEntry) was called unguarded while cert install and the two sibling DNS steps were each best-effort — in the runtime Docker image (USER node, no sudo, read-only /etc/hosts) it threw Command failed with code 1 out of startMitmInternal and killed the whole start, discarding the stderr. The three DNS steps are extracted into a best-effort provisionDnsEntries() where each failure is logged with the full err (stderr included, folded in by systemCommands.ts) and never aborts the start. Regression guard: tests/unit/mitm-dns-graceful-degrade-6127.test.ts (4).

• fix(providers): copilot-m365-web now supports the M365 Education "Starter / OfficeWebIncludedCopilot" tier and no longer returns an empty content:null stream (#6210). Two gaps: (1) buildWsUrl() hardcoded the individual-consumer scenario (OfficeWebPaidConsumerCopilot, isEdu=false) — the EDU tier is now opt-in via providerSpecificData.tier="edu", emitting scenario=OfficeWebIncludedCopilot/isEdu=true (the individual path is unchanged); (2) the EDU/GPT-5.5 path streams deltas via arguments[0].writeAtCursor (incremental) instead of only messages[].text (accumulated snapshots), which the parser dropped — a new accumulateBotContent() folds both formats, with type:2 item.result.message as a last-resort fallback. Regression guard: tests/unit/copilot-m365-edu-writeatcursor-6210.test.ts (10). (thanks @qpeyba)

• fix(providers): GitLab Duo executor now feeds tool results back into the prompt instead of looping (#6220) — buildPrompt() branched only on system/user and took userParts.at(-1), silently dropping the assistant{tool_calls} + tool{result} turns the client appended, so the reconstructed prompt was byte-identical to turn 1 and the model re-emitted the same <tool> call forever. When a tool exchange is present the full conversation is now serialized, folding each tool result back keyed by its tool_call_id; simple conversations keep the legacy shape. Complements the tool_call emission from #6051 (the kilo-duplicate label was a false positive — different, sequential defect). Regression guard: tests/unit/gitlab-tool-result-feedback-6220.test.ts (4).

• fix(providers): opencode-go/opencode-zen can now synthesize the OpenCode CLI identity headers Cloudflare requires on VPS egress (#5997) — on a datacenter VPS, opencode.ai/zen/go/v1/chat/completions 403s (HTML challenge) requests lacking CLI identity, while the reporter's control curl proved that User-Agent: opencode-cli/1.0.0 + x-opencode-client: cli + x-opencode-project: default + fresh request/session UUIDs succeed. Opt-in via OPENCODE_SYNTHESIZE_CLI_HEADERS=true (values overridable via OPENCODE_GO_USER_AGENT/OPENCODE_USER_AGENT/OPENCODE_CLIENT/OPENCODE_PROJECT); it fills only headers the client did not already send. Kept off by default — the forward-only path is deliberate (fabricating a wrong value risks upstream rejection; a prior dedup regressed with opencode/local), so this replaces the fragile local header-injection shim without changing default behavior. Regression guard: tests/unit/opencode-cli-headers-synthesis-5997.test.ts (6). (thanks @aleksesipenko)

• fix(resilience): sticky session affinity now evicts and fails over to another account when the pinned account is exhausted/unavailable (#6219)

• fix(sse): Responses API passthrough now drops internal commentary-phase output before forwarding to clients (gated by RESPONSES_PASSTHROUGH_DROP_COMMENTARY, default on) (#6199)

• fix(sse): tool-call function schemas with a root type: null are now coerced to type: "object" before dispatch (#6359) — clients like the Codex app emit parameters: { type: null, ... } for some tools, which OpenAI-compatible upstreams reject with 400 Invalid schema for function '...': schema must be a JSON Schema of 'type: "object"', got 'type: null', failing the whole request. toolSchemaSanitizer already stripped the null; it now re-adds the mandatory root "object" type (and empty properties/open additionalProperties when absent). Combinator roots (anyOf/oneOf/allOf) and explicit root types are left untouched. Regression guard: 5 new cases in tests/unit/tool-schema-sanitizer.test.mjs.

• fix(docker): AgentBridge no longer fails to start on npm/Electron/VPS installs with "MITM manager stub reached at runtime" (#6344) — v3.8.45 flipped the production bundler default to Turbopack, but next.config.mjs aliased @/mitm/manager to its Docker-only degraded stub unconditionally. That was harmless while Docker (which sets the alias intentionally for #3390 graceful degradation) was the sole Turbopack consumer, but once every artifact built with Turbopack the stub shipped to all non-Docker users and startMitm threw on the first Agent-Bridge start. The alias is now opt-in via OMNIROUTE_MITM_STUB=1 (set only by the Dockerfile) through the shared scripts/build/mitm-stub-flag.mjs helper; default builds bundle the real manager. Regression guard: tests/unit/mitm-stub-alias-6344.test.mjs (4).

• fix(proxy): stop the v3.8.44 proxy regression that leaked the real IP and disabled healthy proxies (#6246). Two coupled defects from the new health scheduler: (1) IP leak — when a proxy assigned to a connection was marked inactive, resolution fell through to a direct egress instead of blocking, exposing the operator's real IP; (2) over-deactivation — the sweep flipped a proxy to inactive on the first failed probe and counted our own 5s timeout / a probe-target 5xx as the proxy's fault, so healthy paid proxies vanished from egress selection ("my proxies are not being used anymore"). Fix: the sweep decision is extracted into a pure, network-free decideProxyHealthAction (src/lib/proxyHealth/decision.ts) — by default the health check now only counts/logs and never downgrades status (a proxy is downgraded/removed only with PROXY_AUTO_REMOVE=true, after PROXY_AUTO_REMOVE_AFTER consecutive conclusive failures); probes are classified tri-state so an inconclusive result (our timeout, or a 5xx from the probe target) never penalizes the proxy, and the probe timeout is raised 5s→15s. Separately, safeResolveProxy now fails closed via the existing policy: a connection whose assigned proxy is dead is blocked instead of leaking direct (hasBlockingProxyAssignment), honoring the explicit proxy off toggles and the PROXY_FAIL_OPEN=true opt-out. Existing proxies stuck inactive by the old behavior need a one-time manual re-activate (the operator owns proxy status). Regression guards: tests/unit/proxy-health-decide-action-6246.test.ts, tests/unit/proxy-assigned-unavailable-6246.test.ts.

• fix(proxy): make "Test All" read-only and add bulk enable/disable (#6246). Complements the core fail-closed / scheduler fix (#6296) with the two remaining reporter asks. (1) The "Test All" button (POST /api/settings/proxies/auto-test) used to flip a proxy to inactive on a failed reachability probe; since the egress selector excludes inactive proxies, a flaky probe (an unreachable httpbin.org, a proxy that blocks HEAD, or a slow paid proxy) silently disabled every proxy that failed — "Test All" is now read-only by default (only the operator sets a proxy active/inactive; opt back into the legacy test-and-set with PROXY_HEALTH_AUTO_DEACTIVATE=true). (2) Adds a bulk enable/disable proxies endpoint + toolbar action (POST /api/settings/proxies/batch-activate) so an operator can re-activate proxies in one click. Regression guard: tests/unit/proxy-health-6246.test.ts. (thanks @tenshiak)

• chatcore (tools): stop the default 128-tool cap from silently dropping opencode's task/MCP tools. opencode (used as an MCP/agent host) sends a large tool list; when it exceeds the speculative MAX_TOOLS_LIMIT (128) default, truncateToolList did a blind tools.slice(0, 128), dropping every tool past index 128 — including opencode's built-in task tool (subagent launch) and many MCP tools, so models routed through OmniRoute could no longer spawn subagents or reach part of their tools. The cap exists to avoid upstream 400s for providers with real hard limits (e.g. grok-cli 200), so it is kept for those: detection of the opencode client (isOpencodeClient — any x-opencode-* header, or opencode in the user-agent) now only bypasses the speculative 128 default, never a known provider ceiling. Precedence is explicit — a proactive/detected provider limit always truncates (even for opencode); otherwise opencode forwards its full tool list; otherwise the unchanged 128 default applies to every other client. Refactors getEffectiveToolLimit into getKnownToolLimit(provider) ?? DEFAULT_LIMIT (byte-identical for existing callers) and fixes a cosmetic debug-log that reported the truncated count instead of the original. Regression guard: tests/unit/tool-limit-detector.test.ts.

• fix(mitm): the macOS MITM-cert install check now matches the system keychain again. security find-certificate -a -Z prints the SHA-1 as a colon-less hex string, but the installed-check compared it against getCertFingerprint()'s colon-separated form, so the substring match never hit — the cert was reported as not-installed and re-prompted for the sudo install on every run. Fingerprints are now normalized (colons stripped, upper-cased) on both sides via the extracted macCertOutputHasFingerprint helper. Regression guard: tests/unit/mitm-cert-mac-fingerprint.test.ts. (#6204, closes #6134 — thanks @rianonehub)

• fix(api): /v1/messages/count_tokens now counts tool_use, tool_result and thinking content blocks (and array-form system prompts) in the local-estimation path, instead of only text. Real agentic conversations keep ~95% of their tokens inside tool results; the previous estimate returned near-zero for them, which silently broke Claude Code's auto-compaction (context grew past the window with no compaction until the upstream API rejected the request). The real provider-side count path is unchanged. Regression guard: tests/unit/messages-count-tokens-route.test.ts. (#6221 — thanks @luweiCN)

• fix(antigravity): strip a trailing assistant prefill turn for Vertex Claude models to avoid upstream 400s (#6114). Regression guard: tests/unit/antigravity-claude-prefill-strip.test.ts. (thanks @anki1kr)

• fix(security): the mutable cloud-agent routes (/api/cloud/credentials/update, /api/cloud/models/alias) now require management auth instead of being treated as public. They were classified as public API routes, so a request without management credentials could update stored cloud-agent credentials and model aliases. They are removed from the public-route set, classified as management routes in the authz pipeline, and gated by requireManagementAuth; cloud read/auth routes stay public. Regression guards: tests/unit/cloud-write-auth.test.ts, tests/unit/authz/classify.test.ts, tests/unit/public-api-routes.test.ts. (#6233 — thanks @vittoroliveira-dev)

• refactor(dashboard): extract the onboarding-wizard "Open provider details" link target into a pure, unit-tested buildProviderDetailsHref(connection) helper. The wizard already routes by connection.id (the node UUID) rather than the provider category slug (#6144/#6145); this hardens that behavior behind a tested helper that guards a missing id/connection. Regression guard: tests/unit/provider-onboarding-href.test.ts. (#6166 — thanks @KooshaPari)

• fix(security): the doubao synthetic device-id generator now derives its digits via an unbiased crypto-random draw (rejection sampling over crypto.randomBytes()) instead of a % 10 reduction, closing a CodeQL js/biased-cryptographic-random finding.

• fix(agentSkills): the GitHub-skills generator now resolves outputDir to an absolute path before writing, fixing a regression introduced by #6366 (relative-to-cwd base path) that could write generated skill files to the wrong directory.

• fix(security): /api/keys/{id}/devices now checks the HTTP method before auth/validation, returning a 405 for non-GET/DELETE verbs instead of a misleading 401/500 (closes a dast-smoke QUERY-method finding).

• fix(quality): clear the last 2 heavy quality-gate reds on the release tip (cycle pre-flight).

• fix(mitm): the test suite and CI can never mutate the OS trust store — OMNIROUTE_SKIP_SYSTEM_TRUST=1 is set globally for tests/CI so installCert/uninstallCert/installTproxyCa skip the privileged OS dispatch (#6310; full detail is under the [3.8.45] section below — this branch received it via the parallel-cycle sync-back).

• fix(api): POST /api/github-skills now Zod-validates its request body; documented the new quality-gate env vars and pinned the merge-integrity GitHub Actions to a commit SHA.

• fix(skills): generate the missing omni-github-skills registry entry and align the agent-skills catalog-count tests (follow-up to #6186).

• fix(quality): clear the cycle's 11 net-new ESLint errors and make validate-release-green suppressions-aware.

• fix(security): proxy-pool random rotation now selects via crypto.randomInt instead of Math.random — silences the post-release CodeQL js/insecure-randomness alerts (#698/#699) that flagged Math.random flowing into the selected proxy's credentials. Load-balancing selection is not a secret, but the crypto source is unbiased and clears the alert at the origin (#6365 follow-up).

📝 Maintenance

• i18n(it): add 118 missing Italian (it) translations (net-additive — no existing keys dropped, valid JSON), improving Italian UI coverage. (#6212 — thanks @serverless83)
• chore(providers): remove deprecated MiMo V2 model entries from the catalogs (xiaomi-mimo, opencode-go, zenmux-free, audio TTS) — the upstream V2 line is superseded by MiMo V2.5; drops mimo-v2-tts, mimo-v2-pro, mimo-v2-omni, mimo-v2-flash, mimo-v2-flash-free and realigns the provider-catalog tests. (#6248 — thanks @backryun)
• chore(release): ~50 commits on this branch are v3.8.45 pre-flight/hardening fixes and CI-perf work that landed here via the parallel-cycle sync-back (sync-next-cycle.mjs, Hard Rule #21) after the v3.8.45 git tag was cut, and are already fully documented under the [3.8.45] section below — listed here only so the per-cycle commit-coverage check (npm run release:uncovered) doesn't flag them as gaps. Provider/catalog/UX/backend fixes: #6041, #6078, #6108, #6135, #6148, #6149, #6154, #6158, #6161, #6162, #6163, #6164, #6165, #6170, #6177, #6178, #6181, #6186, #6187, #6191, #6193, #6194, #6195, #6200, #6205, #6208, #6209, #6211, #6213, #6223, #6224, #6225, #6226, #6227, #6228, #6229, #6230, #6235, #6291, #6292. CI/release-pipeline work: #6167, #6203, #6214, #6215, #6218, #6273, #6275, #6283, #6284, #6285, #6300, #6305.
• chore(release): additional zero-ref release-cycle plumbing on this branch, kept out of release:uncovered on purpose (no #N in the commit subject to cite): opening the v3.8.46 cycle, opening/closing the v3.8.45 cycle, the finalized [3.8.45] CHANGELOG i18n sync-back to 42 mirrors, the v3.8.45 cognitive/cyclomatic and file-size drift rebaselines, ESLint stale-suppression pruning (4,273 → 4,233), and clearing test-masking/docs-all pre-flight reds for v3.8.45.

⚡ Performance & Infrastructure

• perf(release-green): the pre-flight validator (scripts/quality/validate-release-green.mjs) now runs its 4 slow suites (unit / vitest / integration / pack-artifact) concurrently via Promise.all — pre-flight wall time drops from ~the sum of the suites to ~the slowest one (~30min saved per round; Phase 0 was the nº1 bottleneck of the v3.8.45 release benchmark, 2h54 of 6h34 e2e). Guard: tests/unit/validate-release-green.test.ts ("runs the slow suites CONCURRENTLY"). (#6319)
• fix(ci): scripts/release/sync-next-cycle.mjs — two defects found live in its first production run (v3.8.45 Phase 5): (1) the git() helper's default 1 MiB maxBuffer crashed with ENOBUFS on git show origin/main:CHANGELOG.md (the CHANGELOG alone is >1 MiB) — widened to 64 MiB; (2) the i18n resync only propagated the [NEXT] (TBD) section, leaving the just-shipped finalized section as "— TBD" in all 42 mirrors — it now also syncs [prevVersion] bounded by the heading below it (new exported pure helper versionAfter). Guards: +5 tests in tests/unit/sync-next-cycle.test.ts (8/8). (#6327)
• test(ci): concurrency-sensitive flaky tests are quarantined into a serial pass (tests/unit/serial/, --test-concurrency=1, appended to every unit runner incl. sharded variants — the serial pass is sharded too so concurrent shard jobs never self-collide). Initial set: glm-coding-plan-monthly-3580, quota-division-blocks, provider-health-autopilot, combo-health-autopilot — the class behind the ~28min CI wedges/re-runs (two live 1h+ wedges cancelled during this PR's own validation). Discovery + TIA gates track the new glob; systemic root cause (async logger writing after teardown) tracked in #6360. Guard: tests/unit/test-serial-quarantine.test.ts (4). (#6347)

🙌 Contributors

Thanks to everyone whose work landed in v3.8.46:

Contributor	PRs / Issues
@2220258345	direct commit / report
@abdofallah	direct commit / report
@adentdk	direct commit / report
@aleksesipenko	direct commit / report
@anki1kr	direct commit / report
@anungma	direct commit / report
@arpicato	direct commit / report
@backryun	#6248
@binsarjr	direct commit / report
@brightfiscalband	direct commit / report
@chirag127	#6501, #6506
@developerjillur	direct commit / report
@dilneiss	#6499
@dtybnrj	direct commit / report
@Forcerecon	direct commit / report
@hao3039032	direct commit / report
@Iammilansoni	#6150, #6245
@jmengit	direct commit / report
@jordansilly77-stack	direct commit / report
@JxnLexn	direct commit / report
@KooshaPari	#6166
@loopyd	direct commit / report
@luweiCN	#6221
@makcimbx	direct commit / report
@muflifadla38	direct commit / report
@newnol	direct commit / report
@ofekbetzalel	direct commit / report
@ohahe52-dot	direct commit / report
@phidinhmanh	direct commit / report
@powellnorma	direct commit / report
@qpeyba	direct commit / report
@RaviTharuma	direct commit / report
@RCrushMe	direct commit / report
@rianonehub	#6134, #6204
@serverless83	#6212
@swingtempo	direct commit / report
@tenshiak	direct commit / report
@ThongAccount	direct commit / report
@UnrealAryan	direct commit / report
@vinayakkulkarni	direct commit / report
@vittoroliveira-dev	#6233
@warelik	direct commit / report
@xxy9468615	direct commit / report
@xz-dev	direct commit / report
@yanpaing007	direct commit / report
@diegosouzapw	maintainer

---

[3.8.45] — 2026-07-06

✨ New Features

• feat(providers): add Yuanbao (web) as a cookie-session provider (#6196) — yuanbao-web (Tencent Yuanbao, yuanbao.tencent.com) with cookie-only auth (hy_user/hy_token + public agent id), SSE→OpenAI translation incl. reasoning_content, exposing DeepSeek V3/R1 + Hunyuan / Hunyuan-T1. Regression guard: tests/unit/providers-yuanbao-web.test.ts. together-web was deferred (no verifiable web-session endpoint — needs a captured request) and huggingchat-web dropped (the existing huggingchat already is a web-cookie provider). (thanks @chirag127)
• feat(providers): route the built-in agentrouter through the dynamic Claude-Code wire image (#6056) — a small static allow-set (CC_WIRE_IMAGE_BUILTINS in open-sse/services/ccWireImageBuiltins.ts), consulted by isClaudeCodeCompatible / isClaudeCodeCompatibleProvider / applyFingerprint, makes agentrouter adopt the CC wire-image headers + fingerprint while guarding the CC baseUrl/auth branches so it keeps its own registry baseUrl and x-api-key auth. Regression guard: tests/unit/agentrouter-cc-wire-image.test.ts (asserts the wire image is applied AND agentrouter's baseUrl/auth are preserved). Live WAF-acceptance against agentrouter.org is a VPS validation follow-up (Hard Rule #18).
• feat(providers): bulk-add API keys for Cloudflare Workers AI (#6174) — cloudflare-ai is removed from the bulk-add exclusion list and the bulk parser gains a 3-field name|accountId|apiKey mode; the bulk route now builds a per-entry providerSpecificData so each key carries its own accountId (fixing the previous shared-object reuse), and both the create + key-validation paths receive it. Regression guard: tests/unit/bulk-api-key-parser-cloudflare.test.ts. (thanks @muflifadla38)
• feat(dashboard): routing/settings UX clarity (#6147) — (1) weighted combos show the effective routing share % next to each weight when weights don't sum to 100 (WeightTotalBar.tsx); (2) the status widget's user-facing "Cloud Sync" label is renamed to "Remote Settings Sync" (CloudSyncStatus.tsx; internal ids/state untouched); (3) built-in providers gain an opt-in advanced base-URL override (isBaseUrlOverrideEligibleProvider, hidden behind an "Advanced" toggle, reusing the existing providerSpecificData.baseUrl persistence — not globally widened). Regression guard: tests/unit/routing-settings-ux-6147.test.ts.
• feat(combo): add an option to disable session stickiness, per-combo or globally — round-robin / random combos can rotate to a different connection on every request instead of pinning a whole conversation to one connection by its first-message hash. Resolution precedence per-combo config.disableSessionStickiness → global settings.disableSessionStickiness → default false (preserves the #3825 prompt-cache/504 fix); gates both stickiness call sites in open-sse/services/combo.ts. Exposed as a global toggle (Combo Defaults) and a per-combo Inherit/on/off control. (#6168) Regression guard: tests/unit/combo-disable-session-stickiness.test.ts. (thanks @RCrushMe)
• feat(docker): add the OMNIROUTE_NO_SUDO env flag for root-less / user-namespaced deployments — the MITM cert-trust command path (resolveSudoSpawn in src/mitm/systemCommands.ts) now strips the leading sudo when the flag is truthy, in addition to the existing root / sudo-missing cases, so the Proxy Agent runs without sudo (the operator trusts the CA manually, e.g. via NODE_EXTRA_CA_CERTS). Argv-array spawn preserved — no shell interpolation (Hard Rule #13). (#6122) Regression guard: tests/unit/mitm-systemCommands-no-sudo.test.ts. (thanks @powellnorma)
• feat(providers): add Requesty as an OpenAI-compatible gateway provider (BYOK, base https://router.requesty.ai/v1, ~200 free requests/day) — wired through the shared OpenAI-compatible registry with full model passthrough (open-sse/config/providers/registry/requesty/, src/shared/constants/providers/apikey/gateways.ts). (#6120) Regression guard: tests/unit/requesty-provider.test.ts. (thanks @chirag127)
• feat(dashboard): add configured-only / available-only filters to the Free Provider Rankings page (#6150) — hide providers you haven't configured, or whose connections are all rate-limited / out of quota, via server-side query params (?configuredOnly / ?availableOnly on GET /api/free-provider-rankings) backed by a testable lib helper reusing the in-process connection state (no Redis). Both filters default off, so the default view is unchanged; this supersedes the earlier client-side "Configured Only" toggle (#6245) with an available-only dimension and unit-tested logic. Regression guard: tests/unit/freeProviderRankings-filters.test.ts.
• feat(rankings): add a 'Configured Only' filter to the Free Provider Rankings page, so the table can be narrowed to just the providers you have configured connections for (with an empty-state hint when none are configured). New en.json keys and a pure filter helper covered by tests/unit/free-provider-rankings-configured-filter.test.ts. (#6245, closes #6150 — thanks @Iammilansoni)

🔧 Bug Fixes

• fix(mitm): the test suite and CI can never mutate the OS trust store again — OMNIROUTE_SKIP_SYSTEM_TRUST=1 (set by the global test setup and all CI workflows) makes installCert/uninstallCert/installTproxyCa skip the privileged OS dispatch while preserving the #4546 environment-skip contract. Root cause of the self-hosted runner incident: a cert-flow integration test installed a 105-byte fake PEM into /usr/local/share/ca-certificates, breaking ALL system TLS on the VM. Regression guard: tests/unit/system-trust-test-guard.test.ts. (#6310)

• fix(security): /api/keys/{id}/devices answers a clean method-first 405 for undocumented HTTP methods (e.g. the new QUERY) via a dedicated http-method-guard rule — the auth layer was answering 401 first, failing schemathesis's unsupported-methods check. Same pattern as the v3.8.44 TRACE fix. Regression guard: tests/unit/dast-method-not-allowed.test.ts.

• fix(combo): the #6216 empty-stream failover is restricted to truly empty bodies (zero bytes — the Gemini HTTP-200-empty case), restoring the #3399/#3685 pass-through contracts for [DONE]-terminated empty streams and incomplete Claude lifecycles. New guard: #5976 truly EMPTY streaming body → invalid for combo failover (87/87 across both suites).

• fix(combo): 5 streaming-path fixes — locked-stream 500, error-frame-only-if-no-content, Gemini MALFORMED_RESPONSE→content_filter failover, correlationId substring search, per-model-500 lockout skip + request-logger UI detail. Maintainer follow-up: releaseQualityClone cancels the abandoned quality-check tee branch (per-request memory) + regression test. (#6216 — thanks @hartmark)

• fix(skills): generate the missing omni-github-skills registry entry (the #6186 catalog addition never ran the generator — 8 integration assertions split between old/new counts) and align the agent-skills catalog counts across integration + unit suites (43 = 23 API + 20 CLI; 44 with config).

• fix(a2a): finish the #6186 catalog-count update — listCapabilities metadata reported coverage.api.total: 22 (type literal + value) and SkillCoverageSchema pinned z.literal(22), so the schema would REJECT the correct runtime value with 23 API skills. All three aligned to 23.

• fix(github-skills): add a missing import, unit tests and a settings JSON-parse fix for the GitHub agent-skill discovery/import flow. (#6186 — thanks @Moseyuh333)

• fix(api): POST /api/github-skills validates its body with a Zod schema (validateBody) instead of blind request.json() destructuring — a non-array targets would crash .map. Regression guard: tests/unit/github-skills-route-validation.test.ts.

• fix(docker): add id= to the BuildKit cache mounts so strict builders (e.g. buildkitd with strict frontend parsing) accept the Dockerfile. (#6291 — thanks @karimalsalah)

• fix(oauth): register zed in the OAuth PROVIDERS map (fixes "Unknown provider" on the Zed sign-in flow) (#6078 — thanks @anki1kr), and align zed in OAUTH_PROVIDER_IDS + the config enum after the merge.

• fix(doubao-web): switch the Doubao web provider to the Dola global endpoint. (#6235 — thanks @backryun)

• fix(doctor): resolve two false-positive WARNs in the doctor diagnostics (#6163, closes #6162 — thanks @arssnndr)

• fix(providers): refresh the GitHub Copilot model catalog to the current upstream set. (#6154 — thanks @backryun)

• fix(providers): correct the Kiro model catalog to real upstream ids — fabricated claude-opus-4.7/claude-sonnet-4.6 entries removed, real claude-sonnet-5/claude-sonnet-4.5/claude-haiku-4.5 kept. (#6170)

• feat(sse): surface Kiro adaptive-thinking reasoning frames as reasoning_content in the OpenAI-shaped stream. (#6213 — thanks @VXNCXNX)

• fix(cli): use OMNIROUTE_SERVER_HOST instead of the POSIX auto-set HOSTNAME for the bind address (fixes wrong bind on POSIX shells that export HOSTNAME). (#6195, closes #6194 — thanks @Theadd)

• feat(provider): add Claude 5 Sonnet to the Claude Web provider catalog. (#6209, closes #6200 — thanks @Iammilansoni)

• fix(providers): add nvidia to PROVIDER_TOOL_LIMITS (1536) to prevent silent tool-list truncation. (#6177 — thanks @LuisAlejandroVega)

• fix(translator): strip the reasoning param for nvidia z-ai/glm-5.2 (upstream 400s on it). (#6181 — thanks @kanztu)

• fix(dashboard): providers page gains a data-timeout guard and the live-WS standalone wiring (no more indefinite spinner when the data fetch stalls). (#6211)

• fix(sse): surface the ChatGPT-web image silent-drop as an accurate error instead of an empty success. (#6208)

• fix(cline): force upstream streaming for Cline/ClinePass (streaming-only API) — non-stream client requests are served from the buffered SSE. (#6165)

• fix(dashboard): remove the always-on Auto-Routing (combo) banner from the home page — it did not reflect live routing state and reappeared on every fresh browser. Replacement guard: tests/unit/home-no-autorouting-banner.test.ts. (#6164)

• fix(dashboard): stop a model-test error from freezing the page (React #31 object-as-child toast) — errors go through extractApiErrorMessage. (#6161)

• fix(oauth): extract the keychain-import-only guard to its own module, restoring the oauth file-size freeze. (#6158)

• fix(sse): strip zero-width markers from streamed tool-call arguments — a follow-up to #5857. That PR removed injected zero-width joiners (U+200D) from streamed assistant text/reasoning but deliberately left tool-call argument JSON byte-exact. The request-side obfuscation (open-sse/services/claudeCodeObfuscation.ts) injects ZWJ into agent words — including the temp path inside the Bash tool description — and Claude models copy that verbatim into generated commands, which are delivered as tool-call arguments rather than assistant text. As a result the ZWJ survived and corrupted code blocks (e.g. a temp path rendered with an invisible joiner). Now open-sse/handlers/responseSanitizer.ts strips zero-width code points from tool-call argument strings at every emit site (OpenAI non-stream/stream chat tool_calls + legacy function_call, native Responses function_call items, the OpenAI→Responses conversion, and the native Responses streaming response.function_call_arguments.delta/.done events). Only zero-width code points are removed; JSON structure and all other bytes stay identical (no parse/restringify), so normal arguments remain byte-exact. Regression guard: 6 new cases in tests/unit/response-sanitizer.test.ts (suite 50/50).

• fix(nodejs): the default app log path now resolves under DATA_DIR (~/.omniroute/logs/application/app.log) instead of process.cwd() (#6197) — the globally-installed CLI runs from an arbitrary working directory, so anchoring the default to cwd made file logging silently write to (or no-op under) an unrelated directory, contradicting the documented .env.example default. getAppLogFilePath() now computes the default lazily via the pure resolveDataDir() resolver (honours a per-process DATA_DIR, no directory-creation side effect); an explicit APP_LOG_FILE_PATH still wins. Regression guard: tests/unit/logenv-datadir-path-6197.test.ts (3). (root cause independently diagnosed by @subhansh-dev in #6298 — thanks!)

• fix(docker): AgentBridge/startMitm no longer aborts in containers/headless when the Antigravity-default DNS step can't write /etc/hosts (#6127), and the privileged command's stderr now reaches app.log instead of only a bare exit code hitting the toast (#6198). The default DNS step (addDNSEntry) was called unguarded while cert install and the two sibling DNS steps were each best-effort — in the runtime Docker image (USER node, no sudo, read-only /etc/hosts) it threw Command failed with code 1 out of startMitmInternal and killed the whole start, discarding the stderr. The three DNS steps are extracted into a best-effort provisionDnsEntries() where each failure is logged with the full err (stderr included, folded in by systemCommands.ts) and never aborts the start. Regression guard: tests/unit/mitm-dns-graceful-degrade-6127.test.ts (4).

• fix(providers): copilot-m365-web now supports the M365 Education "Starter / OfficeWebIncludedCopilot" tier and no longer returns an empty content:null stream (#6210). Two gaps: (1) buildWsUrl() hardcoded the individual-consumer scenario (OfficeWebPaidConsumerCopilot, isEdu=false) — the EDU tier is now opt-in via providerSpecificData.tier="edu", emitting scenario=OfficeWebIncludedCopilot/isEdu=true (the individual path is unchanged); (2) the EDU/GPT-5.5 path streams deltas via arguments[0].writeAtCursor (incremental) instead of only messages[].text (accumulated snapshots), which the parser dropped — a new accumulateBotContent() folds both formats, with type:2 item.result.message as a last-resort fallback. Regression guard: tests/unit/copilot-m365-edu-writeatcursor-6210.test.ts (10). (thanks @qpeyba)

• fix(providers): GitLab Duo executor now feeds tool results back into the prompt instead of looping (#6220) — buildPrompt() branched only on system/user and took userParts.at(-1), silently dropping the assistant{tool_calls} + tool{result} turns the client appended, so the reconstructed prompt was byte-identical to turn 1 and the model re-emitted the same <tool> call forever. When a tool exchange is present the full conversation is now serialized, folding each tool result back keyed by its tool_call_id; simple conversations keep the legacy shape. Complements the tool_call emission from #6051 (the kilo-duplicate label was a false positive — different, sequential defect). Regression guard: tests/unit/gitlab-tool-result-feedback-6220.test.ts (4).

• fix(providers): opencode-go/opencode-zen can now synthesize the OpenCode CLI identity headers Cloudflare requires on VPS egress (#5997) — on a datacenter VPS, opencode.ai/zen/go/v1/chat/completions 403s (HTML challenge) requests lacking CLI identity, while the reporter's control curl proved that User-Agent: opencode-cli/1.0.0 + x-opencode-client: cli + x-opencode-project: default + fresh request/session UUIDs succeed. Opt-in via OPENCODE_SYNTHESIZE_CLI_HEADERS=true (values overridable via OPENCODE_GO_USER_AGENT/OPENCODE_USER_AGENT/OPENCODE_CLIENT/OPENCODE_PROJECT); it fills only headers the client did not already send. Kept off by default — the forward-only path is deliberate (fabricating a wrong value risks upstream rejection; a prior dedup regressed with opencode/local), so this replaces the fragile local header-injection shim without changing default behavior. Regression guard: tests/unit/opencode-cli-headers-synthesis-5997.test.ts (6). (thanks @aleksesipenko)

• fix(resilience): sticky session affinity now evicts and fails over to another account when the pinned account is exhausted/unavailable (#6219)

• fix(sse): Responses API passthrough now drops internal commentary-phase output before forwarding to clients (gated by RESPONSES_PASSTHROUGH_DROP_COMMENTARY, default on) (#6199)

• fix(proxy): stop the v3.8.44 proxy regression that leaked the real IP and disabled healthy proxies (#6246). Two coupled defects from the new health scheduler: (1) IP leak — when a proxy assigned to a connection was marked inactive, resolution fell through to a direct egress instead of blocking, exposing the operator's real IP; (2) over-deactivation — the sweep flipped a proxy to inactive on the first failed probe and counted our own 5s timeout / a probe-target 5xx as the proxy's fault, so healthy paid proxies vanished from egress selection ("my proxies are not being used anymore"). Fix: the sweep decision is extracted into a pure, network-free decideProxyHealthAction (src/lib/proxyHealth/decision.ts) — by default the health check now only counts/logs and never downgrades status (a proxy is downgraded/removed only with PROXY_AUTO_REMOVE=true, after PROXY_AUTO_REMOVE_AFTER consecutive conclusive failures); probes are classified tri-state so an inconclusive result (our timeout, or a 5xx from the probe target) never penalizes the proxy, and the probe timeout is raised 5s→15s. Separately, safeResolveProxy now fails closed via the existing policy: a connection whose assigned proxy is dead is blocked instead of leaking direct (hasBlockingProxyAssignment), honoring the explicit proxy off toggles and the PROXY_FAIL_OPEN=true opt-out. Existing proxies stuck inactive by the old behavior need a one-time manual re-activate (the operator owns proxy status). Regression guards: tests/unit/proxy-health-decide-action-6246.test.ts, tests/unit/proxy-assigned-unavailable-6246.test.ts.

• fix(proxy): make "Test All" read-only and add bulk enable/disable (#6246). Complements the core fail-closed / scheduler fix (#6296) with the two remaining reporter asks. (1) The "Test All" button (POST /api/settings/proxies/auto-test) used to flip a proxy to inactive on a failed reachability probe; since the egress selector excludes inactive proxies, a flaky probe (an unreachable httpbin.org, a proxy that blocks HEAD, or a slow paid proxy) silently disabled every proxy that failed — "Test All" is now read-only by default (only the operator sets a proxy active/inactive; opt back into the legacy test-and-set with PROXY_HEALTH_AUTO_DEACTIVATE=true). (2) Adds a bulk enable/disable proxies endpoint + toolbar action (POST /api/settings/proxies/batch-activate) so an operator can re-activate proxies in one click. Regression guard: tests/unit/proxy-health-6246.test.ts. (thanks @tenshiak)

• chatcore (tools): stop the default 128-tool cap from silently dropping opencode's task/MCP tools. opencode (used as an MCP/agent host) sends a large tool list; when it exceeds the speculative MAX_TOOLS_LIMIT (128) default, truncateToolList did a blind tools.slice(0, 128), dropping every tool past index 128 — including opencode's built-in task tool (subagent launch) and many MCP tools, so models routed through OmniRoute could no longer spawn subagents or reach part of their tools. The cap exists to avoid upstream 400s for providers with real hard limits (e.g. grok-cli 200), so it is kept for those: detection of the opencode client (isOpencodeClient — any x-opencode-* header, or opencode in the user-agent) now only bypasses the speculative 128 default, never a known provider ceiling. Precedence is explicit — a proactive/detected provider limit always truncates (even for opencode); otherwise opencode forwards its full tool list; otherwise the unchanged 128 default applies to every other client. Refactors getEffectiveToolLimit into getKnownToolLimit(provider) ?? DEFAULT_LIMIT (byte-identical for existing callers) and fixes a cosmetic debug-log that reported the truncated count instead of the original. Regression guard: tests/unit/tool-limit-detector.test.ts. (#6193 — thanks @DKotsyuba)

• fix(mitm): the macOS MITM-cert install check now matches the system keychain again. security find-certificate -a -Z prints the SHA-1 as a colon-less hex string, but the installed-check compared it against getCertFingerprint()'s colon-separated form, so the substring match never hit — the cert was reported as not-installed and re-prompted for the sudo install on every run. Fingerprints are now normalized (colons stripped, upper-cased) on both sides via the extracted macCertOutputHasFingerprint helper. Regression guard: tests/unit/mitm-cert-mac-fingerprint.test.ts. (#6204, closes #6134 — thanks @rianonehub)

• fix(api): /v1/messages/count_tokens now counts tool_use, tool_result and thinking content blocks (and array-form system prompts) in the local-estimation path, instead of only text. Real agentic conversations keep ~95% of their tokens inside tool results; the previous estimate returned near-zero for them, which silently broke Claude Code's auto-compaction (context grew past the window with no compaction until the upstream API rejected the request). The real provider-side count path is unchanged. Regression guard: tests/unit/messages-count-tokens-route.test.ts. (#6221 — thanks @luweiCN)

• fix(antigravity): strip a trailing assistant prefill turn for Vertex Claude models to avoid upstream 400s (#6114). Regression guard: tests/unit/antigravity-claude-prefill-strip.test.ts. (thanks @anki1kr)

• fix(security): the mutable cloud-agent routes (/api/cloud/credentials/update, /api/cloud/models/alias) now require management auth instead of being treated as public. They were classified as public API routes, so a request without management credentials could update stored cloud-agent credentials and model aliases. They are removed from the public-route set, classified as management routes in the authz pipeline, and gated by requireManagementAuth; cloud read/auth routes stay public. Regression guards: tests/unit/cloud-write-auth.test.ts, tests/unit/authz/classify.test.ts, tests/unit/public-api-routes.test.ts. (#6233 — thanks @vittoroliveira-dev)

• refactor(dashboard): extract the onboarding-wizard "Open provider details" link target into a pure, unit-tested buildProviderDetailsHref(connection) helper. The wizard already routes by connection.id (the node UUID) rather than the provider category slug (#6144/#6145); this hardens that behavior behind a tested helper that guards a missing id/connection. Regression guard: tests/unit/provider-onboarding-href.test.ts. (#6166 — thanks @KooshaPari)

• fix(api): relay worker now binds the SSRF guard to a stable const name so minified standalone (Docker) builds resolve it (#6149) — the Vercel/Deno relay generators embedded the shared resolveRelayTarget guard as a bare ${fn.toString()} declaration while the worker body called the hardcoded literal name; SWC minification mangled the source function's name, so the deployed worker defined <mangled> but still called resolveRelayTarget → ReferenceError. Both templates now emit const resolveRelayTarget = ${fn.toString()}; (the const name is a template literal, immune to minification). Regression guard: tests/unit/relay-minified-fn-6149.test.ts (4). (thanks @SeaXen)

• fix(providers): refresh the stale NVIDIA NIM model registry — drop EOL z-ai/glm-5.1, add z-ai/glm-5.2 and nvidia/nemotron-3-ultra-550b-a55b (#6108). Regression guard: tests/unit/nvidia-nim-registry-6108.test.ts. (thanks @andrea-kingautomation)

• fix(backend): GPT-family (codex) models now report a distinct max_input_tokens (272000) below their 400K context_length via an optional maxInputTokens on RegistryModel, so coding agents auto-compact correctly instead of overflowing the real input cap (#6191). Regression guard: tests/unit/gpt-max-input-tokens-6191.test.ts. (thanks @luweiCN)

• fix(backend): call logs now record a reasoning source/char-count (migration 116, reasoning_source/reasoning_chars) for models that emit reasoning_content/<think> but report zero reasoning tokens in usage, so tokens_reasoning no longer silently under-represents reasoning — cost math is unchanged (the priced tokens_reasoning stays usage-derived) (#6187). Regression guard: tests/unit/reasoning-token-source-6187.test.ts. (thanks @andrea-kingautomation)

• fix(auth): a stale/changed STORAGE_ENCRYPTION_KEY now surfaces as a clear 424 storage_encryption_stale ("re-enter the API key") instead of a misleading "Auth failed: 401" — the connection's ciphertext failed to decrypt and was coerced to an empty Bearer, hiding the real cause (#6148). Regression guard: tests/unit/decrypt-stale-key-hint-6148.test.ts. (thanks @chirag127)

• fix(backend): memory injection now keeps the injected system message first for providers that require it (via a PROVIDERS_SYSTEM_MUST_BE_FIRST capability), instead of the cache-safe mid-array splice that made strict providers reject the request with a 400 (#6135). Regression guard: tests/unit/memory-system-first-6135.test.ts.

• fix(services): 9Router embed panel no longer 404s (optional catch-all route) and the supervisor probes the port before spawning to avoid raw EADDRINUSE (#6205). Regression guards: tests/unit/ninerouter-embed-port-6205.test.ts, tests/unit/services/ServiceSupervisor.test.ts. (thanks @jonlwheat2-gif)

• fix(mcp): forward the MCP request extra context through static tool loops so stdio callers keep their scope/identity (#6178)

⚡ Performance & Infrastructure

• perf(test): test-suite loader quick wins (#6214) — the 19 test scripts switch --import tsx → --import tsx/esm (the repo is pure ESM; the unused CJS hook cost ~1.3s per test process × 2,462 processes — CI fast-path unit shards dropped 14.8→7.5 min, −49%), tsx bumped to ^4.23.0 (tsx#809 startup-regression fix), 37 orphan .test.mjs files (224 cases) recovered into the canonical glob (they matched no runner and never ran in any CI job; check:test-discovery now scans .mjs too), and ci.yml/quality.yml unit jobs now call the canonical npm script test:unit:ci:shard (single source of truth — closes two silent drifts: missing setupPolyfill import in CI and memory/+usage/ dirs absent from the fast-path glob). tests/unit/dashboard/** keeps the full tsx hook in its own invocation (@lobehub/icons es/ build internally require()s ESM-syntax files).
• ci: heavy-pipeline dedup (#6215) — the release-PR pipeline ran the unit suite 4× per sync (95 jobs, 208 machine-min; the v3.8.44 cycle fired 123 such runs, 88 cancelled). Now: Node 24/26 compat matrices move to a daily nightly-compat.yml (−28%/run; resolves the active release branch, opens a tracking issue on failure), coverage is collected inside the unit shards themselves via c8/NODE_V8_COVERAGE (−18%/run; the Coverage Shard ×8 matrix is gone — nodejs/node's own CI pattern), the ~40-job per-language i18n matrix becomes 1 job (the account has 20 concurrent-job slots total), and heavy jobs skip draft PRs — paired with /generate-release now opening the living release PR as draft (flipped ready at the new Phase 0a.0a), killing the per-merge churn for the whole cycle. Validated by a full workflow_dispatch of the new pipeline: 35 jobs, 0 failures, 23 min, merged coverage 80.16% (> ratchet baseline).
• feat(quality): no-new-warnings per PR (#6218) — native ESLint bulk suppressions (≥9.24) freeze the pre-existing debt (476 files / 4,273 violations in config/quality/eslint-suppressions.json); npm run lint, lint-staged (pre-commit) and a new fork-aware lint-guard job in quality.yml all run suppressions-aware, so a NEW warning goes red in the PR that introduces it instead of accruing invisibly (+41/+88 per cycle) and being blind-rebaselined at release. 3 warn rules promoted to error in src/** (react-hooks/exhaustive-deps, @next/next/no-img-element, import/no-anonymous-default-export); collect-metrics measures under the frozen baseline (ratchet metric = net-NEW debt; baseline tightened 4,279→0 in-PR per require-tighten); fork PRs run report-only (contributors are never blocked — the maintainer campaigns fix via co-authorship). Baseline stock shrinks via --prune-suppressions at release reconciliation.
• ci: test jobs no longer wait on the Build gate (#6275) — test-unit×8, vitest, integration×2 and security declared needs: build but never download the next-build artifact; they now start at minute 0 (needs: changes, same if as Build), cutting ~15–20 min of wall-clock per heavy run. e2e/package-artifact/electron-smoke keep needs: build (they consume the artifact for real).
• ci(build): the ci.yml Build job compiles Next.js with Turbopack (OMNIROUTE_USE_TURBOPACK=1) (#6273) — Build job 20 min → 6 min 59 s (~2.9×) on ubuntu-latest; the webpack actions/cache step is removed. Validated end-to-end pre-merge via gh workflow run ci.yml --ref <branch>.
• feat(build): Turbopack becomes the default bundler for next build and next dev (#6283) — build-next-isolated.mjs, run-next.mjs and the playwright-runner default to Turbopack; OMNIROUTE_USE_TURBOPACK=0 is the explicit webpack escape hatch. nightly-compat.yml/npm-publish.yml inherit the default. Regression guard: tests/unit/build-bundler-default-turbopack.test.ts.
• feat(docker): the Docker image builds with Turbopack (ENV OMNIROUTE_USE_TURBOPACK=1) (#6285) — the v3.8.27 ImportTracer panic ("unreachable: there must be a path to a root") does not reproduce on Next 16.2.9: amd64 (659 s) and arm64 (qemu) build clean, 0 panics, smoke health 200.
• ci: opt-in self-hosted VPS runners for the release window (#6284) — scripts/vps/release-runner-up.sh/down.sh manage the runner VM, and build/test-unit/vitest pick a dynamic runs-on gated by vars.USE_VPS_RUNNER == 'true' and own-origin (fork PRs never reach self-hosted runners). Wired into /generate-release (VM up at Phase 1, mandatory down at Phase 3).

📝 Maintenance

• quality(release-green): full pre-flight hardening for this release — the cycle's 11 net-new ESLint errors typed/fixed and validate-release-green made suppressions-aware with per-gate logs (_artifacts/release-green/) and a --hermetic mode; test-masking allowlist entries for the cycle's verified-legitimate assert reductions; stale ESLint suppressions pruned (4,273 → 4,233); the 7 net-new as any casts from #6292 typed; githubSkillTools MCP errors routed through sanitizeErrorMessage(); combo-provider-cooldown-sibling added to the Stryker tap set; executors/env docs count fixes.
• ci(quality): merge-integrity fast-gates per PR — check:changelog-integrity (no base CHANGELOG bullet may vanish in the merge result — the auto-resolve "CHANGELOG-eat" pattern) and check:agent-skills-sync (generated SKILL.md ≡ catalog), blocking for own-origin branches and report-only for forks (Princípio Zero). (#6300)
• ci(vps): hermetic nightly-release-green pre-flight on the dedicated omni-release self-hosted runner (dynamic runs-on, clean env); e2e/integration/electron stay on hosted runners (per-VM port collision + concurrent artifact-download limits documented in the PR). (#6305)
• chore(quality): v3.8.45 cycle-close drift rebaselines — file-size (13 files grown by merged cycle PRs), cognitive 867→877, cyclomatic 2028→2035, kiro-translator test debt from #6213; all with dated justification keys.
• docs(architecture): sync stale DB-layer counts (45+/55 → 95+/110+) in REPOSITORY_MAP, the db-schema diagram and llm.txt (+42 i18n mirrors). (#6167)
• chore(release): parallel-cycle flow — sync-next-cycle.mjs + Hard Rule #21 semantics (#6203); v3.8.45 development cycle opened.
• i18n(it): add 118 missing Italian (it) translations (net-additive — no existing keys dropped, valid JSON), improving Italian UI coverage. (#6212 — thanks @serverless83)
• chore(providers): remove deprecated MiMo V2 model entries from the catalogs (xiaomi-mimo, opencode-go, zenmux-free, audio TTS) — the upstream V2 line is superseded by MiMo V2.5; drops mimo-v2-tts, mimo-v2-pro, mimo-v2-omni, mimo-v2-flash, mimo-v2-flash-free and realigns the provider-catalog tests. (#6248 — thanks @backryun)

🙌 Contributors

Thanks to everyone whose work landed in v3.8.45:

Contributor	PRs / Issues
@aleksesipenko	direct commit / report
@andrea-kingautomation	direct commit / report
@anki1kr	#6078
@arssnndr	#6162, #6163
@backryun	#6154, #6235, #6248
@chirag127	direct commit / report
@DKotsyuba	#6193
@hartmark	#6216
@Iammilansoni	#6150, #6200, #6209, #6245
@jonlwheat2-gif	direct commit / report
@kanztu	#6181
@karimalsalah	#6291
@KooshaPari	#6166
@LuisAlejandroVega	#6177
@luweiCN	#6221
@Moseyuh333	#6186
@muflifadla38	direct commit / report
@powellnorma	direct commit / report
@qpeyba	direct commit / report
@RCrushMe	direct commit / report
@rianonehub	#6134, #6204
@SeaXen	direct commit / report
@serverless83	#6212
@subhansh-dev	#6298 (diagnosis, landed via #6234)
@tenshiak	direct commit / report
@Theadd	#6194, #6195
@vittoroliveira-dev	#6233
@VXNCXNX	#6213
@diegosouzapw	maintainer

---

[3.8.44] — TBD

✨ New Features

• feat(resilience): throttle upstream quota fetches on the per-request preflight path (#6009) — a new global min-interval gate (open-sse/services/quotaFetchThrottle.ts) spaces the actual network calls made by the Codex quota fetcher so that many accounts on one IP no longer fetch quota in the same second (which, per router-for-me/CLIProxyAPI#2385, can get a Codex OAuth token revoked). Complements the existing bulk-sync spacing (PROVIDER_LIMITS_SYNC_SPACING_MS) which already serialized the periodic provider-limits sync — this covers the concurrent combo/preflight path it didn't. Cache hits are never delayed; fail-open (only ever awaits a timer). Configurable via OMNIROUTE_QUOTA_FETCH_MIN_INTERVAL_MS (default 250ms, clamped 0..5000; 0 disables). Regression guard: tests/unit/quota-fetch-throttle-6009.test.ts (5). (thanks @powellnorma)
• feat(autoCombo): add per-request Auto-Combo controls via two headers (#6024 / #6025 / #6023) — X-OmniRoute-Mode steers an auto combo's scoring for a single request (friendly presets fast/balanced/quality/cheap/reliable/offline or a raw mode-pack name; balanced forces the default weights), and X-OmniRoute-Budget sets a hard per-request USD cost ceiling. Both override the combo's stored config only for the request that carries them; unknown/garbage values are ignored so the saved config is preserved. The resolvers are pure (open-sse/services/autoCombo/requestControls.ts) and feed the engine's existing config.modePack / config.budgetCap inputs — no engine changes. Regression guard: tests/unit/auto-combo-request-controls-6024.test.ts (5). (thanks @chirag127)
• feat(providers): add the Kenari OpenAI-compatible gateway (BYOK). Regression guard: tests/unit/kenari.test.ts. (thanks @doedja)
• feat(models): add claude-sonnet-5 to the Antigravity model catalog (alias mapping in antigravityModelAliases.ts) (#6103). Regression guard: tests/unit/antigravity-model-aliases.test.ts. (thanks @anki1kr)
• feat(api): add /v1/ocr endpoint (Mistral OCR), an OCR provider category, and Mistral moderation support. (#5950) (thanks @waguriagentic)
• Discovery tool (Phase 2): add the discoveryResults DB module (CRUD over the discovery_results table, migration 074) and wire the opt-in provider-discovery service to persist and read findings through it (persistDiscoveryResult, getDiscoveryResults, getDiscoveryResultById, markVerified, deleteDiscoveryResult) with (provider, method, endpoint) upsert de-duplication. Adds the /api/discovery/* HTTP surface — GET /results, GET|DELETE /results/:id, POST /scan, POST /verify/:id — under strict loopback-only authorization (/api/discovery/ is in LOCAL_ONLY_API_PREFIXES and is NOT manage-scope-bypassable, so the scan route's outbound probes can never be reached from a tunnel/remote origin). Adds a dashboard UI tab (Tools → Discovery, /dashboard/discovery) to run scans and review, verify, or delete findings. The service stays opt-in / default-off. (#5939)
• feat(api): expose a read-only provider plugin manifest at GET /api/v1/provider-plugin-manifest for sidecar/relay discovery. (#6001) (thanks @KooshaPari)
• feat(sidecar): advertise the provider manifest URL to Bifrost/CLIProxyAPI via the X-OmniRoute-Provider-Manifest-Url header (OMNIROUTE_PROVIDER_MANIFEST_URL). (#6007) (thanks @KooshaPari)
• feat(autoCombo): add a latency/speed-optimized routing mode (shared rankBySpeed scoring core) plus the omniroute_pick_fastest_model MCP tool. (#6011) (thanks @KooshaPari)
• feat(providers): refresh The Old LLM (Free) model catalog (#5181) — seed the current free /api/chatgpt tier (GPT-5/5.1/5.2/5.3/5.4, o3/o4-mini, Gemini 3 Pro / 2.5 Pro / 2.0 Flash / 1.5 Flash, Claude 4.6 Opus/Sonnet & 4.5 Haiku, GPT-4o, Grok 4, DeepSeek V3/R1, Sonar Pro) while keeping the legacy alias IDs for saved-preference compatibility. Also fixes a latent routing bug: mapModel() now passes known upstream IDs through unchanged, so Gemini/o-series/Grok/DeepSeek/Sonar models no longer silently collapse onto GPT_5_4. Regression guard: tests/unit/theoldllm-model-refresh-5181.test.ts. (thanks @WslzGmzs)
• feat(resilience): surface Codex banked reset credits per connected account (#5199) — the Codex quota parsers (buildCodexUsageQuotas, parseCodexUsageResponse) now additively read rate_limit_reset_credits.available_count (+ optional rate_limit_reached_type) from the /wham/usage payload OmniRoute already fetches, and the provider-limits dashboard renders a "Banked Reset Credits" row when a positive count is present. Display-only and fail-open — the field is eligibility-gated, so accounts without it are unaffected (parsers never throw on absent/garbage shapes); redemption (an unofficial mutating endpoint) is intentionally out of scope. Regression guard: tests/unit/codex-banked-reset-credits-5199.test.ts (8). (thanks @ofekbetzalel)
• feat(providers): add sign-up geo-restriction notices for SenseNova and StepFun (#5462) — the provider add-form now warns that SenseNova's console appears to require a Chinese (+86) phone number with no documented international path, and that StepFun's default endpoint is its China platform while a global StepFun Open Platform (platform.stepfun.ai, operated by Sparkling AI Pte. Ltd., Singapore) with email/Google/Discord login exists for international users. Informational notice only — neither provider is disabled. Regression guard: tests/unit/regional-provider-cn-notices-5462.test.ts. (thanks @chirag127)
• feat(usage): add on-demand period-scoped usage-data reset (Settings → System Storage) with a purge API and time-window selector. (#5831)
• feat(claude-code): add an opt-in auto-permission classifier compat mode (off/auto/always) for Claude Code, toggleable from the CLI Code settings. (#5810)
• feat(providers): add optional client-identity header profiles for compatible nodes — preset User-Agent/fingerprint headers (e.g. matching a known CLI) merged into the existing customHeaders field. (#5812)
• feat(build): add a backend-only fast build mode (scripts/build/build-next-isolated.mjs + backendOnlyPages.mjs) that skips compiling the dashboard frontend pages, cutting local/CI build time for backend-only changes. (#6119 — thanks @artickc)
• feat(minimax): extract MiniMax M3's raw <think>...</think> leakage into reasoning_content on the 8 OpenAI-format provider tiers, leaving the Claude-format minimax/minimax-cn tiers untouched (they already report reasoning correctly). (#6073 — thanks @KooshaPari)
• feat(services): promote Bifrost (@maximhq/bifrost — Go AI-gateway) from an env-only relay sidecar to a first-class embedded/supervised service, matching the existing cliproxy/9router model — installer, bootstrap SERVICES[] entry, migration 113 DB seed, 7 lifecycle API routes under /api/services/bifrost/ (loopback-only), a dashboard tab, and relay auto-wiring that defaults BIFROST_BASE_URL to the supervised port when running. Implements item #2 of #5670; the broader RouterBackend contract (items #1, #3-#5) stays out of scope. (#5817, part of #5670)
• feat(services): add Mux (coder/mux — local agent-orchestration daemon) as a fourth-tier embedded service on the existing ServiceSupervisor framework — npm-based installer, bootstrap.ts registration, migration 113 DB seed, 7 lifecycle API routes under /api/services/mux/ (loopback-only, defense-in-depth bind to 127.0.0.1), and a dashboard tab reusing the shared service-management components. (#6034)
• feat(xai): surface Grok/xAI usage on the quota dashboard via local usageHistory aggregation (getXaiUsage) — since xAI exposes no per-account quota API, this sums tokens routed to the connection from usage_history and reports them as a cumulative, uncapped quota, mirroring the existing Xiaomi MiMo self-track pattern. (#5806)
• feat(minimax): extract MiniMax M3's raw <think>...</think> tags into a separate reasoning_content field on the 8 provider tiers that register M3 with format:"openai" (trae, huggingchat, bazaarlink, ollama-cloud, opencode, cline, opencode-zen, codebuddy-cn) — previously the thinking text leaked directly into content. Reuses the existing extractThinkingFromContent primitive, extending its allowlist with a minimax-m3-only pattern; the two direct minimax/minimax-cn tiers are untouched since they already surface reasoning natively over Anthropic's Messages format. (Inspired by 9router#2231.) (#6050 — thanks @KooshaPari)
• feat(i18n): auto-detect the browser language on first visit — a pure detectBrowserLocale() matcher (exact match, zh-HK/zh-MO folded to zh-TW, language-prefix match, else null) plus a client-only LocaleAutoDetect component mounted once in the root layout. When no locale cookie is set yet, it reads navigator.languages, computes a match against the supported locales, and persists it via the same cookie/localStorage writer LanguageSelector already used (extracted to shared/lib/persistLocale.ts). (Inspired by 9router#1324.) (#5979)
• feat(cli-tools): add CodeWhale — the actively-maintained successor to DeepSeek TUI (same author, renamed project) — as a dual dashboard entry alongside the existing "deepseek-tui" catalog entry, so existing DeepSeek TUI users keep a working card while new users are steered to CodeWhale. New /api/cli-tools/codewhale-settings route writes ~/.codewhale/config.toml and keeps the legacy ~/.deepseek/config.toml in sync. (Inspired by 9router#1761.) (#5996)
• feat(server): support reverse-proxy basePath deployment via a new opt-in OMNIROUTE_BASE_PATH env var (empty by default), using Next.js's native basePath support so a deployment behind a reverse-proxy subpath (e.g. https://host/omniroute/) works without manual header stripping; the two hardcoded auth-redirect targets in src/server/authz/pipeline.ts now prefix with request.nextUrl.basePath. Default empty basePath is a no-op for existing root-path deployments. (Inspired by 9router#1810.) (#5992)
• feat(providers): add SumoPod (ai.sumopod.com) and X5Lab (api.x5lab.dev) OpenAI-compatible BYOK aggregator gateways, wired via the default executor with bearer API-key auth; both use passthroughModels with a live /v1/models fetcher instead of a hardcoded catalog. Regression guard: tests/unit/sumopod-x5lab-provider.test.ts. (Inspired by 9router#1288.) (#5963)
• feat(providers): add Charm Hyper (hyper.charm.land) as a new OpenAI-compatible, bearer-auth API-key gateway provider with a free tier (100 monthly Hypercredits); models resolve via passthrough (modelsUrl + live /v1/models) since the catalog isn't publicly documented. (Inspired by 9router#2006.) (#5961)
• feat(providers): add Nube.sh (ai.nube.sh) as a new BYOK OpenAI-compatible gateway (LiteLLM proxy), Bearer/API-key auth. Its live model catalog is only reachable with a valid key, so no model IDs are hardcoded — it uses passthroughModels + modelsUrl for live enumeration. (Inspired by 9router#2294.) (#5936 — thanks @whale9820)
• feat(providers): add b.ai (api.b.ai) as a new OpenAI-compatible BYOK provider, distinct from the existing thebai/theb.ai provider, using passthrough model discovery with no hardcoded model list. (Inspired by 9router#963.) (#5969)
• feat(providers): add Qiniu (七牛云) AI inference gateway as a BYOK API-key provider — proxies many upstream models (DeepSeek V3/V4, Claude, Kimi, and more) behind a single key, shipping with an empty static seed and relying on passthroughModels + the live /v1/models catalog instead of a stale hardcoded model id. Regression guard: tests/unit/qiniu-provider.test.ts. (Inspired by 9router#911.) (#5966)
• feat(providers): port ModelScope (Alibaba 魔搭) as a new API-key, OpenAI-compatible provider — verified against ModelScope's own docs that the real production domain is api-inference.modelscope.cn (.cn, not the upstream PR's .ai) and shipped passthroughModels: true with an empty seed + modelsUrl instead of the upstream PR's static 5-model snapshot, since the open-model catalog moves fast. (Ported from 9router#1764.) (#5965 — thanks @tn5052)
• feat(providers): add Augment (Auggie CLI) as a new local, no-auth provider that spawns the user's local auggie CLI and pipes a flattened prompt via stdin, wrapping stdout as an OpenAI-compatible SSE stream or single JSON body. Auth is delegated to auggie login outside OmniRoute (synthetic noAuth: true connection, no DB row required); "Test Connection" spawns auggie --version. Hardened against the untrusted-input spawn sink: no shell: true on Windows (argv passed straight to the OS loader, no metacharacter interpretation), and model is validated against the registry allowlist before spawn (rejecting unknown or --prefixed values) with a trailing -- end-of-options marker. (Inspired by 9router#1200.) (#5972 — thanks @chamdanilukman)
• feat(providers): add NVIDIA NIM image generation — a dedicated nvidia-nim image format/handler (separate host, ai.api.nvidia.com/v1/genai/<model>, native NIM body shape) for the 4 FLUX models (flux.1-dev, flux.1-schnell, flux.1-kontext-dev, flux.2-klein-4b), shaping each model's per-model request body (dimension/mode validation, required input image + aspect ratio, optional edit image) and normalizing the NIM response's varying shapes into the OpenAI {created, data} shape. (Inspired by 9router#1195.) (#5971)
• feat(oauth): import a Codex connection from a raw ChatGPT access token — OmniRoute's only Codex import path previously required both access_token and refresh_token, leaving no path for a user with only a bare ChatGPT website access token. createProviderConnection gains an explicit access_token auth-type branch (intentionally never deduped), a new POST /api/oauth/codex/import-token route (Zod-validated), and OAuthModal's manual-paste path now detects an eyJ-prefixed pasted token and posts it to the new endpoint, mirroring the existing grok-cli raw-token flow. The executor's refreshCredentials() already degrades safely to null without a refresh token, forcing re-auth on expiry. (Inspired by 9router#1290.) (#5995 — thanks @ryanngit)
• feat(dashboard): add a tool-source diagnostics settings toggle — a new Settings → Advanced card lets operators flip the existing logToolSources flag from the UI instead of editing the DB row directly; logToolSources is added to the .strict() /api/settings Zod PATCH schema (previously rejected). (Inspired by 9router#1825.) (#5978 — thanks @DuyPrX)
• feat(dashboard): collapse and sort provider quota rows by remaining percentage — the expanded quota list is sorted highest-remaining-first and collapsed to the first 3 rows by default, with a "Show N more"/"Show less" toggle when a connection reports more than 3 quotas, keeping at-risk quotas visible above a long list of healthy ones. Sort/slice logic extracted into pure, directly-unit-tested helpers (sortQuotasByRemaining, getVisibleQuotas). (Inspired by 9router#1919.) (#5977)
• feat(dashboard): suggest HuggingFace Hub media models — a new GET /api/v1/providers/suggested-models route proxies the public HF Hub models search API (Zod-validated, no token exposed client-side) and ImageExampleCard merges the results into the model picker as a selectable chip row for the huggingface provider; also adds a dedicated huggingface-image format/handler for HF's raw-image-bytes response. (Inspired by 9router#1633.) (#5990)
• feat(cli-tools): add a Crush entry to the dashboard CLI-Tools catalog plus a new /api/cli-tools/crush-settings route (GET/POST/DELETE) — OmniRoute already shipped a crush CLI setup command (bin/cli/commands/setup-crush.mjs) but the dashboard catalog had no matching entry; the new route writes to the same canonical ~/.config/crush/crush.json path so the dashboard and CLI command agree. (Inspired by 9router#1233.) (#5970)
• feat(providers): extend Vercel AI Gateway (vercel-ai-gateway/vag) beyond chat-only to support embeddings and image generation — the gateway's OpenAI-compatible /v1 API also exposes /embeddings and /images/generations, so entries were added to EMBEDDING_PROVIDERS (embeddingRegistry.ts) and IMAGE_PROVIDERS (imageRegistry.ts) modeled on the existing openai entries. (#5968 — thanks @tantai-newnol)
• feat(api-keys): add per-key device/connection tracking — a SHA-256 fingerprint of IP + User-Agent, with a 30-minute TTL and per-key/global caps, tracks distinct client devices seen with each API key (in-memory only, raw IP never stored). A new GET /api/keys/[id]/devices route exposes masked device details, and the API Keys dashboard tab gets a "Devices" count badge alongside the existing Sessions badge. This is a new granularity distinct from the existing maxSessions cap, which limits concurrent sticky-routing sessions rather than tracking device identity. (#5998 — thanks @mugni-rukita)
• feat(proxy): add Webshare (proxy.webshare.io) as a fourth source in the free-proxy provider framework alongside 1proxy, Proxifly, and IPLocate. WebshareProvider paginates the account's /api/v2/proxy/list/ endpoint, upserts proxies into the shared free_proxies table, and tombstones proxies the account no longer lists while never touching rows already promoted into the live proxy pool. Unlike the other sources, Webshare is a paid per-account list, gated on FREE_PROXY_WEBSHARE_API_KEY. (#5993 — thanks @ricatix)
• feat(antigravity): support custom Google Cloud project ID settings from the connection edit modal (Antigravity family). (#5905 — thanks @nickwizard)
• feat(dashboard): add a wildcard-CORS runtime warning banner (Settings → Authorization) when CORS_ALLOW_ALL/* origins are in effect, plus a new docs/security/CORS.md security guide covering the risk and safer alternatives. (#5602, #5759)
• feat(api): add a /v1/audio/translations endpoint (Whisper-style audio translation), a new audioTranslation handler, and translation providers wired into audioRegistry. Regression guard: tests/unit/audio-translations-route.test.ts (8, incl. no-stack-leak). (#5809)
• feat(providers): allow a custom icon URL for compatible provider nodes (migration 113 + nodes.ts + Zod schema + API routes + catalog + ProviderIcon UI). Regression guards: 14 backend + 5 frontend(vitest) + 24 page-utils tests. (#5815)
• feat(xai): register a dedicated XaiExecutor with reasoning-effort suffix parsing. Regression guard: tests/unit/executors/xai-executor.test.ts (6). (#5800)
• feat(webfetch): support self-hosted FireCrawl instances via FIRECRAWL_BASE_URL/FIRECRAWL_TIMEOUT_MS. Regression guard: tests/unit/executors/firecrawl-fetch.test.ts (4). (#5793)
• feat(providers): add ClinePass as a first-class API-key (BYOK) provider — Cline's paid gateway (cline-pass/* models, plain Bearer key), distinct from the existing OAuth cline provider. Regression guard: 16 clinepass tests. (#5942 — thanks @adentdk)
• feat(relay): gate Bifrost auto-routing by the provider plugin manifest — only manifest-eligible providers reach the sidecar; ineligible/unknown providers fall back to the existing TS routing path with explicit reasons. Regression guards: 4 provider-plugin-manifest + 11 relay-routing-backend tests. (#5870 — thanks @KooshaPari)
• feat(providers): wire Claude Sonnet 5 end-to-end across the model pipeline — registries, modelSpecs, pricing (×3), cost, Sonnet-family fallback, 1M-context, and static models. (#5833 — thanks @ggiak)

🔧 Bug Fixes

• dashboard (/dashboard/system/proxy 500 on every render): ProxyRegistryManager called useProxyBatchOperations(load) before the const load = useCallback(...) declaration in the component body, so every server render threw a TDZ ReferenceError: Cannot access 'load' before initialization and the whole proxy page 500'd (#5918 regression, caught by the release-PR e2e smoke — the PR→release fast-gates never render pages). The hook block now sits after the load declaration. Regression guard: tests/unit/ui/ProxyRegistryManager-tdz-render.test.tsx (SSR renderToString — the exact crash mode).

• server (TRACE/TRACK/CONNECT returned raw 500 on every route): methods that undici/fetch cannot represent blew up inside Next's middleware adapter (TypeError: 'TRACE' HTTP method is unsupported.) as an unhandled 500 (caught by the release-PR dast-smoke Schemathesis negative tests on the new /api/keys/{id}/devices endpoint). The raw HTTP method guard now answers a clean 405 + Allow header for these methods on any path, before Next sees the request. Regression guard: tests/unit/dast-method-not-allowed.test.ts (new case).

• i18n (auto-detect refreshed every first visit): LocaleAutoDetect (#5979) called router.refresh() on every cookie-less first visit — even when the detected browser locale was exactly the one the server had just rendered — re-navigating the page mid-interaction (flaky e2e "execution context destroyed" + a visible flash for every new visitor). It now refreshes only when the detected locale differs from the server-rendered <html lang>. Regression guard: tests/unit/ui/LocaleAutoDetect-refresh.test.tsx.

• models (oc/ alias must reach the no-auth OpenCode provider): restore the #2901 routing contract after the #5918 transitive-alias change made the registered no-auth opencode provider unreachable by any prefix (oc/ chained through the manual opencode → opencode-zen slug override and misrouted its combo entries). resolveProviderAlias now stops the alias chain as soon as a hop lands on a registered provider id, while keeping #5918's transitivity across alias-only hops and its loop/depth guards. Regression guards: tests/unit/combo-builder-opencode-prefix.test.ts, tests/unit/provider-alias-transitive-5918.test.ts.

• providers (Auggie executor EPIPE crash): a fast-exiting auggie CLI (e.g. binary present but immediately failing) delivered EPIPE asynchronously as an 'error' event on the child's stdin stream — which a plain try/catch around stdin.write() cannot catch — crashing the request instead of surfacing the sanitized CLI error. Both spawn sites now attach a stdin 'error' handler so the child's own exit/close handlers report the failure. Regression guard: tests/unit/auggie-executor.test.ts (deterministic 3/3 locally).

• dashboard (CoolingConnectionsPanel broke next build): the cooling-connections panel from #6061 imported Card from a shadcn-style path that does not exist in this repo (@/components/ui/card) and pulled the server DB barrel (@/lib/localDb) into a client component — next build failed to compile on the release branch. The panel now renders with repo-native markup and reads formatResetCountdown from the new client-safe src/shared/utils/formatting.ts. Regression guards: tests/unit/format-reset-countdown.test.ts, tests/unit/ui/CoolingConnectionsPanel.test.tsx. (#6155)

• oauth (Zed "Unknown provider" crash): adding Zed from the providers dashboard threw an unhandled OAuth GET error: Unknown provider: zed (500) (#6041). Zed is a keychain-import-only provider — it's listed in the OAuth catalog so the UI shows it, but has no OAuth handler, so the generic /api/oauth/[provider]/[action] route hit getProvider("zed") and crashed. The route now recognizes keychain-import-only providers and returns a clear 400 pointing users at the Import button (for both GET and POST OAuth actions), instead of a 500. Regression guard: tests/unit/oauth-keychain-import-only-6041.test.ts. (thanks @imblowsnow)

• fix(providers): disable the unsupported thinking param for minimax-m2.7 on NVIDIA NIM (the upstream rejects it) (#6102). Regression guard: tests/unit/nvidia-minimax-thinking-strip.test.ts. (thanks @anki1kr)

• fix(mitm): add an in-process guard so concurrent MITM server starts no longer race — a second start while one is already in flight is short-circuited instead of double-binding the listener (#6107). Regression guard: tests/unit/mitm-start-guard.test.ts. (thanks @anki1kr)

• translator (Responses → Chat Completions): strip the Responses-API-only truncation field before forwarding a /v1/responses request to a non-OpenAI Chat Completions upstream (#6109). Strict upstreams (e.g. NVIDIA NIM) rejected it with HTTP 400 Unsupported parameter(s): truncation, breaking Codex-style clients routed to those providers. client_metadata, background, and safety_identifier were already stripped — truncation was the remaining gap. Regression guard: tests/unit/responses-strip-truncation-2311.test.ts. (thanks @TuanNguyen0708)

• combo (prefer known context capacity over unknown): when a combo filters out at least one target for exceeding a known context limit, the router now prefers the remaining known-compatible targets over targets whose context metadata is simply unknown, instead of letting unknown-metadata targets be the only survivors. If no known-compatible context target remains, context-only candidates fall back to the normal strategy order. Regression guard: tests/unit/combo-context-window-filter.test.ts. (#6088 — thanks @Thinkscape)

• models (GLM-5.2 context normalization): stop treating every hosted GLM-5.2 provider alias as the native 1M-context model. Native/bare GLM-5.2 and verified OpenCode / ZenMux routes keep their 1,000,000-token context, while hosted-provider aliases now respect the caps declared in their provider metadata instead of inheriting the native max. Regression guards: tests/unit/model-capabilities-registry.test.ts, tests/unit/models-catalog-route.test.ts. (#6091 — thanks @Thinkscape)

• providers (Gemini Web): refresh the Gemini Web cookie handling and model catalog so live Gemini Web sessions keep authenticating and routing to current models. Regression guard: tests/unit/gemini-web.test.ts. (#6095 — thanks @backryun)

• providers (Perplexity Web): refresh the Perplexity Web model catalog to the current set (GPT-5.4/5.5, Claude Sonnet 5.0 / Opus 4.8, GLM-5.2, Kimi K2.6, Nemotron 3 Ultra) and update the internal mode / model_preference mappings and thinking variants so requests resolve to live upstream models. Regression guard: tests/unit/perplexity-web.test.ts. (#6106 — thanks @backryun)

• dashboard ("Update now" → Internal Server Error): clicking Update now on the dashboard home could crash the page with a blank "Internal Server Error" screen (Minified React error #31). The handler POSTs the loopback-only /api/system/version auto-update endpoint and, on a non-OK JSON response (e.g. a 403 when the dashboard is reached through a reverse proxy / non-loopback origin), passed the raw error envelope object { error: { code, message, correlation_id } } straight to notify.error(), which rendered the object as a React child and threw #31. The update-error path now funnels the body through extractApiErrorMessage() (the same safe extractor added in #5340), so a readable string always reaches the toast. Regression guard: tests/unit/ui/home-update-error-render-5991.test.ts. (#5991)

• fix(onboarding): route the provider-details link in the onboarding wizard by the node's stable id instead of the composite provider slug, which could point at the wrong provider details page for multi-account/fingerprint nodes. Regression guard: tests/unit/onboarding-wizard-details-link-6145.test.ts. (#6145 — thanks @chirag127)

• fix(cli): give setup-claude a fallback profile generator mirroring setup-codex, so profile generation no longer silently no-ops when the primary generator path is unavailable. Regression guard: tests/unit/cli/setup-claude.test.ts (new cases). (#6138 — thanks @derhornspieler)

• fix(glm): suppress a leaked </think> close marker in the GLM Anthropic transport, which was surfacing the raw reasoning-close tag in visible response content instead of being consumed as part of the thinking-block framing. Regression guard: tests/unit/glm-think-close-marker-leak.test.ts. (#6133 — thanks @dhaern)

• fix(provider-limits): close a TOCTOU race in quota-recovery clearing by moving the check-then-clear to a CAS (compare-and-swap) primitive in src/lib/db/providers.ts, so two concurrent recovery paths can no longer both observe stale state and double-clear/re-lock a connection. Regression guard: tests/unit/provider-limits-recovery.test.ts. (#6139 — thanks @janeza2)

• fix(provider-limits): clear transient rate-limit state (rateLimitedUntil, lastError, backoffLevel) as soon as quota recovers, instead of leaving stale rate-limit fields behind that could keep a now-healthy connection looking unavailable. Regression guard: tests/unit/provider-limits-recovery.test.ts. (#6128 — thanks @janeza2)

• combos (OpenCode/MiMo fingerprint accounts): expand fingerprint-scoped OpenCode/MiMo accounts into their full per-fingerprint set in the combo builder, which previously showed only the first matching account entry and hid the rest from combo target selection. Regression guard: tests/unit/combo-builder-fingerprint-expansion.test.ts. (#6092, closes #6087 — thanks @anki1kr)

• fix(auth): persist quota-preflight account lockouts until the reset window elapses, instead of losing the lockout on process restart and letting a still-quota-exhausted account be selected again immediately. Regression guards: tests/unit/sse-auth.test.ts, tests/unit/opencode-quota-fetcher.test.ts, tests/unit/usage-service-hardening.test.ts. (#6090 — thanks @Thinkscape)

• combo (fingerprint-based provider expansion): expand fingerprint-based providers into per-fingerprint combo targets (open-sse/services/combo/fingerprintExpansion.ts) so a combo referencing a fingerprint-scoped provider fans out to every matching fingerprint account instead of collapsing onto one. Regression guards: tests/unit/combo-fingerprint-expansion.test.ts, tests/integration/fingerprint-expansion.test.ts. (#6082 — thanks @pizzav-xyz)

• fix (safety-net redirect reqId crash): fix a reqId ReferenceError thrown inside the safety-net combo redirect path in src/sse/handlers/chat.ts, remove dead code in src/domain/quotaCache.ts, and rename the stray root DESING.md to DESIGN.md. Regression guard: tests/unit/chat-safetynet-reqid-6097.test.ts. (#6097 — thanks @fix2015)

• fix(compression): send a patch-only body to PUT /api/settings/compression from CompressionHub, instead of round-tripping the full settings object and risking clobbering fields changed elsewhere between load and save. Regression guard: tests/unit/ui/CompressionHub-patch-only.test.tsx. (#6077, closes #6039 — thanks @anki1kr)

• fix(codex): use access_token.exp instead of id_token.exp when computing expiresAt on Codex auth import, since the id_token can expire far sooner than the actual access token, causing imported connections to be treated as expired while still usable. Regression guard: tests/unit/codex-auth-import-expiry.test.ts. (#6084, closes #6075 — thanks @anki1kr)

• fix(security): persist the IP allow/block-list configuration (it was resetting to Disabled and clearing configured IPs on every restart/update) and actually enforce it in the authz pipeline (src/server/authz/pipeline.ts), where it was previously validated but never applied. Regression guards: tests/unit/ip-filter-persistence-6131.test.ts, tests/unit/authz/ip-filter-enforcement-6131.test.ts, tests/unit/ip-filter.test.ts. (closes #6131, #6132)

• fix (Claude tool_result adjacency): reattach an OpenAI-shaped tool_result to sit directly adjacent to its originating tool_use before translating to Claude's message format (open-sse/translator/request/openai-to-claude/toolResultAdjacency.ts), since Claude's API rejects/mishandles a tool result separated from its tool call by intervening messages. Regression guard: tests/unit/translator-openai-to-claude.test.ts (new cases). (#6035 — thanks @KooshaPari)

• fix(config): externalize ws/bufferutil/utf-8-validate in next.config.mjs so the copilot-m365-web executor's WebSocket masking path works at runtime — chat requests through it were silently timing out because the bundler was inlining ws instead of leaving it as a real Node dependency. Regression guard: tests/unit/next-config.test.ts. (#6130, closes #6062 — thanks @anki1kr, whose #6098 fix it re-lands)

• fix(registry): update grok-cli model context lengths to match the actual Grok CLI /context capacities — grok-build 128k→256k, grok-composer-2.5-fast 128k→200k — so context-aware routing stops filtering these models out for exceeding a stale, too-low limit. Registry-only. (#5913 — thanks @Chewji9875)

• fix(providers): strip an orphan tool_result (one with no preceding tool_use) on the Antigravity MITM path before translating to OpenAI format, since an unpaired tool result upstream caused request failures. Regression guard: tests/unit/antigravity-orphan-toolresult-6026.test.ts. (closes #6026, #6115)

• fix(providers): emulate OpenAI-style tool_calls in the GitLab Duo executor (new open-sse/executors/gitlabResponses.ts), since the executor previously didn't emulate tool-call semantics for Duo, breaking tool-using clients routed to GitLab Duo. Regression guard: tests/unit/gitlab-duo-toolcalls-6051.test.ts. (closes #6051, #6111)

• fix(429 / accountFallback): persist the per-account 429 cooldown cascade across the request boundary and classify OpenCode's "Monthly usage limit. Resets in N days." message as a connection-scoped quota exhaustion with an N-day cooldown (instead of a ~5s transient retry), so an exhausted account stops being re-selected until its window resets. (#6061 — thanks @KooshaPari / @anki1kr, whose superseded #6086 carried the same day-parser approach)

• combo (sibling-model fallback on per-model-quota 500s): when a combo held multiple models from the same provider (e.g. two Gemini models) and the first returned a server 500, the router retried the same locked model and surfaced a 429 "cooling down" instead of trying the sibling — markConnectionLevelExhaustion was wrongly tripped by a model-level 500 for per-model-quota providers (gemini, github, passthrough, compatible), and the retry loop didn't check isModelLocked before re-hitting the same model. Both gaps are fixed; the combo now falls through to the untried sibling model. Regression guard: tests/unit/combo/combo-target-exhaustion.test.ts (21 cases). (#5976 — thanks @hartmark)

• providers (Cline non-streaming envelope): Cline can return OpenAI-compatible chat completions wrapped as { success, data: { choices, usage, ... } }; the non-streaming path checked the top-level body for empty content before unwrapping, so a valid wrapped response could be misclassified as malformed/empty. The envelope is now unwrapped immediately after provider-envelope handling, before empty-content detection, usage extraction, and translation. Regression guard: tests/unit/cline-response-envelope.test.ts. (#6046 — thanks @KooshaPari)

• providers (kimi-web, qwen-web): align the kimi-web model catalog and request-scenario selection with www.kimi.com's live GetAvailableModels response, and stop aliasing qwen3-coder-plus on qwen-web now that it is present as its own model in the live Qwen web catalog. (#5915 — thanks @janeza2)

• translator (Antigravity/Gemini tool schemas): strip multipleOf from function-declaration parameters before forwarding to Antigravity/Gemini — it is not part of the Gemini OpenAPI 3.0 schema subset accepted upstream and triggered a hard 400 ("Unknown name multipleOf"). Added to GEMINI_UNSUPPORTED_SCHEMA_KEYS so it is stripped at every schema level; minimum/maximum are unaffected since Gemini accepts them. (Ported from 9router#2309, reported by @abil0321.) (#6052)

• translator (Kiro system prompt leak): Kiro/CodeWhisperer has no system role, so system messages were normalized into a bare user turn — the full Claude Code system prompt then appeared as raw user text, polluting model context. System-origin content is now wrapped in <system-reminder> tags before merging into the Kiro user message; real user turns are unaffected. (Ported from 9router#2306, reported by @VitzS7.) (#6053)

• fix(codex): convert Chat Completions json_schema response_format → Responses API text.format on the Codex path, and preserve an existing text.format through verbosity normalization. Regression guards: 48 translator-openai-responses-req + 8 codex-verbosity tests. (#5933 — thanks @yusufrahadika)

• fix(thinking): only inject the redacted_thinking replay block when tool_use is present and thinking is enabled, avoiding a fabricated replay block on plain (non-tool) turns. (#5945, #5953)

• fix(resilience): honor active codex session affinity over per-request reset-aware re-scoring, so an in-flight session sticks to its pinned account instead of being re-scored away mid-conversation. New src/sse/services/sessionAffinityPin.ts module. Regression guard: tests/unit/codex-session-affinity-reset-aware-5903.test.ts. (#5903, #5943)

• fix(resilience): compute per-window is_exhausted and honor the quota-exhaustion preflight for priority combos, so a combo no longer keeps routing to a target whose current window is already exhausted. New open-sse/services/combo/quotaExhaustionCutoff.ts. Regression guard: tests/unit/combo-priority-quota-exhaustion-cutoff-5923.test.ts. (#5923, #5941)

• fix(providers): strip a /v1 suffix from the base URL unconditionally in both models-discovery paths, avoiding a doubled /v1/v1/models fetch error (e.g. Api Airforce). Regression guard: tests/unit/airforce-v1-double-prefix-5899.test.ts. (#5899, #5920 — thanks @anki1kr)

• fix(api): relax provider-scoped chat completion validation on /api/providers/[provider]/chat/completions. Regression guard: tests/unit/provider-scoped-chat-completions-validation.test.ts. (#5907 — thanks @nickwizard)

• fix(providers): validate v0 Platform (Vercel) API keys via the /chats endpoint instead of a probe that rejected valid keys. Regression guard: tests/unit/provider-validation-specialty.test.ts. (#5954 — thanks @vittoroliveira-dev)

• fix(mcp): auto-recover stale streamable HTTP MCP sessions on initialize instead of failing the reconnect. Regression guard: tests/unit/mcp-session-sweep.test.ts. (#5957 — thanks @Chewji9875)

• fix(translator): enforce strict Anthropic content-block compliance when converting an antigravity → openai request. Regression guard: tests/unit/translator-antigravity-to-openai.test.ts (9). (#5935)

• fix(sse): strip ANSI/VT100 escape codes from gemini-cli stream frames using a ReDoS-safe pattern. Regression guard: tests/unit/gemini-cli-ansi-sanitization.test.ts (5). (#5934 — thanks @anki1kr)

• fix(discovery): resolve a doubled /v1 discovery path and a REDIRECT_BLOCKED probe-loop abort in the model-discovery route. Regression guard: tests/unit/provider-models-route.test.ts. (#5904 — thanks @hamsa0x7)

• fix(providers): Perplexity Web now emits real tool_calls in streaming mode — previously only non-streaming requests (hasTools && !stream) converted <tool>{...}</tool> text into OpenAI tool_calls; streaming requests (the default for agentic coding clients) got the raw <tool> text as plain delta.content and never emitted a tool_calls SSE delta. Now mirrors the chatgpt-web toolMode helpers (buildToolModeResponse()/toolCompletionToSseStream(), extended with a caller-supplied idSeed so tool-call ids stay provider-specific), buffering the completion and emitting a terminal SSE replay carrying delta.tool_calls + finish_reason: tool_calls regardless of the caller's stream flag. (#5927, #5937)

• providers (openai-family model inference no longer hijacks cataloged models): resolveModelByProviderInference() had an unconditional /^gpt-/i heuristic that hijacked any model id starting with gpt-/o1/o3 into provider openai, even when the id is cataloged under other providers — breaking bare (non-combo) requests for open-weight models like gpt-oss-120b (served by fireworks/cerebras/scaleway/byteplus/sambanova/heroku), which don't exist on openai's catalog, producing a 404 with no fallback. The heuristic is now gated on providers.length === 0 so it only fires for genuinely uncataloged openai-family ids. Regression guard: tests/unit/gptoss-provider-inference-5852.test.ts. (#5852, #5938)

• fix(providers): deepseek-web reliability — auto-refresh the session on 401/403, refresh the v2.0.0 client headers, and fix the token-kind bulk import path. Regression guards: tests/unit/deepseek-web-autorefresh-401-response.test.ts, tests/unit/bulk-web-session-import.test.ts. (#5988 — thanks @backryun)

• fix(api): guard the shared frontend API client (handleResponse in src/shared/utils/api.ts) against non-JSON error responses — it previously called response.json() unconditionally and read data.error directly, throwing an unrelated parse error (or undefined) instead of a useful message when an upstream/proxy returned a non-JSON error body. Now routes through parseResponseBody/getErrorMessage to build a safe message regardless of body shape. Regression guard: tests/unit/shared-api-utils.test.ts. (#5973)

• fix(embeddings): forward the connection-level proxy configuration to embedding requests — src/lib/embeddings/service.ts previously ignored a connection's configured proxy when making embedding calls, so proxy-only network setups leaked embedding traffic outside the proxy. Regression guard: tests/unit/embeddings-proxy-forwarding.test.ts. (#5975)

• fix(resilience): parse Retry-After from a 429's JSON body for cooldown calculation, not just the HTTP header — a new retryAfterJson.ts helper extracts a retry-after hint from common JSON error-body shapes and accountFallback.ts's cooldown path now prefers it when the header is absent. Regression guard: tests/unit/account-fallback-retry-after-json.test.ts. (Includes #6013's retry-after-json extraction.) (#5974 — thanks @KooshaPari)

📝 Maintenance

• release close (release-PR one-pass CI sweep): restore Zod validation on the provider-scoped chat route with a .passthrough() schema that keeps #5907's relaxed semantics (t06 route-validation gate); point /api/keys/{id}/devices' 401 response at the management error envelope in docs/openapi.yaml (Schemathesis schema-conformance); rebaseline i18nUiCoverage.pct 77.5→76.8 (~1352 new en.json UI keys from the cycle await the async translation workflow — same shape as the v3.8.39 rebaseline); dismiss 2 CodeQL js/incomplete-url-substring-sanitization false positives on unit-test asserts (v3.8.35 precedent).

• release close (Phase 0 pre-flight): align cycle-stale tests with merged behavior — provider count 166→167 (Kenari #6104), Linux-regenerated translate-path golden (+kenari), OpenCode quota scope provider→connection (#6061) — and absorb cycle ratchet drift (file-size caps for oauth/[provider]/[action]/route.ts 960, providerLimits.ts 998, chat.ts 1662, auth.ts 2426, with #6158 tracked to restore the oauth-route freeze). The test-masking gate gains a narrowly-scoped _deletedWithReplacement allowlist section (deletion is exempt ONLY when the declared replacement test file exists in HEAD — used for targetExhaustion.test.ts → tests/unit/combo/combo-target-exhaustion.test.ts, which has MORE coverage: 21 cases/52 asserts vs 13/37), plus 5 new gate unit tests and reduction-allowlist entries for the verified-legitimate #5958/#6088/#5816 assert migrations.

• test (deflake setup-claude): tests/unit/cli/setup-claude.test.ts failed ~50% of runs with Unable to deserialize cloned data due to invalid or unsupported version at file teardown (all subtests passed), randomly reddening Unit Tests fast-path (2/2) / Fast Quality Gates across the PR→release queue. Root cause: node --test streams each file's report to the parent as V8-serialized frames on fd 1 (stdout), and the CLI helper under test (syncClaudeProfilesFromModels) prints progress via console.log — that stdout output interleaved with the serialized frames and corrupted the stream. The test now silences the stdout-writing console methods for the file's duration (no assertion inspects stdout), making it deterministic (15/15 green locally). (#5959) (#6021)

• API validation: add a validatedJsonBody(request, schema) helper in src/shared/validation/helpers.ts that fuses JSON body parsing and Zod validation into a single call, returning either the type-narrowed data or a ready-to-return 400 NextResponse with the standard error envelope. Salvaged from the closed refactor PR #5075 (Tier 1 portable helper) with a focused 6-case regression test. Co-authored-by: KooshaPari KooshaPari@users.noreply.github.com

• repo (Windows case-conflict cleanup): remove the stale root DESIGN.md, which case-conflicted with design.md and broke checkouts/clones on case-insensitive Windows filesystems. (#6140 — thanks @backryun)

• i18n(zh-CN): translate the CHANGELOG entries and section headings, adopting zh-CN as a fully translated locale alongside the existing supporting docs. (#6043 — thanks @studyzy)

• docs (env-doc-sync base-red): document BIFROST_PORT in .env.example / docs/reference/ENVIRONMENT.md — the Bifrost embedded-service merge referenced process.env.BIFROST_PORT (default 8080) without documenting it, so check:env-doc-sync failed on the release tip and reddened Fast Quality Gates for every open PR→release. Docs-only (8d7e3e28f).

• test (CI-runner-independent translate-path golden): normalize OS/arch-derived request headers (X-Stainless-Os/X-Stainless-Arch, (OS;arch) User-Agent segments, and Antigravity's os.platform()-derived platform substring) in the provider translate-path golden snapshot, so the test no longer depends on the OS/arch of the CI runner that generated it — a Mac-literal Antigravity UA was failing on Linux CI. Regression guard: tests/unit/provider-translate-path-golden.test.ts. (#6076 — thanks @KooshaPari)

• release-green base-reds (#5695 regex + file-size rebaseline): tests/unit/ui/quick-start-api-keys-link-5695.test.ts now tolerates Prettier splitting a multi-line <Link href=...> so the step1Desc regex matches the /dashboard/api-manager link instead of skipping to step2's single-line /dashboard/providers link (test was brittle, not the code). Also rebaselines 5 files that grew via already-merged release-tip PRs in config/quality/file-size-baseline.json (ApiManagerPageClient 3017→3058, OAuthModal 969→989, cliRuntime 1090→1100, webProvidersA 805→809, deepseek-web.test 1081→1092), with shrink tracked in #3501. (#6093)

• release close (LEDGER-4 base-red): the cline-pass provider's minimax-m3 registry entry was missing supportsVision, breaking the LEDGER-4 registry-consistency test (every minimax-m3 entry must set supportsVision to match lite.ts — the model is multimodal). Flagged it to match every other minimax-m3 entry (trae, bazaarlink, cline, ollama-cloud, ...). (#6003)

• release close (stryker tap.testFiles drift): additional release-green cleanup clearing the qoder registry's minimax-m3 supportsVision LEDGER-4 base-red and stryker.conf.json's tap.testFiles drift. (#6012)

• install (pnpm 11+ support): pnpm 11 introduced ERR_PNPM_IGNORED_BUILDS for native addon packages — without explicit allowBuilds approval, packages silently skip their build scripts and OmniRoute fails to start with missing native modules. Sets allowBuilds=true for all 13 native addon packages in pnpm-workspace.yaml (@parcel/watcher, @swc/core, better-sqlite3, core-js, esbuild, keytar, koffi, libxmljs2, onnxruntime-node, protobufjs, sharp, tls-client-node, unrs-resolver) and migrates onlyBuiltDependencies from the deprecated package.json field to a new pnpm.json. (commit 39349da18 — thanks @chirag127)

• refactor (Block J hot-path decomposition): extract pure leaves with no behavior change from the executor, translator, combo, and SSE hot paths — orphaned executor tests moved to top-level so a runner collects them, and handleComboChat's auto-strategy/target-timeout regions split into named helpers. (#6063, #6049, #6036, #6030, #6020, #6018, #6017, #6016, #6015, #6014, #6008, #6006, #6000, #5999, #5994, #5967, #5962, #5960, #5947, #5949, #5940, #5932)

• chore (quality/CI housekeeping): rebaseline residual ESLint/cognitive-complexity/file-size drift accumulated over the v3.8.44 cycle, move orphaned executor tests to a top-level location so a runner actually collects them, harden the release pipeline with a test-masking pre-flight gate plus contributors/uncovered helpers, and make the pr-evidence FAIL output tell the author to push (a body edit alone does not re-run the gate). (#5926, #5944, #5952, #6027, #5928, plus a #5975-collateral test hardening pinning a seeded connection to direct egress in route-edge-coverage)

• docs (housekeeping): normalize mixed-language documentation content, restore the OpenAPI coverage ratchet by documenting 9 newly-added routes, record Hard Rule #22 (cross-session safety — git stash + in-flight PR bans), and document the compression-engine's upstream sync policy for the RTK/Caveman engines. (#6105, #5955, #5948, plus docs-only commit 926b08aa8)

🙌 Contributors

Thanks to everyone whose work landed in v3.8.44:

Contributor	PRs / Issues
@adentdk	#5942
@anki1kr	#5899, #5920, #5934, #6039, #6061, #6062, #6075, #6077, #6084, #6086, #6087, #6092, #6098, #6130
@artickc	#6119
@backryun	#5988, #6095, #6106, #6140
@chamdanilukman	#5972
@Chewji9875	#5913, #5957
@chirag127	#6145
@derhornspieler	#6138
@dhaern	#6133
@doedja	direct commit / report
@DuyPrX	#5978
@fix2015	#6097
@ggiak	#5833
@hamsa0x7	#5904
@hartmark	#5976
@imblowsnow	direct commit / report
@janeza2	#5915, #6128, #6139
@KooshaPari	#5870, #5974, #6035, #6046, #6050, #6061, #6073, #6076, #6086
@mugni-rukita	#5998
@nickwizard	#5905, #5907
@ofekbetzalel	direct commit / report
@pizzav-xyz	#6082
@powellnorma	direct commit / report
@ricatix	#5993
@ryanngit	#5995
@studyzy	#6043
@tantai-newnol	#5968
@Thinkscape	#6088, #6090, #6091
@tn5052	#5965
@TuanNguyen0708	direct commit / report
@vittoroliveira-dev	#5954
@waguriagentic	direct commit / report
@whale9820	#5936
@WslzGmzs	direct commit / report
@yusufrahadika	#5933
@diegosouzapw	maintainer

---

[3.8.43] — 2026-07-02

✨ New Features

• usage (quota percentages + provider USD drilldown): @@om-usage and the HTTP usage endpoint now report personal API-key quotas as remaining percentages (USD amounts stay out of the command output), provider quota remaining is scaled by the configured quota cutoff so the protected reserve reads as 0% left, and the quota dashboard regains a provider USD cost drilldown (/api/usage/provider-window-costs + ProviderUsdCostModal, management-auth gated). Also honors observed provider quota resets: a same-resetAt reset (usage dropping back to the reset floor) is detected and preferred over stale recorded weekly events for provider USD windows and API-key USD quotas. New src/lib/usage/providerWindowCosts.ts. Regression guards: tests/unit/provider-window-costs.test.ts, tests/unit/internal-usage-command.test.ts, tests/unit/api-key-usage-limits.test.ts, tests/unit/lib/quota-reset-events.test.ts. Extracted from #5863 by @Witroch4.

• dashboard (live WS behind reverse proxy): the live dashboard WebSocket can now be fronted by a reverse proxy or Cloudflare Tunnel via NEXT_PUBLIC_LIVE_WS_PUBLIC_URL (e.g. wss://ws.my-ai.com/live-ws). The URL is honored both at build time (env inlined into the bundle) and at runtime for prebuilt Docker/npm images: the /api/v1/ws?handshake=1 handshake now echoes a lazily-read live.publicUrl (only ws:///wss:// values are accepted; anything else is rejected to null), and useLiveDashboard resolves the URL from that handshake before connecting, falling back to the previous ws(s)://hostname:20129 default. Also documents LIVE_WS_ALLOWED_HOSTS and aligns the GitLab Duo OAuth scopes line in .env.example with the live config (ai_features read_user). Regression guard: tests/unit/live-ws-public-url.test.ts (5). (#5877 by @ianriizky)

• providers (CLI profile auto-sync): opt-in toggles to auto-regenerate CLI tool profiles after a provider model sync. When enabled, a model-catalog change (re)writes that tool's profile files from the live catalog — Codex (~/.codex/*.config.toml) and now Claude Code (~/.claude/profiles/<name>/settings.json, via an extracted syncClaudeProfilesFromModels + a new claudeProfileAutoSync.ts mirroring the Codex path). Both are off by default and never touch the active/default CLI config; they are backed by the OMNIROUTE_AUTO_SYNC_CODEX_PROFILES / OMNIROUTE_AUTO_SYNC_CLAUDE_PROFILES feature flags (DB/dashboard override > env > default "false") and additionally gated behind the existing CLI_ALLOW_CONFIG_WRITES write-guard. A "CLI profile auto-sync" card on the CLI Code dashboard toggles each (moved from the providers dashboard in #5778 — thanks @rdself). Regression guards: tests/unit/claude-profile-auto-sync-gate.test.ts, tests/unit/codex-profile-auto-sync-gate.test.ts, tests/unit/cli/setup-claude.test.ts (follow-up to #5737).

• cli (startup banner): the serve startup banner now prints the running OmniRoute version (v3.8.x) beneath the ASCII logo, so the active version is visible at a glance without a separate --version call. Regression guard: tests/unit/cli-serve-version-banner.test.ts. Thanks @chirag127 (#5752).

• analytics (subscription cost): flat-rate providers now show $0 in cost analytics instead of an inflated per-token estimate. Subscription / coding-plan providers (every cookie-web provider — ChatGPT Web, grok-web, … — plus the dedicated Minimax Coding, Kimi Coding, GLM Coding, Alibaba Coding Plan, and Xiaomi MiMo plans) bill a flat fee, not per token, yet still carry per-token pricing rows used for estimates — so the analytics dashboard over-reported their cost. A new flat-rate classifier (src/lib/usage/flatRateProviders.ts) is consulted by the analytics surfaces (analytics route, usage stats, usage analytics) via an opt-in flatRateAsZero cost option, so those providers read $0 while budget / quota / routing keep estimating unchanged. Deliberately NOT zeroed: codex/cx (OmniRoute actively tracks Codex token cost — Fast-tier multipliers, GPT-5.x pricing — and Codex can be a metered account), byteplus (metered ModelArk), minimax-cn (metered China API). Regression guard: tests/unit/flat-rate-cost-5552.test.ts. (#5552)

• mcp (RTK): expose the RTK tool-output learn/discover workflow as two new MCP tools so an agent can grow the RTK filter catalog without leaving the protocol. omniroute_rtk_discover analyzes recently captured raw tool output (discoverRepeatedNoise / suggestFilter) and returns candidate noise patterns plus a suggested filter; omniroute_rtk_learn lists the captured command samples (listRtkCommandSamples) and resolves a command to its RTK filter id (commandToId). Both are read-only (scope read:compression), wrap the existing RTK discovery primitives (no new logic in the engine), and log to the MCP audit trail. Regression guard: tests/unit/compression/rtk-mcp-tools.test.ts (4). gaps v3.8.42 — T07.

• compression (LLM tier): add an opt-in, default-off LLM-tier compression engine (llm) that condenses the prose of non-system messages via a pluggable chat-completion backend. It mirrors the llmlingua engine's contract but is safe by construction: the default backend is a no-op pass-through (the engine never mutates the payload until an operator both enables it and wires a real backend via setLlmCompressorBackend()), it is not part of the default stacked pipeline, enabled defaults to false, fenced code blocks and system messages are never sent to the model, and every backend error fails open (the original segment/body is kept, never thrown). A minTokens floor skips small prompts. The real production backend is intentionally a VPS-validated follow-up (Hard Rule #18), exactly as the llmlingua worker backend is gated. New open-sse/services/compression/engines/llm/index.ts. Regression guard: tests/unit/compression/llm-compressor-engine.test.ts (8). gaps v3.8.42 — T05/C3.

• memory (typed decay): add opt-in typed memory decay (TV6) so the conversational memory store stops accumulating stale episodic noise. Each injected memory now tracks an access_count + last_accessed_at (always-on, non-destructive telemetry; migration 111_memory_typed_decay), and an opt-in, default-off sweep (MEMORY_TYPED_DECAY_ENABLED, default false) deletes memories that are past a per-type TTL and not immune. Only episodic decays by default (30d, env-tunable); factual/procedural/semantic are immune, and any memory accessed >= 3 times earns access immunity (mirroring "guardrail/convention/decision never decay"). The decay clock re-bases on the last access, so used memories survive. Deletions reuse deleteMemory (SQLite + sqlite-vec + Qdrant stay in sync) and fail open; an optional periodic sweep is doubly opt-in (also needs MEMORY_TYPED_DECAY_SWEEP_INTERVAL>0). With the flag off nothing is ever deleted (Rule #20 spirit). New src/lib/memory/typedDecay.ts. Regression guard: tests/unit/memory/typed-decay.test.ts (15). gaps v3.8.42 — T10/TV6.

• dashboard (combos): the named-combos editor now lets you drag to reorder the stacked-compression pipeline instead of only editing fixed-position steps. A new pure model (src/shared/components/compression/compressionPipelineModel.ts) owns add/remove/move/update with the engine→intensity invariant and a never-empty guarantee, and a @dnd-kit/sortable editor (CompressionPipelineEditor.tsx, matching the sidebar reorder pattern) replaces the inline list in CompressionCombosPageClient. Order persists through the existing combos endpoint. Regression guards: tests/unit/compression-pipeline-model.test.ts (11) + tests/unit/ui/compression-pipeline-editor.test.tsx (4). A dedicated tests/e2e/compression-studio.spec.ts (Tela A render + tab switch) closes the studios e2e gap the combo-live spec did not cover. gaps v3.8.42 — T06 + T03.

• compression (pipeline): add an opt-in, default-off per-engine circuit-breaker to the stacked compression pipeline (T02). When an engine throws repeatedly across requests, its breaker opens and the stacked loops skip that engine (keeping the body verbatim for that step — fail-open) for a cooldown, then probe once (lazy half-open); success closes it, a failed probe re-opens it. This is distinct from the provider circuit-breaker (src/shared/utils/circuitBreaker.ts, provider-scoped + DB-persisted) — the new pipelineEngineBreaker.ts is engine-scoped, process-local, and adds zero DB/IO on the hot path. It composes with the existing per-request TV1 bail-out (which skips within a single request); the breaker adds cross-request memory. Default off (COMPRESSION_PIPELINE_BREAKER_ENABLED=false) → byte-identical to the pre-breaker pipeline (a throwing engine still propagates unless TV1 is separately enabled). Configurable per-call, per-CompressionConfig, or via env (_THRESHOLD/_COOLDOWN_MS). Regression guard: tests/unit/compression/pipeline-circuit-breaker.test.ts (9, incl. a throwing-engine integration); existing strategySelector/bail-out suites stay green. gaps v3.8.42 — T02 (2.2).

• compression (CCR): the CCR retrieval-feedback (H8) is now graduated instead of a binary cliff. Previously a block retrieved >= 3 times was flagged do-not-compress and everything below that stayed fully compressible. Now each prior retrieval raises a block's effective minChars linearly (effectiveMinChars), so frequently-retrieved content is compressed progressively less; the >= 3 exclusion is preserved (as Infinity). The ramp is controlled by a retrievalRampFactor (default 2, per-combo config or COMPRESSION_CCR_RETRIEVAL_RAMP_FACTOR); 1 reproduces the exact legacy binary behavior. Per-(principal, hash) isolation is unchanged. Regression guard: tests/unit/compression/ccr-retrieval-ramp.test.ts (12); existing CCR suites (51) stay green. gaps v3.8.42 — T08/H8.

• compression (cache-aware): add an opt-in, default-off usage-observed prefix freeze (H5). The cache-aware guard previously preserved the system prompt only for providers a static heuristic recognized as caching. It now also learns which system prompts actually recur: once a system prompt has been observed >= a threshold across requests, it is treated as a stable cacheable prefix and preserved from compression even for providers the static check misses — recovering prompt-cache hits that a prefix-compressing mode would otherwise bust. Content-addressed by a hash of the system prompt (OpenAI / Claude / Gemini shapes), in-memory + bounded, zero DB/IO; a "freeze" only preserves the prefix, so it never mutates a payload. Default OFF (COMPRESSION_PREFIX_FREEZE_ENABLED, threshold _THRESHOLD); respects the never preserve-mode (never freezes). New open-sse/services/compression/prefixFreeze.ts, wired into resolveCacheAwareConfig. Regression guard: tests/unit/compression/prefix-freeze.test.ts (10); 44 existing cache-aware / preserve-mode tests stay green. gaps v3.8.42 — T08/H5.

• compression (read-lifecycle): add a new opt-in, default-off read-lifecycle engine (H7) that collapses stale/superseded file-Read tool results. In agentic conversations the same file is Read repeatedly; an earlier Read becomes stale once the same path is re-read (superseded by a newer view) or modified by a later Write/Edit. The engine replaces those earlier Read results with a short stub — keeping only the current (last, un-superseded) Read intact — recovering the tokens the model no longer needs. Unlike session-dedup (identical-content) or ccr (reversible markers), this is semantic + lossy, so it is opt-in (enabled defaults false). Conservative by construction: matches only well-known Read/Write tool names, compares exact paths, collapses a Read only when a strictly-later invocation touches the same path, and fail-opens on any unexpected shape. Supports both the Anthropic (tool_use/tool_result) and OpenAI (tool_calls + role:"tool") shapes. New open-sse/services/compression/engines/readLifecycle/index.ts. Regression guard: tests/unit/compression/read-lifecycle.test.ts (10). gaps v3.8.42 — T08/H7.

• observability (correlation IDs): requests now carry a correlation id threaded through logs so a single request can be traced end-to-end across the pipeline. (#5834 — thanks @hartmark)

• cli (startup banner — boot time): the serve ready banner now shows how long startup took, so slow-boot conditions are visible at a glance. (#5799 — thanks @ishatiwari21)

• api (quota-policy bypass scope): add an opt-in API-key provider quota-policy bypass scope, so a designated key can be exempted from provider quota enforcement without disabling quotas globally. (#5731 — thanks @Witroch4)

• providers (Ollama local): add a first-class Ollama local-provider card to the providers dashboard so the local LLM runtime can be configured like any other provider. (#5712 — thanks @diegosouzapw)

• codex (fallback profiles): generate fallback CLI profiles for Codex-compatible models so compatible models get a usable profile automatically. (#5701 — thanks @skyzea1)

• api (response-body validation + failover): add a configurable response-body validation step that can fail a target over to the next candidate when the upstream returns a structurally-invalid body (routing/#4985). (#5684 — thanks @diegosouzapw)

• providers (SenseNova): complete the SenseNova free Token Plan — chat completions plus Text-to-Image (ported from 9router#2233). (#5679 — thanks @diegosouzapw)

• db (self-correcting context windows): add self-correcting model context-window overrides so a model whose advertised context length is wrong is corrected automatically (models/#5004). (#5667 — thanks @diegosouzapw)

• routing (latency strategy): optimize the latency routing strategy using observed per-target performance metrics for better candidate selection. (#5629 — thanks @KooshaPari)

• compression (preserveSystemPrompt mode): add a preserveSystemPrompt mode enum (always | whenNoCache | never) with legacy back-compat, giving operators explicit control over when the system prompt is protected from compression (T05/C5). (#5653 — thanks @diegosouzapw)

• commandCode (vision): add multimodal image support for Command Code vision models. (#5557 — thanks @Stazyu)

• compression (read-lifecycle engine): T08/H7 (2.5) — an opt-in read-lifecycle engine that collapses superseded file reads so stale earlier reads of the same file are pruned from the context. (#5754 — thanks @diegosouzapw)

• compression (usage-observed prefix freeze): T08/H5 (2.4) — opt-in prefix freeze driven by observed usage, keeping a stable cached prefix from being rewritten by downstream engines. (#5744 — thanks @diegosouzapw)

• compression (CCR retrieval-feedback ramp): T08/H8 (2.3) — a graduated Context-Compression-Ratio retrieval-feedback ramp that tunes compression aggressiveness from retrieval signals. (#5739 — thanks @diegosouzapw)

• compression (per-engine circuit breaker): T02 — an opt-in per-engine pipeline circuit-breaker that disables a misbehaving compression engine without failing the whole pipeline. (#5735 — thanks @diegosouzapw)

• compression (LLM-tier engine): T05/C3 — an opt-in LLM-tier compression engine that uses a model pass for higher-ratio semantic compression. (#5702 — thanks @diegosouzapw)

• dashboard (compression pipeline editor): T06/T03 — a drag-to-reorder compression pipeline editor plus a compression-studio e2e flow. (#5727 — thanks @diegosouzapw)

• memory (typed decay): T10/TV6 — opt-in typed memory decay so aged, low-value memories fade on a per-type schedule. (#5723 — thanks @diegosouzapw)

• mcp (RTK tools): T07 — expose the RTK learn/discover capabilities as first-class MCP tools. (#5691 — thanks @diegosouzapw)

• providers (CLI profile auto-sync): opt-in CLI profile auto-sync toggles, including Claude Code auto-sync, so generated CLI profiles can track provider changes automatically. (#5755 — thanks @diegosouzapw)

🔧 Bug Fixes

• fix(opencode): stop fabricating User-Agent: opencode/local and x-opencode-client: cli headers when the client sends none — the executor-dedup refactor (#5720) accidentally re-introduced header fabrication, violating the forward-only contract (inventing opencode-internal values risks upstream rejection). Restored to forward-only: those headers are emitted only when a real client source is present. Regression guard: tests/unit/opencode-executor.test.ts. (thanks @diegosouzapw)

• fix(executors): resolveEffectiveKey returns undefined (not "") when no API key is present — a type-coercion cleanup (#5798) changed apiKey ?? "" to satisfy the typechecker, silently mutating auth-key resolution semantics. Widened the return type to string | undefined and reverted the coercion so OAuth-only credentials resolve correctly. Regression guard: tests/unit/refactor-buildHeaders-preamble.test.ts. (thanks @diegosouzapw)

• fix(translator): restore the terminal message_delta + message_stop on Responses→Claude streams — the doubled-tool-args dedup (#5828) guarded the finish handler on the shared state.finishReason, which the openai-responses→openai leg sets first in the hub path, so the openai→claude leg dropped its terminal events and the stream ended after content_block_delta. The dedup now uses a dedicated state.claudeFinishEmitted flag. Regression guard: tests/unit/claude-code-rendering-fixes.test.ts. (thanks @diegosouzapw)

• fix(pricing): add the Kiro claude-sonnet-5 pricing row so the newly-catalogued model (#5796) no longer reports $0.00 usage. Regression guard: tests/unit/catalog-updates-v3x.test.ts. (thanks @diegosouzapw)

• fix(github): keep Copilot access-token sessions active. GitHub Copilot device-flow accounts can hold a GitHub access token plus a short-lived Copilot token without a refresh token; the proactive health check treated that as terminal no_refresh_token and marked the connection expired minutes after login. The health check now keeps those sessions active, clears stale no_refresh_token state, and refreshes the Copilot sub-token when needed. Regression guard: tests/unit/token-health-no-refresh-token-expired-5326.test.ts. Extracted from #5863 by @Witroch4.

• fix(kiro): bound the Claude model-id dash→dot normalization to a 1–2 digit minor so date-suffixed ids (e.g. claude-opus-4-20250514) are no longer corrupted. (thanks @voravitl)

• fix(usage): preserve (bounded) tool definitions in request logs even when the request body is truncated, so the request-details view can still show available tools. (thanks @noir017)

• fix(providers): route OpenAI responses-only models to /v1/responses instead of 404ing on /v1/chat/completions. The curated gpt-5.5-pro / gpt-5.4-pro entries never worked (OpenAI only serves *-pro reasoning models via the Responses API), and "Test all models" surfaced the same 404s. The registry entries now carry targetFormat: "openai-responses" (reusing the existing per-model translation plumbing shared with gh/codex), DefaultExecutor.buildUrl swaps the openai endpoint to /responses in lockstep (honoring custom base URLs), and a -pro suffix heuristic covers dynamically-synced ids such as o1-pro / gpt-5.2-pro (same spirit as the gh executor's /codex/i routing, 9router#102). Legacy completions-only ids (e.g. gpt-3.5-turbo-instruct) are out of scope — they are not in the catalog and OmniRoute has no legacy /v1/completions upstream. Regression guard: tests/unit/openai-responses-only-models-5842.test.ts (8). Thanks @maikokan. (#5842)

• fix(image): keep bare codex image aliases (e.g. gpt-5.5) resolving to the codex image pipeline even when a combo shares the same name. A chat combo named gpt-5.5 used to shadow the bare image alias in resolveImageRouteModel, hijacking /v1/images/* requests to a chat target (regression path adjacent to #5887); codex bare models are now reserved before bare-combo resolution, while non-codex aliases (e.g. gpt-image-2) remain user-shadowable (#3214/#3215 behavior preserved). Regression guard: tests/unit/image-routes-combo-edits-3214-3215.test.ts (9). (#5902 by @KooshaPari)

• fix(ci): re-green the release/v3.8.43 fast-gates queue — every PR→release was inheriting base-reds (#5798). Five distinct blockers cleared: (1) stale modelContextOverrides entry in the check:db-rules intentionally-internal allowlist (#5827 allowlisted it while the #5609 fix re-exported it from localDb.ts; the re-export stays, the obsolete entry goes, classification guard re-pinned to 33); (2) LIVE_WS_ALLOWED_HOSTS / NEXT_PUBLIC_LIVE_WS_PUBLIC_URL documented in docs/reference/ENVIRONMENT.md (env/docs contract, from #5877); (3) the Router Backends ADR's references to the not-yet-merged registry (#5868) marked as landing-with-PR so check:fabricated-docs --strict passes; (4) antigravity-429-quota-tdd + middleware-header-strip-5849 added to stryker tap.testFiles (check:mutation-test-coverage); (5) file-size / complexity / cognitive-complexity ratchets rebaselined with justification notes — all drift measured identical on the pristine tip and this PR (net-zero). Regression guard: tests/unit/check-db-rules-classification.test.ts. (#5798)

• providers (codex image auto-routing regression): an unprefixed gpt-5.5 request from a codex-only setup (no OpenAI connection) now correctly infers the codex provider again — the OpenAI static-catalog short-circuit in resolveModelByProviderInference was preempting the codex-preference block, so gpt-5.5 (added to the OpenAI catalog) stopped auto-routing to Codex image generation. Users with an active OpenAI connection are unaffected (OpenAI stays default). Regression guard: tests/unit/codex-gpt55-routing-5887.test.ts. (#5887)

• api (proxy header hygiene): upstream x-middleware-* control headers (emitted by providers hosted behind Next.js, e.g. synthetic.new) are now stripped from proxied responses instead of forwarded verbatim — forwarding x-middleware-rewrite made Next 16 throw NextResponse.rewrite() was used in a app route handler and return 500 despite a successful upstream call. Applies to both streaming and JSON paths. Regression guard: tests/unit/middleware-header-strip-5849.test.ts. (#5849)

• docs (pnpm global install): replaced the unsupported pnpm approve-builds -g step with the install-time pnpm add -g omniroute@latest --allow-build=better-sqlite3 flag across README + Setup Guide (and i18n mirrors), fixing native-build approval for pnpm v11 global installs. (#5554)

• dashboard (token badge): the red "Token Expired" connection badge no longer flashes for OAuth refresh-capable providers (Antigravity/Gemini) whose access token merely lapsed but is auto-refreshed — it now shows only when the connection is terminally expired (testStatus === "expired"). Continuation of #5326. Regression guard: tests/unit/ui/connection-row-token-badge-5836.test.tsx. (#5836)

• db (auto backup toggle): full pre-write SQLite backups now honor the persisted backup.autoBackupEnabled dashboard setting — previously only the DISABLE_SQLITE_AUTO_BACKUP env var was checked, so disabling auto-backup in the UI had no effect and ~70MB pre-write snapshots kept firing. Manual and pre-restore backups still always run. Regression guard: tests/unit/db-backup-autobackup-setting-5871.test.ts. (#5871)

• providers (auto/ routing for custom providers): custom OpenAI-/Anthropic-compatible providers (dynamic *-compatible-* connection IDs) are no longer excluded from auto/ routing — the Auto-Combo virtual factory previously skipped any connection whose provider was absent from the static registry. It now falls back to the connection's defaultModel. Regression guard: tests/unit/auto-custom-provider-5873.test.ts. (#5873)

• middleware (hook sandbox): operator-authored pre-request hook code now runs inside a hardened Node vm sandbox (minimal context, no ambient globals/process.env, execution timeout, no require) instead of new Function() in the main process — closing the Hard Rule #3 / SonarCloud S1523 exposure. Regression guard: tests/unit/middleware-hook-sandbox-5872.test.ts. (#5872)

• mcp-server (auth forwarding): the per-caller MCP identity forwarded via withMcpHttpAuthContext now wins over the static OMNIROUTE_API_KEY env fallback in the internal-fetch helpers (apiFetch, omniRouteFetch) — previously the env key was spread after the forwarded headers and clobbered the caller's Authorization. Regression guard: open-sse/mcp-server/__tests__/httpAuthContext.test.ts. (#5819)

• dashboard (Modal provider — two-field auth): the Modal provider connection form now exposes two fields — Token ID + Token Secret — instead of a single API-key input, since Modal authenticates with Authorization: Bearer <token-id>:<token-secret>. The dashboard combines the two fields into the id:secret credential before saving (combineModalCredential, trims both parts), while a value pasted in the legacy single-field format keeps working verbatim (empty secret → passthrough), so existing saved connections need no migration; the key-help link points at Modal's token settings. Regression guard: tests/unit/modal-credential-combine.test.ts (5). (#5881, closes #5446) Follow-up: the Validation Model Id field is now pre-filled for Modal with the same model the server-side validator probes (Qwen/Qwen3-4B-Thinking-2507-FP8, shared via MODAL_DEFAULT_VALIDATION_MODEL_ID in src/shared/constants/modal.ts), closing the last checklist item of #5446. Regression guard: tests/unit/modal-validation-model-prefill.test.ts.

• api (chat completions — early SSE keepalive gate): the /v1/chat/completions route wrapped the response in the early-stream keepalive whenever stream was not explicitly false, so a client that omitted stream and asked for JSON (Accept: application/json) could receive premature SSE framing. The keepalive wrapper is now gated on an explicit stream: true in the body or an Accept header that forces SSE (acceptHeaderForcesStream); the parsed body is passed to the chat handler untouched, so the actual stream/JSON framing stays decided by chatCore/resolveStreamFlag — preserving OmniRoute's legacy streaming default when stream is omitted and the per-key streamDefaultMode: "json" opt-in. Regression guard: tests/unit/chat-combo-live-test.test.ts ("returns JSON without early SSE framing when stream is omitted and Accept is application/json"). (#5866 by @rdself)

• fix(github): drop a trailing assistant prefill before dispatching to GitHub Copilot chat to avoid 400 errors. (thanks @baslr)

• fix(oauth): prevent cross-IdP account overwrites by disambiguating OAuth connections on username when present, not email alone. (thanks @KunN-21)

• fix(mitm): best-effort revert privileged /etc/hosts entries on exit when a sudo password is cached, instead of always leaving orphaned state. (thanks @manhdzzz)

• providers (Kiro — Claude Sonnet 5): the Kiro provider's model catalog was missing claude-sonnet-5, so the model could not be selected or routed even on accounts that already had access to it ("claude-sonnet-5 is not supported"). Added the model to the Kiro registry (open-sse/config/providers/registry/kiro/index.ts) as a 1M-context / 128K-output Claude model, mirroring the existing Claude entries; the registry models[] feeds both the model selector and the live CodeWhisperer ListAvailableModels fallback, so the model is now selectable and routable. Regression guard: tests/unit/kiro-claude-sonnet-5-2267.test.ts. (thanks @openbioinfo)

• settings (model aliases — self-heal after restart): the Settings → Routing page showed "No exact-match aliases configured" after a server restart even though the aliases were persisted in the DB. Aliases are held in a module-local _customAliases map in modelDeprecation.ts that the boot path hydrates, but Next.js compiles the app-route module graph separately from the startup graph (the same webpack chunk-splitting class as #5312), so the GET /api/settings/model-aliases handler read a different, un-hydrated copy. The handler now self-heals: when its in-memory alias map is empty it reads settings.modelAliases from the DB (via the existing getSettings() db module — no raw SQL in the route) and repopulates the map, so the UI reflects the persisted aliases on the first GET after a restart. Follow-up: the root cause is now also fixed — the _customAliases store in modelDeprecation.ts is backed by globalThis (key __omniroute_customAliases__), so the startup and app-route module graphs share one store and the route reads the boot-hydrated aliases directly (the DB self-heal remains as a harmless fallback), mirroring the same globalThis singleton pattern already applied to thinkingBudget.ts/backgroundTaskDetector.ts (#5312). Regression guards: tests/unit/model-aliases-settings-route-selfheal.test.ts + tests/unit/model-aliases-globalthis-5777.test.ts. (#5777 — thanks @jleonar2)

• providers (grok-cli token auto-refresh): grok-cli OAuth tokens were never proactively refreshed before their real expiry. mapTokens hardcoded expiresIn: 21600 (6 h) regardless of the token's actual lifetime, so the persisted expiresAt was always "now + 6 h" and the proactive tokenHealthCheck sweep (refresh when expiresAt - now < 5 min) fired 6 h after import instead of shortly before the token really expired. mapTokens now computes expiresIn from the authoritative expires_at field in ~/.grok/auth.json (ISO → epoch-seconds) with a fallback to the JWT exp claim (payload-only decode, no signature trust); the hardcoded 21600 is kept only when neither is present. An already-expired token (real expires_at/exp in the past) is now clamped to a positive expiresIn via Math.max(1, …), so the import route stores a near-future expiresAt and AutoCombo refreshes the connection instead of reading a past date and excluding it outright. Regression guards: 5 cases in tests/unit/grok-cli-oauth.test.ts (JWT exp, JSON expires_at, the 21600 fallback, and the two expired-token clamps). (#5775 — thanks @Chewji9875)

• compression (CCR retrieve via MCP HTTP): the omniroute_ccr_retrieve MCP tool returned "CCR block not found" for blocks stored earlier in the same session when called over the MCP HTTP transports (SSE / Streamable HTTP), e.g. from OpenCode in a Docker deployment. Compression stores each block keyed by the API-key principal (String(apiKeyInfo.id)), but the tool resolved the caller via extra.authInfo.clientId — which the MCP SDK never populates for API-key auth — so it fell back to "anonymous" and the compound store-key never matched. The retrieve tool now resolves the caller's API-key id from the MCP HTTP auth context (httpAuthContext) using the same getApiKeyMetadata lookup used at storage time, so retrieval matches storage. Cross-tenant IDOR isolation is preserved: a different key resolves to a different id → miss; no key → the anonymous bucket only. Regression guard: tests/unit/compression/ccr-mcp-principal-5649.test.ts (extraction, distinct-principal isolation, fail-closed, end-to-end store→retrieve). (#5649)

• compression (context-editing telemetry): streaming responses now record Context Editing savings. Anthropic surfaces context_management.applied_edits[] on the final message_delta snapshot of an SSE stream, but the streaming reconstruction (buildStreamSummaryFromEvents → Claude branch) dropped context_management entirely and no telemetry hook was wired into the streaming finalizer — so the delegated server-side context-clear savings (cleared_input_tokens / cleared_tool_uses) surfaced under engine context-editing in compression analytics only for non-streaming responses. The collector now preserves context_management from the final snapshot (last-writer-wins), and onStreamComplete mirrors the non-streaming recordContextEditingTelemetryHook (best-effort, Claude-only, HTTP 200 only). Purely additive telemetry — no payload mutation, no new env flag, no behavior change when the stream carries no context_management. Regression guard: tests/unit/context-editing-streaming-telemetry.test.ts (3). gaps v3.8.42 — T01 (5.1).

• proxy (relay test diagnostics): the Proxy Pool "Test" button showed a bare "failed" with nothing in the server logs when a relay (Vercel / Deno / Cloudflare) responded with a non-200 — e.g. a 401 from an auth-token mismatch after a STORAGE_ENCRYPTION_KEY rotation. The relay success-path response set success: false but carried no error field, so the dashboard had no reason to show and the server logged nothing. The test now returns an actionable error (the HTTP status, plus an auth/encryption-key hint on 401/403) and logs the failure server-side; the SOCKS5/HTTP proxy path now logs its failures too. Shaping extracted to buildRelayTestResult with a regression guard (tests/unit/proxy-relay-test-error-5716.test.ts). Note: this surfaces why a relay fails — it does not repair a genuinely broken/misconfigured relay. (#5716)

• fix(dashboard): add error boundaries for the Combos and MITM Proxy pages so a render error shows a recoverable fallback instead of a blank page. (thanks @wahyuzero)

• providers (onboarding wizard — unsupported validation): adding a provider whose credentials have no live validator (LMArena, PiAPI, …) failed silently in the Add-Provider wizard. The /api/providers/validate endpoint returns HTTP 400 + { unsupported: true } for these (#5565/#5567), but the wizard's validateOnboardingApiKey ran it through expectOk, which threw on the non-200 — so the flow jumped to the error step and the connection was never created. The wizard now treats unsupported: true as a non-blocking "can't verify" and proceeds to save, mirroring AddApiKeyModal. Regression guard added to tests/unit/provider-onboarding-wizard.test.ts. (related to #5692)

• dashboard (Quick Start step 1): the Quick Start "Create API key" step told users to "Go to Endpoint → Registered Keys" and linked to /dashboard/endpoint, but API keys are created on the API Manager page (/dashboard/api-manager, sidebar "API Keys") — the Endpoint page has no "Registered Keys" section, so users followed the link and could not find where to create a key. Step 1 now reads "Go to API Keys" and links to /dashboard/api-manager. Regression guard: tests/unit/ui/quick-start-api-keys-link-5695.test.ts. (#5695)

• providers (DashScope/Alibaba setup link): the "Get API key" link for the Alibaba and Alibaba (China) providers pointed at the bare API host (dashscope-intl.aliyuncs.com / dashscope.aliyuncs.com), which returns 404 in a browser — API hostnames have no homepage. Repointed to the consoles where keys are actually issued: bailian.console.alibabacloud.com (international) and dashscope.console.aliyun.com (China). Same class as #5572/#5574/#5576; regression guard added to tests/unit/provider-setup-links-5572.test.ts. (#5665)

• thinking / runtime-config (module-graph fix): operator-configured proxy settings that are hydrated at boot but read per-request were silently ignored in production. Next.js compiles instrumentation.ts (boot hydration via applyRuntimeSettings / restore hooks) as a separate webpack module graph from the app-route / open-sse executors, so a module-local let _config singleton is duplicated — the boot copy is hydrated but the request path reads a different, un-hydrated copy. Live VPS validation proved the Thinking-Budget hydration ran to completion at boot yet base.ts still saw the passthrough default (this is why #5312 fix A stayed broken even after the boot-wiring fix). Fixed by backing the singletons with globalThis (the pattern systemPrompt.ts already uses for the Global System Prompt, #2470), so all module-graph copies share one instance: thinkingBudget.ts (the dashboard Thinking-Budget mode now reaches the executor), backgroundTaskDetector.ts (the opt-in background-model degradation now actually fires on requests), and systemTransforms.ts (operator pipeline overrides now reach the request path). payloadRules.ts was already safe (it lazily self-loads from the DB per request, #2986). Regression guards: tests/unit/thinking-budget-globalthis-5312.test.ts + tests/unit/runtime-config-globalthis-5312.test.ts (assert globalThis-backed sharing; a module-local let fails them). (#5312)

• thinking (Claude OAuth): restore the proxy-level Thinking-Budget config on startup. The dashboard mode (auto/custom/adaptive) is persisted under settings.thinkingBudget, but the boot-time hydration (hydrateThinkingBudgetConfig) was only wired into src/server-init.ts — an unused module that never runs in production — so the operator's choice silently reverted to the passthrough default on every restart (#5312 fix A was non-functional, even though its direct unit test passed). The hydration now runs in the real boot path (src/instrumentation-node.ts), alongside the Global System Prompt restore. Surfaced by live Anthropic-OAuth validation on the VPS. Regression guard: tests/unit/thinking-budget-boot-wiring-5312.test.ts (asserts the production boot module calls the hydration, not just the function in isolation). (#5312)

• translator/chatcore (hardening): re-apply two defensive review-fixes that were dropped in a branch rebuild before #5661 / #5662 landed. (1) mergeConsecutiveSameRoleContents (OpenAI→Gemini) now shallow-copies each entry and its parts array instead of pushing the input reference, so the consecutive-same-role merge never mutates the caller's objects. (2) defaultClaudeToolType (Claude tool defaults) now passes any non-object array entry (null / primitive) through unchanged instead of spreading it into a fabricated { type: "custom", … } tool. No behavior change on real payloads (Gemini contents are freshly built; Claude tools are always objects); both properties are now locked by regression tests in tests/unit/translator-gemini-consecutive-role-2191.test.ts and tests/unit/claude-tool-type-default-2195.test.ts.

• providers (grok-cli): truncate the tool list when it exceeds a provider's hard limit, so grok-cli (cli-chat-proxy.grok.com, max 200 tools) no longer rejects requests with Maximum tools limit reached. Adds a proactive PROVIDER_TOOL_LIMITS map (grok-cli: 200, consulted before the reactive cache), a corrected limit-parsing regex that captures the stated maximum (200) instead of the supplied count (427), and removes the broken < MAX_TOOLS_LIMIT truncation gate so truncation now fires whenever tools.length exceeds the effective limit. Regression guard: tests/unit/tool-limit-detector.test.ts. (#5563 — thanks @Chewji9875)

• resilience (antigravity): record model lockout for Antigravity 429 rate_limit_exceeded errors. Antigravity's "Resource has been exhausted (e.g. check quota)." text was matched by overly broad QUOTA_PATTERNS and misclassified as QUOTA_EXHAUSTED, so the combo retry path was skipped (providerExhausted) and the model was never cooled down. Classification now prefers the structured error code — classifyErrorText(structuredError?.code || errorText) — so a rate_limit_exceeded code is treated as a transient rate-limit (not quota), and the two broad patterns (/resource.*exhaust/i, /check.*quota/i) were replaced with Antigravity-specific ones (individual quota reached, enable overages). (#5579 — thanks @Chewji9875)

• providers (OpenAI-compatible): Codex MCP / tool_search deferred discovery (and apply_patch) now works through a Custom OpenAI-compatible provider. When such a provider received a Responses-API-shaped request that carried MCP / tool_search tools, OmniRoute downgraded it to /chat/completions, which drops the deferred tool-discovery mechanism — so the MCP namespaces never surfaced to the model and apply_patch was mis-handled as a JSON tool. The executor now detects a Responses-shaped request (input / previous_response_id / max_output_tokens / reasoning) that carries namespace / tool_search* tools and routes it to the upstream /responses endpoint natively instead of downgrading (it can also be forced via providerSpecificData._omnirouteForceResponsesUpstream). This is a distinct code path from the official Codex OAuth backend (#3033 / #4539, which the earlier fix never touched). Regression guard: tests/unit/executor-default-base.test.ts. Thanks to @KooshaPari for the fix. (#5483)

• dashboard (routing): selecting the fusion strategy on the Global Routing defaults tab now reveals fusion-specific config instead of only the generic resilience fields. Fusion's engine knobs — judgeModel (the model that synthesizes the panel answers) and fusionTuning (minPanel / stragglerGraceMs / panelHardTimeoutMs) — already existed in the schema and the per-combo editor, but the Global Routing tab never surfaced them, so picking "fusion" there was effectively a no-op. The fields are now shown (extracted into a new FusionDefaultsFields component). Voting / aggregation-mode / per-provider-weight are intentionally not shown — those don't exist in the fusion engine. Regression guard: tests/unit/ui/combo-defaults-fusion-5598.test.tsx. (#5598)

• dashboard (free proxy pool): the free proxy pool "Sync All" no longer fails silently with Total: 0. Three fixes: (1) the IPLocate source fetched …/protocols/<proto>.json and parsed it as JSON, but the upstream list is plain text (<proto>.txt, one ip:port per line) — every protocol 404'd / failed to parse; it now fetches .txt and parses the line list. (2) The sync route isolates each source in its own try/catch, so one provider throwing (e.g. a TLS handshake failure) no longer aborts the whole sync — the working sources still populate the pool. (3) The UI now surfaces the per-source errors the route already returns, instead of discarding the response, so a partial/empty sync explains itself. Regression guards: tests/unit/free-proxy-providers.test.ts, tests/unit/proxy-pool-sync-4878.test.ts, tests/unit/free-pool-tab.test.tsx. (#5595)

• dashboard (memory engine): the memory engine status page no longer mixes English and Portuguese. The embedding / vector-store / rerank status detail strings were hardcoded in Portuguese in the backend (resolveEmbeddingSource, engineStatus), e.g. auto: nenhuma fonte de embedding disponível and sqlite-vec ativo, dim=…, while the surrounding UI labels render from the English i18n bundle — so an English user saw a half-translated page. The backend detail strings are now English (auto: no embedding source available, sqlite-vec active, dim=…, etc.), matching the rest of the page. Regression guard: tests/unit/memory-engine-status.test.ts. (#5596)

• providers (cline): stop falsely mapping valid Cline (OAuth) responses to 502 empty_choices + account cooldown. detectMalformedNonStream only recognized choices[].message.content as a string, but some OpenAI-compatible upstreams — Cline via OAuth among them — return content as an array of Anthropic-style text blocks inside an OpenAI envelope. A non-empty response (recvBytes > 0) was therefore classified as empty_choices and turned into a 502 that also cooled the account down. The malformed-response detector now also treats a content array carrying at least one non-empty text block as real output. Regression guard: tests/unit/diagnostics.test.ts. (#5559)

• embedded services (Windows): fix CLIProxyAPI install failing instantly with spawn unzip ENOENT on Windows. The binary extractor spawned unzip, which is not a Windows system command — it only ships inside Git for Windows' usr/bin, a directory Node's spawn PATH never sees, so even users with Git installed hit the error. On Windows the extractor now uses PowerShell's built-in Expand-Archive (via execFileAsync, no shell — paths pass as a single non-interpreted arg, with ''-escaping + -LiteralPath as defense in depth); other platforms keep using unzip. This is distinct from #5379 (that was npm.cmd needing shell: true). Regression guard: tests/unit/binary-manager-extract-zip-5590.test.ts. (#5590)

• storage (daemon): fix a Node.js out-of-memory crash on startup when storage.sqlite grows large (~170 MB+). The boot-time call-log cleanup (cleanupExpiredLogs → rotateCallLogs) ran two unbounded SELECT … FROM call_logs … .all() queries — listReferencedArtifacts (every artifact path) and deleteCallLogsBefore (every id before the retention cutoff). node:sqlite's StatementSync.all() materializes the entire result set as JS objects at once, so on a large table the V8 heap blew up and the process crashed before binding (FATAL ERROR: … heap out of memory, native frame node::sqlite::StatementSync::All). Both queries now page through call_logs in bounded 5 000-row chunks (new src/lib/usage/callLogsBoundedQueries.ts), keeping peak memory flat regardless of table size — no more manual --max-old-space-size bump required. Regression guard: tests/unit/call-log-oom-unbounded-5618.test.ts. (#5618)

• dashboard (provider setup): fix three provider setup links that pointed at 404 pages. Ollama Cloud / ollama-search linked to ollama.com/settings/api-keys → corrected to ollama.com/settings/keys (the page moved; Ollama Cloud is a real keyed service, so the field stays). SearchAPI linked to the bare searchapi.io/docs (404) → searchapi.io/docs/google. You.com linked to you.com/docs/search/overview (404) → you.com/business/api/ (the developer portal). All three replacements were verified live. Regression guard: tests/unit/provider-setup-links-5572.test.ts. (#5572, #5574, #5576)

• providers (AI/ML API): the model-import step now loads the live AI/ML API catalog (400+ models) instead of falling back to a stale 6-model seed. The registry had no modelsUrl, so the route silently used the bundled catalog with an "API unavailable — using local catalog" warning even when the key was valid. AI/ML API exposes its full catalog at the public, auth-free https://api.aimlapi.com/models endpoint (a bare array of { id, type, info }, distinct from the OpenAI-compat /v1/models); it's now wired into the models route's discovery config, with the bundled catalog kept as the offline fallback. Regression guard: tests/unit/provider-models-route.test.ts. (#5570)

• providers (CablyAI): mark CablyAI deprecated — cablyai.com no longer resolves (DNS NXDOMAIN, verified 2026-06-30); the domain is gone. The provider is removed from the models-route discovery config so the import step returns a clean error instead of an unhandled 500 crash (the dead-domain fetch threw with no local-catalog fallback), and the registry entry now carries deprecated: true / riskNoticeVariant: "deprecated" so the dashboard flags existing connections (same treatment as the shut-down glhf/kluster.ai gateways). Regression guard: tests/unit/provider-models-route.test.ts. (#5568)

• dashboard (provider add): non-LLM search/agent providers no longer fail the model-import step with a red Provider <id> does not support models listing. Jules (Google Labs coding agent), linkup-search (Linkup web search), ollama-search (Ollama Cloud web search — distinct from the local Ollama LLM), and searchapi-search (SearchAPI SERP) have no /v1/models endpoint, so the import surfaced a failure for expected behavior. Each now ships a small static catalog of its selectable capability ids — Linkup's fast/standard/deep search depths, SearchAPI's google/bing/youtube/… engines, a single Jules/Ollama-web-search entry — so the import step returns a usable list (source: local_catalog) instead of an error. Regression guard: tests/unit/provider-models-route.test.ts. (#5569, #5571, #5573, #5575)

• dashboard (provider add): providers without a live key/cookie validator (e.g. LMArena (Free), PiAPI) can now be saved. The Add-connection modal treated the backend's "Provider validation not supported" response as a hard Invalid state and blocked Save entirely, leaving those providers impossible to add. The validate route now returns unsupported: true alongside the message, and the modal treats that as a non-blocking warning — the "Check" badge still shows "validation not supported" (informational), but Save persists the credential as-is. Regression guards: tests/unit/ui/add-api-key-modal-unsupported-save-5565.test.tsx (Save proceeds) and tests/unit/providers-validate-route.test.ts (wire-format). (#5565, #5567)

• providers (codex): fix the Codex Responses WebSocket path (/v1/responses), which regressed in v3.8.40 with a client-visible Invalid JSON body and bypassed the configured proxy. (1) #5591 — PR #5237 bumped the impersonation TLS profile to chrome_149, but wreq-js@2.3.1 only supports up to chrome_147; the unknown profile produced a degenerate fingerprint and ChatGPT rejected the upstream upgrade. The Codex WS path is reverted to the proven chrome_142 (the v3.8.39 value), and the over-bumped grok-web/claude-web profiles (masked by their circuit-breaker but silently dropping TLS impersonation) are restored to chrome_146. A new regression guard asserts every configured chrome_* profile exists in the installed wreq-js typings (tests/unit/tls-profiles-valid-5591.test.mjs). (2) #5611 — the upstream wreq-js.websocket() connect ignored the Proxy Registry, so a no-direct-egress Docker container failed with a DNS error; the prepare route now resolves the Global/provider proxy and threads it through to the WS connect. Regression guard in tests/unit/responses-ws-proxy.test.mjs. (#5591, #5611)

• providers (GLM): GLM 5.1 / 5.2 now keep the system role instead of having the system prompt folded into the first user turn. roleNormalizer.ts matched every glm* id with a blanket startsWith("glm") / startsWith("glm-") prefix, so the next-generation models — which z.ai documents as supporting the system role (GLM > 5.0) — were normalized as if they rejected it, degrading instruction-following. The matcher is now version-aware: it strips the system role only for bare glm, the 4.x family, and the 5.0 generation, and preserves it for glm-5.1/glm-5.2 (and the Fireworks glm-5p1 point alias). The ZenMux vendor-prefixed z-ai/glm-* compressed-history rule and the ERNIE rule are unchanged. Regression guards in tests/unit/role-normalizer.test.ts. (#5610)

• Security hardening follow-ups (v3.8.15): the auth_token cookie now sets an explicit 30-day maxAge so sessions persist as intended (Seg3); the management bootstrap warns at boot when INITIAL_PASSWORD is left at the insecure CHANGEME default (Seg2); VS Code path-token endpoints (/api/v1/vscode/raw/[token]) emit a once-per-process security warning since the API key travels in the URL and can leak via logs/proxies (Seg4); the system version route resolves the real global install path via npm root -g instead of a hardcoded /app (Bug3); and auto-update mode detection segment-matches node_modules instead of substring-matching, eliminating false "global install" positives (Bug1).

• fix(cli): rename the Node process title to omniroute so it shows correctly in ps/htop. (thanks @waguriagentic)

• dashboard (model picker): guard against null model-alias values so opening Create Combo for a custom provider node no longer crashes. ModelSelectModal's custom-provider branch filtered modelAliases entries with a raw fullModel.startsWith(...), which threw a TypeError whenever an alias value was null/undefined (a stale/partial entry persisted to settings). The filter/map logic is extracted into a new buildNodeAliasModels helper (mirroring the sibling passthrough-alias guard, #485) that requires typeof fullModel === "string" before calling .startsWith. Regression guard: tests/unit/model-select-null-alias-guard-2247.test.ts. (thanks @wahyuzero)

• fix(translator): strip orphaned tool results (results with no matching tool call) across request formats to avoid upstream 400s. (thanks @warelik)

• fix(kiro): stop injecting a placeholder user turn on trailing tool-result turns so agentic loops aren't disrupted. (thanks @jetmiky)

• fix(translator): prevent doubled tool arguments in OpenAI-to-Claude responses (duplicate finish_reason guard + string tool-input passthrough). (thanks @vishalrajv)

• codex (agent goal streams): protect long-running agent goal streams so extended agent runs are no longer cut off prematurely. (#5772 — thanks @nguyenxvotanminh3)

• sse (zero-width markers): strip zero-width markers from streamed responses, matching the non-streaming path so streamed output is byte-clean parity. (#5857 — thanks @DKotsyuba)

• usage (om-usage endpoint): restore the om-usage HTTP endpoint. (#5859 — thanks @Witroch4)

• sse (stream readiness): tune adaptive stream-readiness timeouts so slow-first-token upstreams are handled more reliably. (#5767 — thanks @nguyenxvotanminh3)

• security (provider node URL): harden provider node URL validation. (#5760 — thanks @nguyenxvotanminh3)

• cli (Windows doctor): correct rootDir resolution in doctor.mjs on Windows. (#5845 — thanks @arssnndr)

• providers (Antigravity): fix a 429 hang on credit exhaustion and apply a precise reset-time model lockout instead of stalling — cleaned re-implementation of #5823. (#5846 — thanks @Chewji9875 / @diegosouzapw)

• providers (qwen-web): unblock the validator and chat completion — the retired endpoint is replaced and the missing SPA version header is now sent. (#5855 — thanks @janeza2)

• providers (kimi-web): migrate to the www.kimi.com Connect-RPC API after kimi.moonshot.cn was retired. (#5858 — thanks @janeza2)

• dashboard (CSRF): unify the dashboard CSRF origin fallback so dynamic/public origins validate correctly. (#5856 — thanks @rdself)

• db (health check interval): preserve healthCheckInterval=0 across connection create/update instead of coercing it to a default. (#5822 — thanks @atomlong)

• sse (claude→codex streaming): stop the reasoning-summary drop and duplicated deltas on claude→codex streaming — reasoning snapshots are now synthesized in TRANSLATE mode and the sequence-number watermark is tracked per-stream (#5786). (#5832 — thanks @diegosouzapw)

• deps (runtime): add the missing runtime dependencies @toon-format/toon and safe-regex so the published package resolves them at runtime. (#5771 — thanks @chirag127)

• system (Windows auto-update): route in-app auto-update npm calls through the win32 shell helper so updates run correctly on Windows (#5542). (#5797 — thanks @diegosouzapw)

• dashboard (validation badge): show a neutral badge for unsupported validation and make OAuth error messages clickable links (#5442, #5486). (#5795 — thanks @diegosouzapw)

• providers (metadata): correct stale/broken provider metadata (#5487, #5461, #5534, #5470). (#5790 — thanks @diegosouzapw)

• providers (local-catalog imports): import intentional local-catalog-only providers instead of surfacing a 502 (#5460, #5465). (#5787 — thanks @diegosouzapw)

• proxyfetch (failover): skip the failover retry for non-replayable request bodies so a consumed stream isn't re-sent empty. (#5770 — thanks @Ardem2025)

• batch (recovery): persist batch item checkpoints during recovery so an interrupted batch resumes from where it left off. (#5753 — thanks @ag-linden)

• memory (Qdrant): enabling Qdrant now activates it as the retrieval engine (the auto default never selected it) and adds inline guidance (#5597). (#5741 — thanks @diegosouzapw)

• chat (non-streaming aggregation): harden non-streaming SSE aggregation against malformed upstream event sequences. (#5746 — thanks @rdself)

• sse (cooldown parsing): the anti-thundering-herd guard now tolerates numeric-epoch cooldown values. (#5747 — thanks @diegosouzapw)

• api (body size): raise the LLM API payload limit for the responses routes so larger requests aren't rejected. (#5652 — thanks @JxnLexn)

• providers (HuggingChat): fix HuggingChat web-session routing (#5592). (#5592 — thanks @backryun)

• sse (heap pressure): bound the chat hot-path heap — pressure-aware admission, response cap, and clone reductions — to avoid OOM under load (#5152). (#5425 — thanks @josevictorferreira)

• providers (M365 Copilot): validate M365 Copilot web credentials. (#5432 — thanks @skyzea1)

• providers (chatgpt-web): restore the dot-form Pro model ids. (#5549 — thanks @Thinkscape)

• security (error stacks): avoid rendering error stacks in responses. (#5624 — thanks @KooshaPari)

• security (linkify): restrict linkifyText hrefs to an explicit http(s) scheme allowlist. (#948d2d7 — thanks @diegosouzapw)

• translator (doubled tool args): prevent doubled tool-call arguments in the OpenAI→Claude translation path. (#5828 — thanks @diegosouzapw)

• translator (orphaned tool results): strip orphaned tool-result turns across request formats so an upstream doesn't reject a tool result with no matching call. (#5805 — thanks @diegosouzapw)

• translator (Gemini/Claude hardening): re-apply lost defensive hardening for the Gemini merge path and Claude tool defaults. (#5706 — thanks @diegosouzapw)

• kiro (tool-result turns): stop injecting a placeholder user turn on tool-result turns, which corrupted otherwise-valid Kiro conversations. (#5807 — thanks @diegosouzapw)

• providers (Kiro catalog): add claude-sonnet-5 to the Kiro model catalog. (#5796 — thanks @diegosouzapw)

• oauth (connection disambiguation): disambiguate OAuth connections on username so two different identity providers no longer overwrite each other. (#5803 — thanks @diegosouzapw)

• github (Copilot prefill): drop the trailing assistant prefill for Copilot chat, which some Copilot models rejected. (#5802 — thanks @diegosouzapw)

• mitm (hosts cleanup): clean up privileged /etc/hosts entries on exit when possible so a crashed/interrupted run doesn't leave stale redirects behind. (#5808 — thanks @diegosouzapw)

• dashboard (model picker): guard null modelAliases values in the model picker so a connection with no aliases no longer throws. (#5792 — thanks @diegosouzapw)

• dashboard (error boundaries): add error boundaries for the Combos and MITM Proxy pages so a render error no longer blanks the whole dashboard. (#5788 — thanks @diegosouzapw)

• cli (process title): rename the running process title to omniroute. (#5791 — thanks @diegosouzapw)

• compression (context-editing telemetry): record Context Editing telemetry on the streaming path, not just the non-streaming path. (#5761 — thanks @diegosouzapw)

• security (v3.8.15 hardening follow-ups): land the Seg2/Seg3/Seg4/Bug3 hardening follow-ups from the v3.8.15 security review. (#5512 — thanks @diegosouzapw)

📝 Maintenance

• docs (architecture): add docs/architecture/ROUTER_BACKENDS.md — an ADR pinning down how the routing engines (ts native, bifrost, cliproxy, 9router, VibeProxy-compatible) relate to each other along two orthogonal axes (lifecycle: in-process / supervised / external vs. relay selection backend), answering the architecture questions raised in #5603 (backend interface model, why CLIProxy spawns a process, feature-flag swapping, actionable route-contract errors). The typed router-backend registry the ADR describes lands separately via #5868. (#5891)

• tests (autoCombo): stabilize the getTaskFitnessWithSource identifies fitness_table as source for known models unit test, which flaked whenever the models.dev capabilities DB was populated in CI: the fixture model gpt-4o is a real models.dev catalog id, so the fitness resolution chain returned models_dev_tier instead of the expected static fitness_table source. The fixture now uses claude-sonnet (a shortened alias absent from the models.dev catalog, matching the sibling resolution-chain test), which deterministically falls through to the static table — the exact source and score assertions are preserved (0.95 = FITNESS_TABLE.coding["claude-sonnet"]). (#5890) — thanks @KooshaPari

• oauth (dead-code removal): delete the superseded legacy OAuth service-class hierarchy under src/lib/oauth/services/. The live OAuth flow runs through src/lib/oauth/providers.ts + src/lib/oauth/providers/ (wired into the generic oauth/[provider]/[action] route); the old per-provider class *Service extends OAuthService implementations plus their barrel had zero production or test references. Removed oauth.ts (base class), openai.ts, github.ts, claude.ts, codex.ts, antigravity.ts, qwen.ts, qoder.ts, and the index.ts barrel (−1559 LOC). Kept the three still-live files that routes import directly by path: kiro.ts (Kiro import/exchange routes), cursor.ts (Cursor import route), and codexImport.ts (utility fns for the Codex bulk-import route). Proven safe by typecheck:core staying green (any live reference would fail the build) + a filesystem guard tests/unit/oauth-legacy-services-removed.test.ts pinning the removal against re-introduction. Salvage of the closed PR #5039. gaps v3.8.42 — T10 (5.7).

• refactor (god-file decomposition): extracted pure leaf modules across db, sse, usage, api, memory, evals, models, resilience, and dashboard god-files (types/mappers/helpers/pure-transform leaves; behavior-preserving, test-guarded): db/providers, db/proxies, db/models, db/settings, usageAnalytics, migrationRunner (#5714, #5717, #5705, #5709, #5722, #5721); sse openai-to-gemini / cursor-protobuf / rate-limit-headers / reasoning-tag (#5824, #5794, #5736, #5734); usage families / callLogs / usageHistory / providerLimits (#5782, #5725, #5728, #5730); api provider-models discovery / unified-catalog (#5758, #5699); memory retrieval scoring (#5733); evals golden-set suites (#5740); modelsDevSync transform layer (#5743); resilience settings split (#5745); dashboard sidebarVisibility split (#5683); executor shared-utility dedup + tests (#5720 — thanks @pizzav-xyz). — thanks @diegosouzapw

• chore (Bun script runner): adopt Bun 1.3.10 as a locked, allow-listed build/dev script runner for a small set of validated TS gate/generator scripts (Node stays the published runtime): locked runtime dependency, CI script-checks + validated-scripts run under Bun, and a bun-safe pack validator. (#5615, #5617, #5612, #5643 — thanks @KooshaPari; docs #5703 — thanks @diegosouzapw)

• docs (sync & housekeeping): i18n CHANGELOG mirror sync for the [3.8.43] section (#5789); MCP tool count synced to 95 + routing-strategy count (#5732); README faster/leaner install notes, refreshed metrics/badges, 17-strategy + Quota-Share listing, provider counts, and grammar fixes (#5713, #5738 — thanks @chirag127); security docs for banned-keyword/account-ban detection (#5756) and the full LOCAL_ONLY route set + GHSA advisory + audit path (#5748); relay backend-routing contract clarification (#5621 — thanks @KooshaPari); release-freeze scoped to /generate-release only (#5839); .editorconfig repository standards (#5879 — thanks @shiva24082). — thanks @diegosouzapw

• test/ci (stabilization & ratchets): guard the tsx/esm→esbuild boot transform (#5773); align t3-web web-session metadata (#5835); repoint the sidebar quota-share placement scan (#5711); lightweight health probe for batch e2e (#5651 — thanks @KooshaPari); make release-green pre-flight gates visible + bounded (#5644); stabilize nightly-mutation (tap.testFiles drift guard + anti-flake eps) (#5682); close the QG v2 tail (#5681); normalize check route paths on Windows (#5613 — thanks @KooshaPari); pass sonar.projectVersion to the SonarQube scan (#5880); plus stryker tap.testFiles registration, compression-studio smoke re-anchoring, rtk_discover de-flake, and v3.8.43-cycle ratchet rebaselines (deadExports 225→227, complexity 1981→1982, cognitive-complexity 842→845, eslintWarnings 4121→4158→4199). — thanks @diegosouzapw

• refactor (oauth): remove dead legacy OAuth service classes. (#5838 — thanks @diegosouzapw)

🙌 Contributors

Thanks to everyone whose work landed in v3.8.43:

Contributor	PRs / Issues
@ag-linden	#5753
@Ardem2025	#5770
@arssnndr	#5845
@atomlong	#5822
@backryun	#5592
@baslr	direct commit / report
@Chewji9875	#5563, #5579, #5846
@chirag127	#5738, #5771
@DKotsyuba	#5857
@hartmark	#5834
@ishatiwari21	#5799
@janeza2	#5855, #5858
@jetmiky	direct commit / report
@josevictorferreira	#5425
@JxnLexn	#5652
@KooshaPari	#5613, #5621, #5624, #5629, #5643, #5651, #5890
@KunN-21	direct commit / report
@manhdzzz	direct commit / report
@nguyenxvotanminh3	#5760, #5767, #5772
@noir017	direct commit / report
@pizzav-xyz	#5720
@rdself	#5746, #5856
@shiva24082	#5879
@skyzea1	#5432, #5701
@Stazyu	#5557
@Thinkscape	#5549
@vishalrajv	direct commit / report
@voravitl	direct commit / report
@waguriagentic	direct commit / report
@wahyuzero	direct commit / report
@warelik	direct commit / report
@Witroch4	#5731, #5859, #5863
@diegosouzapw	maintainer — cycle reconciliation, release-close base-red fixes, god-file decomposition, compression/memory features

---

[3.8.42] — 2026-06-30

✨ New Features

• compression (pipeline): add an honest default-on inflation guard to the stacked compression pipeline (T02 / Headroom H1). If the fully-stacked engines produce a body that did not actually shrink — its token count is >= the original — the compressed body is discarded and the verbatim original request is sent upstream instead, with a pipeline-inflation-guard warning recorded in the compression stats. This is safe by construction (the only fallback is the unmodified original, always a valid payload) and complements the existing opt-in per-step TV1 bail-out, which governs step-to-step advancement rather than the final output. New open-sse/services/compression/pipelineGuards.ts; wired at the single finalizeStackedResult choke point shared by the sync and async stacked paths. Regression guards (incl. an inflating-engine integration test) in tests/unit/compression-pipeline-inflation-guard.test.ts.

• compression (caveman): complete the German, French, and Japanese rule packs with the dedup (repeated-context collapsing) and ultra (abbreviation / terse) categories they were missing — these three languages previously shipped only context/filler/structural, while en/es/id/pt-BR had all five. So a de/fr/ja conversation compressed at higher intensities now collapses repeated boilerplate ("wie bereits besprochen" → "Siehe oben.", "comme mentionné précédemment" → "Voir ci-dessus.", "前述のとおり" → "（上記参照）") and abbreviates dense technical vocabulary (Datenbank→DB, Authentifizierung→Auth; base de données→BD, authentification→auth; データベース→DB, アプリケーション→app). Patterns mirror the existing es pack and stay ReDoS-safe (bounded literal alternations; the CJK pack uses no \b since Japanese has no word boundaries). Regression guard: tests/unit/caveman-packs-de-fr-ja.test.ts (packs load + validate + shrink a representative sample). gaps v3.8.42 — T05/C2.

• compression (caveman): add a Chinese (zh / wenyan 文言) input-side rule pack — the counterpart of the existing output-side terse-cjk style. New rules/zh/{dedup,filler,ultra}.json collapse repeated context ("如前所述" → "见上。"), drop pleasantries/hedging ("请帮我…/谢谢/我觉得"), strip sentence-final modal particles ("吗/呢/吧"), and abbreviate dense technical terms ("数据库"→"DB", "应用程序"→"app"). Chinese is now auto-detected: detectCompressionLanguage distinguishes zh from ja by Han-without-kana (kana is Japanese-exclusive, so a Han-heavy Japanese sentence still resolves to ja), and zh is listed in listSupportedCompressionLanguages. Patterns are ReDoS-safe (bounded literal alternations, no \b since CJK has no word boundaries). Regression guard: tests/unit/caveman-packs-zh-wenyan.test.ts (packs load + validate + shrink; zh/ja/non-CJK detection). gaps v3.8.42 — T05/C6.

• compression (RTK): add Gradle and .NET CLI (dotnet) to the RTK tool-output filter catalog. Tool output for gradle/gradlew and dotnet build|test|restore|publish is now recognized (both by command and by output content) and compressed: Gradle daemon/welcome banners and no-op > Task … UP-TO-DATE/SKIPPED/FROM-CACHE lines are dropped while BUILD SUCCESSFUL/FAILED, "What went wrong", and stack traces are preserved; the .NET build banner, copyright, and Determining projects to restore/Restored … chatter are dropped while Build succeeded/FAILED, error CS####/warning CS####, and test summaries are preserved. New builtin filters engines/rtk/filters/{gradle,dotnet}.json (with inline tests run by the catalog gate) plus gradle/dotnet entries in the command detector. Regression guard: tests/unit/rtk-gradle-dotnet-filters.test.ts. gaps v3.8.42 — T07/R9.

🔧 Bug Fixes

• providers (chatgpt-web): fix 502 ChatGPT sentinel failed: Digest method not supported on the Electron desktop app, which made every chatgpt-web/* request fail. The sentinel proof-of-work hashed with native createHash("sha3-512"), but Electron's Node is built against BoringSSL, which does not implement the SHA-3 family (electron/electron#30530), so the digest threw at construction — the provider was unusable on the desktop build (works under plain Node/OpenSSL). The PoW now hashes through a new runtime-portable helper (open-sse/utils/sha3-512.ts) that prefers the native digest and transparently falls back to a dependency-free pure-JS Keccak-f[1600] when native SHA-3 is absent. The fallback is validated bit-for-bit against native createHash("sha3-512") (300 random inputs) and the published FIPS-202 known-answer vectors. Regression guards in tests/unit/chatgpt-web-sha3-boringssl-5531.test.ts. (#5531)

• providers (bytez): fix Bytez key validation ("Provider validation endpoint not supported") and the chat base URL, verified live with a real key. Bytez is OpenAI-compatible at …/models/v2/openai/v1, but the registry stored the bare …/models/v2 base, so the validation chat-probe hit …/models/v2/chat/completions → 404 → the misleading "endpoint not supported". Two parts: (1) the registry baseUrl now carries the full OpenAI-compat chat path (…/models/v2/openai/v1/chat/completions); (2) key validation no longer uses a chat probe — a Bytez account only serves models explicitly added to its catalog, so even valid keys 404 on any model id. A dedicated validateBytezProvider instead probes the auth-only GET …/models/v2/list/tasks endpoint (200 ⇒ valid, 401/403 ⇒ invalid), which is independent of catalog provisioning. Regression guard: tests/unit/bytez-validation-5422.test.ts. (#5422)

• dashboard (provider add): two provider-add UX fixes. (1) #5420 — the "Import Models" button now stays hidden for tool-only providers (web search / web fetch), not just *-search ids: firecrawl and jina-reader (declared serviceKinds: ["webFetch"]) previously showed an Import button that hit the 400 "does not support models listing" route. A new capability check (providerLacksModelListing over the resolved serviceKinds) gates the section without ever hiding an LLM/media provider. (2) #5426 — Coze key validation no longer leaks the raw upstream envelope ({code,msg,logId,from}) into the UI; the Coze-shaped error becomes a friendly Coze rejected the key: <msg> (code <n>) message (scoped to provider === "coze" so no other provider is affected). Regression guards: tests/unit/model-listing-capability-5420.test.ts, tests/unit/coze-validation-error-5426.test.ts. (#5420, #5426)

• providers (friendliai, novita): fix two provider registry endpoints that rejected valid keys (verified live with real keys). FriendliAI pointed at …/dedicated/v1/chat/completions, which 403 Forbiddens a serverless flp_* token — switched to …/serverless/v1/chat/completions (+ a serverless modelsUrl). Novita pointed at the legacy …/v3/… base with a typo'd model id ai-ai/llama-3.1-8b-instruct (both 404) — switched to the OpenAI-compatible …/openai/v1/… base + the valid meta-llama/llama-3.1-8b-instruct id. Regression guard: tests/unit/provider-endpoints-friendliai-novita.test.ts. (#5430, #5455)

• providers (muse-spark): align the Muse Spark Web (Meta AI) cookie copy with the live cookie name. The default session cookie migrated from the retired abra_sess to ecto_1_sess (META_AI_DEFAULT_COOKIE), but the provider form hint and one 401 auth-failure message still told users to paste abra_sess — a cookie that no longer exists. Both strings now name ecto_1_sess. Regression guard: tests/unit/muse-spark-cookie-copy-5449.test.ts. (#5449)

• dashboard (provider add): fix three rough edges in the Add-API-Key / model-import flow reported across the provider-catalog audit. (1) The Validation Model and Account ID form fields shipped untranslated i18n stub copy ("Validation Model Id Label", "Account Id Placeholder", …) that surfaced verbatim in the modal — replaced with real labels/placeholders/hints in en.json. (2) Model import silently fell back to the cached/local catalog: the route already returned a warning ("API unavailable — using local catalog"), but useModelImportHandlers only read models/error and dropped it, so the user got local models with no indication — the warning is now surfaced as an import log line (new pure helper extractImportWarning). (3) The required connection-name field defaulted to "", which let browser autofill inject garbage (e.g. wiw) — it now defaults to "main". Regression guard: tests/unit/provider-add-ux-i18n-import-warning.test.ts. (#5421, #5428, #5429, #5431, #5435)

• services (installer): fix spawn EINVAL when installing an embedded service (9Router / CLIProxy) on Windows + Node.js 24+. Node 24 stopped letting child_process.execFile() run .cmd batch files without a shell (nodejs/node#52554), and npm on Windows is npm.cmd, so runNpm() threw EINVAL the moment a user clicked Install. runNpm now enables shell on win32 only. To keep Hard Rule #13 intact under a shell — where the shell, not execFile, parses argv — the install --prefix (a DATA_DIR path that can legitimately contain spaces, e.g. C:\Users\John Doe\.omniroute\…) is now passed via the npm_config_prefix environment variable instead of an argv path, and the user-supplied install version is constrained to a dist-tag/semver shape (SERVICE_VERSION_PATTERN) at the route boundary so it can never carry shell metacharacters. With the prefix in the environment and the version validated, every remaining argv entry is a static flag. Regression guards: tests/unit/services/installers/runNpm-shell-5379.test.ts (+ existing ninerouter.test.ts aligned to npm's npm_config_prefix env). (#5379)

• cli (serve): restore dist/tls-options.mjs to the npm tarball — the opt-in native HTTPS/TLS sidecar (#5361) was copied into the staged dist/ by the build but then pruned by the prepublish allowlist step, so omniroute serve crashed on the published 3.8.41 with ERR_MODULE_NOT_FOUND (dist/server-ws.mjs imports ./tls-options.mjs). Added tls-options.mjs to APP_STAGING_ALLOWED_EXACT_PATHS (survives the prune) and dist/tls-options.mjs to PACK_ARTIFACT_REQUIRED_PATHS (the check:pack-artifact gate now fails loudly if it ever vanishes again — same guard pattern as webdav-handler.mjs). Regression guards in tests/unit/pack-artifact-policy.test.ts. (#5452 — thanks @KooshaPari for the parallel fix #5494)

• dashboard: fix the Add Provider / onboarding wizard button silently doing nothing. The /dashboard/providers/new route was a redirect stub (it bounced straight back to /dashboard/providers), so every "Add Provider" button and dashboard widget link opened nothing, and the fully-built ProviderOnboardingWizard component stayed orphaned (never rendered by any route). The route now renders the wizard directly; auth is enforced centrally by the (dashboard) layout, same as the sibling provider routes. Regression guard in tests/unit/onboarding-wizard-route-5427.test.ts. (#5427)

• db (import): fix EBUSY: resource busy or locked when importing a database on Windows. The import route deleted the live storage.sqlite + WAL/-shm/-journal sidecars with a plain fs.unlinkSync immediately after resetDbInstance(), but Windows releases the SQLite file handle asynchronously after close() (mmap / antivirus), so the unlink raced and threw EBUSY. The route now deletes via unlinkFileWithRetry (EBUSY/EPERM backoff) — the same helper the restore path already uses. Regression guard in tests/unit/db-import-ebusy-5406.test.ts. (#5406, consolidated under #5161)

• build: keep ioredis out of the client/CLI bundle — a dast-smoke regression revealed the module was being pulled into browser/Electron client-side chunks; adding it to the SPAWN_CAPABLE_PREFIXES leaf excludes it from client bundles while keeping it available on the server path. (#5546)

• providers (mimocode): route per-account traffic through SOCKS5 proxy dispatchers — each mimocode account's requests are now dispatched via its configured SOCKS5 proxy rather than the default direct connection. (#5521 — thanks @pizzav-xyz)

• providers: persist the Configured provider filter selection across page reloads — the filter was resetting to "All" on every navigation. (#5510 — thanks @KooshaPari)

• providers (chatgpt-web): support GPT-5.5 Pro model handoff — adds the model mapping and handoff routing needed for the GPT-5.5 Pro tier. (#5536 — thanks @Thinkscape)

• dashboard: keep onboarding schemas browser-safe — the schema module imported a server-side db reference that crashed the browser bundle; it is now imported only on the server path. (#5525 — thanks @KooshaPari)

• routing (bifrost): add auto-fallback cooldown for bifrost targets — prevents rapid re-selection of a failing bifrost backend within the cooldown window, complementing the existing circuit-breaker mechanism. (#5519 — thanks @KooshaPari)

• providers (opencode-plugin): bump the opencode plugin to v0.2.0 and wire auto-publish on release so the plugin package tracks OmniRoute releases automatically. (#5363 — thanks @herjarsa)

• rate-limit: normalize queue refresh settings — aligns the queue-refresh interval configuration across rate-limit strategies so stale queues are released on a consistent schedule. (#5499 — thanks @KooshaPari)

• fallback: normalize provider error-rule header extraction — ensures fallback retry decisions correctly read all response headers regardless of casing, fixing cases where a provider's Retry-After or custom error header was silently dropped. (#5473 — thanks @KooshaPari)

• routing: gate Claude adaptive-thinking defaults behind the feature flag — prevents the thinking budget from being injected into requests for models that do not support the extended-thinking parameter, avoiding upstream 400 errors on non-thinking Claude variants. (#5480 — thanks @KooshaPari)

• ci: fix post-merge CI regressions introduced by the dead-code sweep — restores test imports and type references broken when the ratchet landed before downstream consumers were updated. (#5467 — thanks @KooshaPari)

• sse: treat terminal stream cancels as complete — an aborted SSE stream was being left in a partial state, causing downstream consumers to wait indefinitely for a final event that would never arrive. (#5491 — thanks @JxnLexn)

• api: fix framing of non-streaming JSON responses — stream: false chat-completions responses were returned without correct content-length framing, causing some clients to misparse the response body. (#5416 — thanks @rdself)

• dashboard (tests): protect dynamic dashboard endpoint tests with CSRF validation — the test suite was exercising dashboard API routes without CSRF tokens, masking a coverage gap for those endpoints. (#5405 — thanks @rdself)

• providers: remove the dead Phind provider (service shut down) and deduplicate the HuggingChat catalog listing that had accumulated a stale duplicate entry. (#5530 — thanks @backryun)

• providers (longcat): correct the LongCat free tier — LongCat-2.0 is now GA; the one-time 10M-token promo (KYC required) is correctly reflected in the catalog, replacing the stale legacy beta entry. (#5508 — thanks @backryun)

📝 Maintenance

• dashboard (refactor): consolidate the duplicate caveman on/off toggle from the compression settings tab onto the single-source panel (T11), eliminating the stale off-sync copy. (#5524)

• tests: add quota guard for Claude-Code identity version lockstep (Phase 2) — asserts that the Claude-Code version reported in quota accounting stays in sync with the deployed version, preventing silent drift. (#5514)

• docs: add relay backend strategy guide documenting supported relay backend types, selection criteria, and configuration patterns. (#5547)

• docs: clarify bifrost relay backend environment variables — documents which env vars control bifrost's relay backend selection and failover behavior. (#5520 — thanks @KooshaPari)

• tests: add relay routing fallback header behavior tests — regression guard asserting that fallback-triggered relay requests carry the correct forwarded headers through the routing layer. (#5526 — thanks @KooshaPari)

• ci: add npm fetch-retry configuration and codify the release-freeze protocol (Hard Rule #21) — reduces transient npm registry fetch failures in CI and establishes the documented procedure for freezing releases. (#5506)

• deps: bump 11 production dependencies to their latest compatible versions. (#5414)

• deps: bump Electron from 42.4.1 to 42.5.1 in /electron. (#5413)

• deps: bump the development dependency group with 9 updates. (#5415)

• maintenance (dead-code): repo-wide sweep of unused exported symbols, types, and schemas — removes 35 no-longer-referenced exports across cloud-agent, a2a, SSE, memory, quota, skills, gamification, codex, qdrant, playground, provider catalog, and combo modules, reducing the exported API surface and eliminating stale misleading types. (#5372, #5373, #5374, #5375, #5376, #5377, #5378, #5380, #5381, #5382, #5383, #5384, #5385, #5386, #5387, #5388, #5389, #5390, #5391, #5392, #5393, #5395, #5396, #5397, #5398, #5399, #5400, #5401, #5402, #5403, #5404, #5463, #5464, #5466, #5468 — thanks @JxnLexn)

• maintenance (DRY): DRY consolidation of shared helpers — extracts 17 duplicated utilities into single shared modules: vscode metadata helpers, proxy route handlers, auth zip extractors, combo-builder model options, vscode tokenized-request helpers, quota strategy ranking helpers, recharts donut card, provider-specific validation, batch response formatter, Redis runtime helpers, version-manager request parsing, media-generation route helpers, service install helpers, settings transform schemas, relay stream finalizer, machine-id fallback, and node SQLite adapter. (#5471, #5472, #5475, #5477, #5479, #5482, #5484, #5485, #5488, #5490, #5492, #5493, #5495, #5496, #5497, #5498, #5500 — thanks @JxnLexn)

---

[3.8.41] — 2026-06-29

✨ New Features

• feat(relay): selectable relay backend (TS / Bifrost / auto) — the OpenAI-compatible relay endpoint can now route its hot path through a native Bifrost sidecar without clients changing URLs. OMNIROUTE_RELAY_BACKEND / RELAY_ROUTING_BACKEND = ts | bifrost | auto: defaults to the existing TypeScript relay; auto selects Bifrost when BIFROST_BASE_URL is set (and BIFROST_ENABLED ≠ 0) and falls back to TS automatically if the sidecar is unreachable; bifrost keeps strict failure behavior. Auth, per-IP/token rate limits, prompt-injection checks, and model allowlists still run in the Next relay route before dispatch (control plane stays in the app); responses carry X-Routing-Backend / X-Routing-Fallback. Regression guards: tests/unit/api/v1/relay-routing-backend.test.ts, tests/unit/api/v1/bifrost-sidecar.test.ts. (#5315, #5316 — thanks @KooshaPari)

🔧 Bug Fixes

• translator (claude): synthesize a minimal user turn when an OpenAI→Claude request carries only system/developer messages, so the request stops failing with [400]: messages: at least one message is required. openaiToClaudeRequest hoists every system/developer turn into Claude's top-level system field and filters them out of messages; an all-system input (OpenCode compaction / title-generation requests) left messages: [], which the Messages API rejects — surfacing in OpenCode as a mid-task stream error that drops the conversation. The guard fires only when messages would otherwise be empty (system instructions still drive the response), so non-empty requests are unaffected. (#5342 — thanks @wild-feather)
• providers (gemini): drop retired Google AI Studio model ids and align the catalog to what the live GenAI API actually serves (verified 2026-06-29 against the official deprecations page). Removes long-retired gemini-1.5-pro/gemini-1.5-flash, the shut-down gemini-2.0-flash/gemini-2.0-flash-lite, and dead experimentals; renames gemini-3.1-flash-lite-preview → the GA gemini-3.1-flash-lite; swaps the retired text-embedding-004 for the live gemini-embedding-001/gemini-embedding-2; and adds graceful modelDeprecation forwards so legacy/renamed ids redirect to the GA model instead of 404ing. Native AI-Studio-direct image/video/music registration is intentionally out of scope (needs real executor work; those models stay reachable via Antigravity/Vertex/aggregators). (#5337 — thanks @backryun)
• services (dashboard): fix the embedded-services dashboard failures (#5298) — service supervisors are now lazily initialized from /api/services/[name]/logs so cliproxy/9router logs no longer 404 before bootstrap registers a supervisor; lifecycle buttons send JSON (empty install bodies default to version: "latest", malformed JSON still returns 400 Invalid JSON body); lifecycle and log-stream failures surface as actionable UI errors instead of silently showing no logs; Tailscale CGNAT 100.64.0.0/10 peers count as private-LAN local for local-only service access; a parent /dashboard/context → /dashboard/context/settings redirect stops RSC prefetch 404s; and /api/v1/providers/{cliproxyapi,9router}/models return synced embedded-service models instead of invalid_provider. (#5299, #5298 — thanks @KooshaPari)
• thinking (claude): fix three independent defects in Claude adaptive-thinking on the OpenAI-compatible path (Cursor → Claude OAuth). (A) the dashboard Thinking-Budget setting was dropped on every restart — setThinkingBudgetConfig was never called at boot, so a saved {mode:"adaptive"…} silently reverted to passthrough; it's now hydrated from settings in server-init. (B) the Claude executor force-injected adaptive thinking after translation, ignoring the operator's budget — it now honors mode:"auto" (strip) while keeping the default (passthrough) behavior byte-identical so native Claude Code is unaffected, and remaps an operator thinking.type:"enabled" to the adaptive shape Opus 4.7/4.8 require (enabled → 400). (D) on replay, signature-less reasoning_content was reconstructed as a thinking block carrying a fabricated signature → Anthropic 400 "Invalid signature in thinking block"; it now emits a signature-less redacted_thinking block (real signatures are still preserved verbatim). Regression guards: tests/unit/thinking-budget-hydration-5312.test.ts, base-thinking-budget-config-5312.test.ts, openai-to-claude-redacted-replay-5312.test.ts (existing #5123/#4479/#2454 suites stay green). The </think> content-marker channel mismatch (RC-C, shared with #5245) is tracked as a follow-up pending a live Anthropic validation. (#5312 — thanks @vitalNohj)
• opencode (proxy pool): the OpenCode Free per-account proxy modal now offers the global Proxy Pool dropdown (by-id reference) instead of forcing manual Host/Port/credentials on every account — Gap 1 of #5217. A Saved / Custom toggle: "Saved" picks a pre-saved proxy from GET /api/settings/proxies and stores {fingerprint, proxyId}, so updating that pool proxy applies to every account using it; "Custom" keeps the manual inputs (stored inline) as an escape hatch. Resolution happens server-side (resolveAccountProxiesFromRegistry) so the executor still receives a resolved proxy unchanged; existing inline entries keep working and an unknown/deleted proxyId degrades safely to direct. Regression guards: tests/unit/noauth-proxy-resolution.test.ts, tests/unit/ui/noauth-account-card.test.tsx. (#5217 Gap 1 — thanks @daniij)
• thinking (claude): let reasoning_content-native clients (e.g. Cursor) opt out of the </think> close-marker so it no longer leaks an orphan </think> into visible content (RC-C of #5312, shared with #5245). The marker-suppression machinery already existed (UA allowlist, #5348) but Cursor's UA was deliberately excluded; this adds an explicit request header x-omniroute-thinking-marker: off (also on/keep to force-keep) that overrides the UA policy. With the header absent the behavior is byte-identical — Claude Code/Cursor-composer clients that scan content for the marker (#4633) still receive it. Regression guard: tests/unit/think-close-marker-suppress-5245.test.ts (#5123 case-b + #4479 stay green). (#5312, #5245 — thanks @vitalNohj, @wild-feather)
• cors: browser/Electron clients (e.g. Wayland AI) can now use OmniRoute as an OpenAI-compatible provider out-of-the-box. The token-authenticated API surface (/v1/*, /v1beta/*) now returns a permissive Access-Control-Allow-Origin (echoes the request Origin, * when absent) by default — matching 9router and the OpenAI-compatible ecosystem — so a renderer fetch can read the response instead of failing CORS-blocked as "site not found" / empty catalog (while curl, which sends no preflight, worked). This is safe: those routes auth via Authorization/x-api-key headers browsers never auto-attach (no credentialed-session/CSRF exposure), and Access-Control-Allow-Credentials is never paired with the echo/wildcard. Cookie-authed MANAGEMENT/dashboard routes stay exactly fail-closed; CORS_ALLOW_ALL/CORS_ALLOWED_ORIGINS still take precedence. Regression guards: tests/unit/cors/origins.test.ts, tests/unit/authz/pipeline.test.ts. (Bug 2 of #5242 — thanks @jonlwheat2-gif)
• grok-web: forward the Cloudflare clearance cookies and stop mislabeling IP-reputation blocks as a bad cookie. "Check cookie" returned Invalid SSO cookie even with a valid, complete browser session — but the cookie parser was never the problem (it robustly extracts sso/sso-rw from a full DevTools header). Two real gaps fixed: (1) buildGrokCookieHeader now forwards cf_clearance and __cf_bm when pasted (it dropped them before; AIClient2API forwards them too) — strictly additive, a bare sso blob still yields exactly sso=…; (2) when the user supplied a cf_clearance, a 401 / invalid-credentials-403 from grok.com is now surfaced as an IP-reputation/anti-bot block (cf_clearance is IP+TLS+UA-pinned and can't be replayed from a different machine) instead of the misleading "Invalid SSO cookie — re-paste". A bare cookie with no clearance still gets the re-paste hint. Regression guards in web-cookie-auth.test.ts + provider-validation-specialty.test.ts. (#5350 — thanks @SeaXen)
• cli (serve): opt-in native HTTPS/TLS for omniroute serve — so strict-CSP Electron apps and browsers can reach OmniRoute over https:// instead of plain http://localhost. Provide --tls-cert <path> --tls-key <path> (or OMNIROUTE_TLS_CERT/OMNIROUTE_TLS_KEY) and the standalone server terminates TLS on the same listener (no extra port/proxy); WebSocket upgrade (live dashboard + /v1 streaming) works over wss:// unchanged since https.Server extends http.Server. With no TLS flags the HTTP path is byte-identical to before; only one of cert/key, or an unreadable path, logs a warning and stays HTTP (never half-enables, never crashes). Auto-generated self-signed certs for localhost are a follow-up; for now provide an explicit cert/key (or front OmniRoute with a TLS terminator). Regression guard: tests/unit/tls-options.test.ts. (Bug 1C of #5242 — thanks @jonlwheat2-gif)
• opencode/observability: make OpenCode Free account/proxy rotation visible and fix two real defects surfaced alongside it. (1) the per-request rotation selection log (dispatch via account … through proxy …) was debug (hidden at default APP_LOG_LEVEL=info) — promoted to info so the shuffle/cooldown lifecycle is auditable (token stays masked). (2) [ProxyEgress] reported proxy=direct even when an account proxy was applied, because the egress logger ran outside the executor's nested proxy context — the effective applied proxy is now captured (via an applied-proxy sink threaded through the proxy AsyncLocalStorage) and reflected in the egress log. (3) [callLogs] too many SQL variables — deleteCallLogRowsByIds deleted up to 5000 ids in one IN (…), exceeding SQLite's ~999 bound-param cap and aborting log trimming/retention; ids are now chunked (≤500 per statement). Regression guards: tests/unit/call-log-trim-sql-vars-5217.test.ts, apply-executor-proxy-info-5217.test.ts, extended opencode-proxy-rotation-4954.test.ts. The Proxy Pool dropdown (by-id) UI (Gap 1) is a follow-up requiring browser validation. (#5217 — thanks @daniij)
• chatgpt-web: wire tool/function calling into the chatgpt-web provider. It was the only web-session executor that never read body.tools — both response builders hardcoded finish_reason:"stop" and emitted only content, so tool calls were silently dropped (the model answered in prose). It now uses the shared webTools prompt-emulation shim (a <tool>-contract system message + <tool>{…}</tool> response parsing) exactly like its 9 sibling executors (qwen-web, perplexity-web, …) — it was simply omitted from the #3259 rollout. Tool mode buffers and emits tool_calls + finish_reason:"tool_calls" (gated off the image-gen path); plain chat is unchanged. Regression guard: tests/unit/chatgpt-web-tools-5240.test.ts. (#5240 — thanks @Rougler)
• oauth/dashboard: fix the persistent/false Antigravity "Token Expired" badge (continuation of #3679/#3850). Two causes: (1) new OAuth connections never set tokenExpiresAt (only expiresAt), so the dashboard badge — which prefers tokenExpiresAt || expiresAt — fell back to the original grant clock and could flash a false "Token Expired" until the first background refresh. Creation now mirrors expiresAt into tokenExpiresAt across all 5 OAuth create paths (a shared buildOAuthConnectionCreatePayload), consistent with every refresh path which already writes both. (2) when a refresh-capable connection has no usable refresh token, the health-check sweep silently skipped it, leaving testStatus="active" forever while the cosmetic badge showed expired; it now surfaces a terminal testStatus="expired" ("needs re-auth"), tightly gated so it never clobbers non-refresh providers or already-terminal/cooldown states. Regression guards: tests/unit/oauth-connection-tokenexpiresat-5326.test.ts, tests/unit/token-health-no-refresh-token-expired-5326.test.ts. (#5326)
• routing: auto-disable a depleted API key on upstream 402 "Insufficient account balance" for API Key Round-Robin connections (multiple keys in one connection's extraApiKeys). The per-connection path already terminalized 402 (→ credits_exhausted), but the per-KEY health tracker (recordKeyHealthStatus) only recorded failures for 401, so a 402-depleted key stayed in rotation and kept getting retried. Now a 402 marks the current key invalid immediately (terminal — balance won't recover mid-session) via a new recordKeyTerminal, so the rotator skips it and falls over to the next healthy key; the state persists across restarts. Also added insufficient balance/insufficient_balance/insufficient account balance to the credits-exhausted body signals so non-402 out-of-credit responses terminalize too. Regression guard: tests/unit/key-health-402-disable-5239.test.ts. (#5239 — thanks @muflifadla38)
• cli: omniroute serve no longer discards a user-set NODE_OPTIONS=--max-old-space-size=…. It used to unconditionally overwrite NODE_OPTIONS (and pass an explicit --max-old-space-size CLI arg) with the calibrated default, so a user who exported --max-old-space-size=8192 still ran at the old cap and OOM'd (#5238 reporter set 8192, crashed at ~505MB). Now it mirrors the Electron and standalone launchers: if NODE_OPTIONS already pins the heap, that value wins (and the duplicate CLI arg is suppressed); otherwise the calibrated --max-old-space-size is appended, preserving unrelated flags. Regression guard: tests/unit/serve-node-options-preserve-5238.test.ts. (Defect C of #5238; the b.mask/OOM-root parts are tracked separately.)
• dashboard: restore the {active}/{total} active model-count badge in a provider's Available Models toolbar (provider detail page). It was dropped during the v3.8.13 god-file decomposition (#3327) — the ModelVisibilityToolbar still received activeCount/totalCount but they were orphaned as unused _-prefixed params and the rendering <span> was never carried over (the modelsActiveCount i18n key stayed). Re-wired the existing props to the existing key; zero data-layer or i18n change. Regression guard: modelVisibilityToolbarActiveCount.test.tsx. (#5264)
• rerank: /v1/rerank no longer rejects SiliconFlow and DeepInfra Qwen3-Reranker models with 400 "Invalid rerank model" even though /v1/models lists them. The model-ID parser was never the problem (it already splits on the first slash, so siliconflow/Qwen/Qwen3-Reranker-8B parses correctly) — siliconflow and deepinfra were just missing from the rerank provider registry. Added both: SiliconFlow as Cohere-compatible, DeepInfra via a new deepinfra adapter (model in the URL path POST /v1/inference/<model>, {queries,documents} request, positional {scores} response mapped to Cohere results[]). Regression guard: tests/unit/rerank-providers-5332.test.ts. (#5332 — thanks @maikokan)
• authz/dashboard: stop rejecting every dashboard mutation with 403 INVALID_ORIGIN when the dashboard is reached over a LAN IP / non-localhost host. The origin-pinning check (#5278) only accepted the configured *_PUBLIC_BASE_URL (typically http://localhost:20128) plus the internal request.url origin — which Next.js standalone reports as the bind host, not the real Host. So opening the dashboard at e.g. http://192.168.0.15:20128 made the browser's same-origin Origin match no candidate, and every POST/PUT/DELETE (save API key, save provider, test connection) failed while GETs still worked. Two fixes: (a) the request Host (or a trusted X-Forwarded-Host) is now accepted as a valid mutation origin, gated by two independent checks — the token-stamped socket peer must be loopback/private-LAN and the Host itself must be a loopback/private-LAN IP literal, so a DNS-rebinding domain (which classifies as remote) can never become a trusted origin and the protocol is pinned to the actual connection; (b) the INVALID_ORIGIN response now carries an actionable message (set OMNIROUTE_PUBLIC_BASE_URL) and the dashboard surfaces API error .message via a shared extractApiErrorMessage helper instead of rendering the raw error object. Regression guards: tests/unit/authz/public-origin.test.ts (direct LAN/loopback + DNS-rebinding defense), tests/unit/api-error-message-5340.test.ts. (#5340)

📝 Maintenance

• chore(dead-code): repo-wide sweep of unused exported symbols and a matching dead-code baseline ratchet — trimmed unused exported helpers, validation/settings/encryption-config schemas, utility/domain/static-constant/formatting helpers, runtime test helpers, the request-timeout fetch wrapper, event-bus, semantic-cache (maintenance + expiry), correlation-middleware, MCP-scope, service-registry, build-profile, api-key-format, authz-class, models.dev-context, embedding-cache, provider-limits-scheduler, search-validator, webhook-example, agent-skills-repo-URL and command-code-auth-cleanup exports. Pure dead-code removal validated by typecheck:core (no remaining referencing site) — no behavior change. (#5321, #5322, #5324, #5325, #5328, #5329, #5330, #5331, #5333, #5334, #5335, #5336, #5338, #5339, #5353, #5354, #5355, #5356, #5357, #5359, #5362, #5364, #5365, #5366, #5368, #5369, #5371 — thanks @JxnLexn)

---

[3.8.40] — TBD

In development — bullets added per PR; finalized at release.

---

[3.8.39] — 2026-06-28

✨ New Features

• feat(oauth): remote Antigravity login via local helper + paste-credentials — Antigravity (and other Google "native/desktop" OAuth providers) use Google's firstparty/nativeapp consent, which only releases the auth code when the loopback redirect (127.0.0.1:<port>) is reachable from the approving browser. On a remote VPS install that loopback lives on the server, so the consent hangs forever and never emits a code — the "paste the callback URL" fallback has nothing to paste (a Google-side constraint, identical in upstream 9router). A new omniroute login antigravity CLI helper runs the OAuth on the user's own machine (where 127.0.0.1 works), exchanges the code, and prints a single-line omniroute-cred-v1.… credential blob; the dashboard's Antigravity Connect → Step 2 field now accepts that blob (alongside callback URLs) and persists the connection via a new paste-credentials action (server-side onboarding, provider-allowlisted, with the blob's embedded provider required to match the route). The SSH local-forward tunnel is documented as a zero-tooling alternative. See docs/guides/REMOTE-MODE.md. (#5203)
• feat(agent-bridge): graceful cert-install fallback for containers / headless — when the MITM root CA can't be installed into the system trust store automatically (Docker / headless / no sudo / read-only trust store), the Agent Bridge no longer hard-fails on start with a generic "Certificate install failed". It now starts in skip mode and the dashboard surfaces a platform-specific manual-install guide (plus a CA download link) so the operator can trust the certificate by hand. The trust-cert endpoints return a structured { skippable, manualGuide } response (HTTP 200) for environment failures instead of a 500; an explicit user cancellation is still reported distinctly. (#4546 — thanks @phuchptty)
• feat(compression): CCR ranged/grep/stats retrieval (ReDoS-safe, backward-compat) — extends the omniroute_ccr_retrieve MCP tool and /api/compression/retrieve endpoint with optional range (byte/line slice), grep (ReDoS-safe literal or bounded-pattern match against stored lines), and stats (byte/line/word counts) parameters so agents pull exactly the slice or summary they need instead of re-expanding the entire stored block. All parameters are optional — no parameters returns the full block byte-identical to the existing behavior; the CCR store written by ionizer/fuzzy/headroom is fully compatible. Sixth item of the compression roadmap. (#5187)
• feat(compression): TOON best-of-N candidate encoder + encoder A/B table — adds @toon-format/toon as a candidate encoder in the headroom compression engine via a best-of-N scheme: both GCF and TOON run per prompt and the shorter result is kept, rather than hard-swapping encoders (GCF already encodes the headroom block and TOON is not a lossless universal win). An encoder A/B comparison table (GCF vs TOON vs JSON — bytes and cl100k tokens) is now surfaced in the compression studio. Fifth item of the compression feature-extraction roadmap (bench: #5080, gate: #5127, fuzzy/gate: #5143, ionizer: #5148). (#5163)

🔧 Bug Fixes

• fix(oauth): Antigravity refresh no longer nulls the stored refresh_token on an empty upstream response — Google's OAuth token endpoint uses non-rotating refresh tokens: a refresh response normally OMITS refresh_token and occasionally returns it as an empty string. The Antigravity executor's refreshCredentials used typeof tokens.refresh_token === "string" ? tokens.refresh_token : credentials.refreshToken, and because typeof "" === "string" is true, an empty-string response overwrote the good token with "" — nulling it on first refresh. The check now treats a non-string or empty value as absent and preserves the stored token, matching the canonical refreshGoogleToken (tokens.refresh_token || refreshToken) semantics. (#3850 — thanks @3xa228148)

• fix(api): LAN/Tailscale dashboard access — ws: CSP scheme, GET-exempt version route, surface combo field errors — three failures when opening the dashboard from a non-loopback host: (1) CSP connect-src allowed the ws: scheme only for loopback origins, blocking the dashboard's ws://<lan-host>:* Live WebSocket from LAN/Tailscale clients; the bare ws: scheme is now permitted (symmetric with the bare wss: already allowed), kept declarative in next.config.mjs with no global middleware (the project has none by design); (2) GET /api/system/version was blocked by LOCAL_ONLY_API_PREFIXES for all methods despite only POST spawning child processes (git/npm/pm2) — a new LOCAL_ONLY_API_GET_EXEMPTIONS set exempts safe read methods for this path while keeping POST/PUT/PATCH/DELETE strictly loopback-only; (3) COMBO_002 validation errors only surfaced the generic message — firstField/firstMessage are now extracted from the first Zod issue and included in the response body. (#5083 — thanks @KooshaPari for the diagnosis and original PR #5084)

• fix(sse): defer </think> close so it never leaks before tool_calls in Claude→OpenAI streaming — when a Claude thinking block was followed by a tool_use block, the translator unconditionally emitted a content: "</think>" chunk at content_block_stop, injecting a spurious assistant text chunk immediately before the tool_calls delta and corrupting OpenAI-compatible clients (e.g. Kimi Coding). The close marker is now deferred: it is flushed at the first text_delta that follows the thinking block (preserving the #4633 / decolua/9router#454 behavior for Claude Code / Cursor) or at stream finish when no tool_calls were collected. Tool-use streams never get a text_delta after the thinking block, so </think> is never emitted into content before tool_calls. (#5123)

• fix(sse): normalize array user-message content in the Command Code executor to prevent upstream 400 — when a client sends a user turn whose content is an array of content parts (e.g. [{type:"text",text:"…"}, …]), the raw array was forwarded verbatim to the Command Code upstream, which requires messages[N].content for the user role to be a plain string — resulting in expected string, received array / HTTP 400 on DeepSeek V4-Pro and other Command Code models. The user branch of convertMessages now calls normalizeContentText() (already used by system, assistant, and tool branches) so multi-part user content is joined to a string before dispatch. Partially addresses (#5166); the 0-output-token symptom on reasoning-only models is tracked separately.

• fix(mcp): return HTTP 404 (not 400) for an unknown/expired Streamable HTTP session id — when an MCP session is terminated or idles out and the client reuses the stale Mcp-Session-Id header, the Streamable HTTP transport replied with HTTP 400. The MCP spec (2025-03-26 and 2025-11-25, Session Management) mandates HTTP 404 Not Found in that case, and spec-compliant clients only re-initialize a session on 404 — so the 400 was non-recoverable. The handler now returns 404 for a present-but-unknown session id, while a missing session id on a non-initialize request correctly stays 400. (#5169 — thanks @czer323)

• fix(api): blocking "Auto (Zero-Config)" in Security settings now removes auto/* from /v1/models — the built-in auto/* combo advertiser (#4164 / #4235) at the top of the models catalog ignored settings.blockedProviders, so checking Auto (Zero-Config) under Security → Blocked Providers had no effect and the model picker kept listing every auto/* entry. The injection loop now skips the entire auto/* block when the system provider auto (its id and alias are both auto) is blocked, consistent with how every other provider is filtered from the catalog. (#5192 — thanks @WslzGmzs)

• fix(cli): auto-calibrate the server V8 heap from physical RAM instead of a fixed 512MB default — the server was spawned with a hard-coded --max-old-space-size=512 (omniroute serve) or with no heap flag at all (Electron desktop, which then inherited the runtime's low ~512MB default), so RAM-rich machines still OOM-crashed under load (FATAL ERROR: Ineffective mark-compacts near heap limit … ~500MB at code=134) with many providers/accounts and large model catalogs (one report: 16GB RAM, 65 providers, ~100 accounts, ~2600 models). A new calibrateHeapFallbackMb(os.totalmem()) helper derives the default heap as ~35% of physical RAM, clamped to [512, 4096], and is wired into both bin/cli/commands/serve.mjs and electron/main.js. An explicit OMNIROUTE_MEMORY_MB (or a pre-set --max-old-space-size) still wins, so the #2939 override contract is unchanged. (#5172, #5160, #5152 — thanks @manchairwang, @Xyzjesus)

• fix(oauth): Antigravity login no longer hangs — fire-and-forget onboarding + bounded post-exchange — the dashboard's Antigravity OAuth login spun indefinitely because postExchange awaited the onboardUser retry loop inline (up to 10 × 5 s per attempt, each fetch with no timeout), blocking the /exchange response forever. Matching the upstream 9router web flow: onboardUser now runs fire-and-forget in a background task; the /exchange endpoint is bounded by a 10 s hard timeout so it always returns; a progress endpoint lets the dashboard poll onboarding completion state. (#5193)

• fix(antigravity): retry Antigravity accounts by quota family before escalating the combo — when one Antigravity account returns a quota or rate-limit 429 for a Gemini model (e.g. gemini-3.5-flash-medium), combo orchestration could prematurely advance to the next combo model instead of trying other eligible Antigravity accounts for the same quota family. Antigravity quota-family awareness is now added to the fallback path so a 429 on one account triggers a bounded same-model retry across other Antigravity accounts sharing that quota bucket before the combo degrades to a lower-tier model. (#5180 — thanks @Ardem2025)

• fix(translator): accept Claude Messages shape in the non-stream malformed-200 guard — when a Claude client (e.g. Claude Code) is routed to a non-Claude provider, the translated non-streaming response body is in Claude Messages shape (type: "message", content[]) produced by convertOpenAINonStreamingToClaude. detectMalformedNonStream only recognized OpenAI choices[].message and Responses API output[], so this shape fell through to empty_choices → 502. The guard now recognizes the Claude Messages shape: text, tool_use, and thinking blocks carrying a signature count as valid output, while a genuinely empty content: [] is still flagged. (#5156 — thanks @NomenAK)

• fix(sse): resolve nameless deepseek-web <tool> blocks via parameter-schema match — when chat.deepseek.com emits a <tool> block with no <name> child, no JSON body name/type key, and no tag suffix, every name-resolution path in extractCall returned null and the raw XML leaked to the client as plain text. A conservative schema-based fallback now compares the block's extracted parameter names against each declared tool's schema keys; if exactly one tool matches, its name is used. Zero or ambiguous (>1) matches still return null so no calls are misattributed. (#5154, #5173)

• fix(stream): normalize provider safety finish reasons to content_filter — Gemini and Antigravity can return safety/prohibited terminal reasons (SAFETY, RECITATION, BLOCKLIST, PROHIBITED_CONTENT) that OpenAI-compatible downstream clients do not recognize. A shared finish-reason normalization helper now maps these to the standard content_filter value, applied in both the streaming and JSON collection paths for both providers. (#5197 — thanks @rdself)

• fix(responses): normalize non-array Responses API input before routing — the OpenAI Responses API accepts input as a string, object, or list, but OmniRoute only handled list-shaped payloads; a string or object input was silently dropped on the Responses→Chat Completions path. The translator now normalizes input to a list before dispatch; the Codex-native Responses path also normalizes before forwarding (preventing upstream 400 Input must be a list); and the prompt-injection and PII sanitizer extraction paths are guarded against object-valued input so security checks do not throw. (#5204 — thanks @wilsonicdev)

• fix(zenmux): normalize vendor-prefixed GLM system roles for Z.AI models — ZenMux exposes Z.AI GLM via vendor-prefixed OpenAI-compatible IDs such as z-ai/glm-5.2. The existing GLM detection only matched bare glm-*/glm ids, so zenmux/z-ai/glm-5.2 kept system messages in place; Z.AI rejects compressed histories ending with a system turn before assistant(tool_calls) → tool sequences. The fix extends GLM detection to cover z-ai/glm-* prefixes and routes them through the existing normalizeSystemRole path. (#5158 — thanks @Thinkscape)

• fix(xai): add OAuth connection test probe + normalize xAI reasoning effort aliases — xAI rejects unsupported reasoning effort values (max, xhigh) with HTTP 400 after a provider update; the xAI translator now maps max and xhigh to high before forwarding. Additionally, xAI OAuth connections had no dashboard test configuration, so provider tests returned "Provider test not supported"; a dedicated OAuth test probe is now wired for xAI accounts with regression coverage for the effort normalization. (#5157 — thanks @nguyenxvotanminh3)

• fix(serve): honour HOSTNAME from .env instead of hardcoding 0.0.0.0 — bin/cli/commands/serve.mjs spread process.env into the child-process environment but immediately overwrote HOSTNAME with a literal "0.0.0.0", silently discarding any user-configured bind address even though HOSTNAME is documented in .env.example and docs/reference/ENVIRONMENT.md. dist/server.js already read process.env.HOSTNAME correctly; only the CLI wrapper was overriding it. The fix applies process.env.HOSTNAME || "0.0.0.0" so the env value takes effect. (#5134, #5170 — thanks @anki1kr / @Angelo90810)

• fix(cli): force NODE_ENV to match dev/start run mode in the custom Next server — when .env.example ships NODE_ENV=production, starting npm run dev via scripts/dev/run-next.mjs forwarded that value to the programmatic next() entry, which — unlike the next CLI — does not normalize it to match the run mode. The resulting production flag caused PostCSS to skip Tailwind's CSS transform, surfacing as Module parse failed: Unexpected character '@' on globals.css. The custom server now explicitly forces NODE_ENV=development for the dev path and NODE_ENV=production for the start path regardless of .env. (#5189 — thanks @backryun)

• fix(cli): raise dev server Node heap limit to 8 GB to prevent OOM — npm run dev crashed with FATAL ERROR: Ineffective mark-compacts near heap limit — Allocation failed - JavaScript heap out of memory while compiling heavy dashboard routes because node scripts/dev/run-next.mjs ran on V8's ~4 GB default with no --max-old-space-size flag. The dev npm script now passes --max-old-space-size=8192 at invocation time (the only point where this flag can be set for that process). (#5198 — thanks @backryun)

• fix(cli): re-enable Turbopack as the default npm run dev bundler — PR #4092 forced webpack because an earlier Turbopack 16.2.x panic (internal error: entered unreachable code: there must be a path to a root in turbopack-core/module_graph) blocked the OmniRoute module graph. That panic no longer reproduces on the pinned Next 16.2.9, so OMNIROUTE_USE_TURBOPACK is flipped from 0 to 1 in .env.example, aligning it with docs/reference/ENVIRONMENT.md which had already documented the default as 1. (#5206 — thanks @backryun)

• fix(auth): allow synthetic no-auth fallback for mimocode — mimocode connections without explicit credentials were blocked before reaching the executor. The auth layer now permits a synthetic no-auth fallback for the mimocode provider so credential-free access patterns work as intended. (#5205 — thanks @KooshaPari)

• fix(combo): reject empty Responses API output: [] as a fail-over trigger — a non-streaming Responses API body with object: "response" and output: [] was accepted as a valid HTTP 200 by the combo response-quality validator, allowing a combo target to stop rather than fail over to the next leg. The non-stream validator now inspects Responses-API-shaped bodies before the generic output shortcut and rejects an empty output: [] as empty_choices; structural non-empty output (e.g. function_call) remains valid. (#5207 — thanks @KooshaPari)

• fix(proxy): close cached dispatchers when clearing the proxy cache — cached proxy and direct-retry dispatchers were not closed on cache clear, leaking open connection handles. The cache-clear path now calls close() on all evicted dispatchers; dispatcher cache and lifecycle helpers have been extracted from the oversized proxy-dispatcher module into a dedicated helper for reuse. (#5202 — thanks @KooshaPari)

• fix(proxy): coalesce concurrent fast-fail health probes per proxy URL — under high concurrency each simultaneous request opened its own TCP health probe for the same proxy URL, creating a thundering-herd burst. Concurrent proxy fast-fail checks are now coalesced so only one TCP probe runs per proxy URL at a time; the completed-result health cache is preserved so subsequent same-URL checks return immediately. (#5109, #5208 — thanks @KooshaPari)

• fix(pwa): prefer cached navigation before showing the offline page — the service worker was too eager to display /offline on transient navigation failures. It now caches successful navigation responses and consults the cached route or app shell before falling back to /offline; /offline remains the final fallback when no cached navigation or app shell exists. (#5165, #5209 — thanks @KooshaPari)

• fix(request-logger): never render a negative percentage in the compression badge — when every prompt token was compressed (totalIn = 0, compressed > 0), the compression pill displayed (-100%) because the badge format hard-coded a leading - before the percentage value. The badge now omits the negative sign in this case, correctly representing the saving as a positive ratio. (#5201 — thanks @KooshaPari)

• fix(dashboard): use amber for home update-step warning icon — the warning-state icon in the home update steps (HomePageClient.tsx) used text-yellow-500 (Tailwind #eab308), which has poor contrast on light backgrounds (~1.9:1, below WCAG AA) and is inconsistent with the amber warning convention used by every sibling element in the same component. Switched to text-amber-500 — a one-line className change with no behavior change. (#5176)

📝 Maintenance

• test(combo): deterministic context-relay universal-handoff coverage — covers the universal (provider-agnostic) session-handoff path in context-relay (combo.ts:2099–2139), which previously had only a definition-order assertion and a TODO(phase-2). The test drives the real pipeline via session seams (x-session-id → relayOptions.sessionId → maybeGenerateUniversalHandoff) without live infrastructure. (#5168)
• test(combo): end-to-end quota-share DRR routing-decision coverage (matrix parity) — adds the missing E2E test for the quota-share strategy, driving the real handleChat → chatCore → selectQuotaShareTarget → executor pipeline via in-process seams and asserting which connection is dispatched. The DRR selector already had 29 unit tests; this closes the E2E gap and brings quota-share to parity with the 17-strategy public matrix. (#5179)
• test(combo): deterministic context-relay codex quota-handoff coverage (closes last gap) — covers the codex-specific handoff block of context-relay (combo.ts:2143–2183), which #5168 left documented-but-untested because it requires a codex connection. All seams (fetchCodexQuota, handoff generation, session relay) are mocked deterministically without live infra. (#5195)
• test(ci): wire antigravity-quota-family test under test:vitest (fix test-discovery orphan) — open-sse/services/__tests__/antigravity-quota-family.test.ts (introduced by #5180) was not collected by any active runner, causing check:test-discovery to report a new orphan and gate every subsequent PR on the release branch. The file is now added to vitest.mcp.config.ts include and the corresponding orphan-allowlist entry is removed. (#5196)
• test(security): regression guard — PII redaction stays opt-in (default off) + Hard Rule #20 — adds a test asserting both PII_REDACTION_ENABLED and PII_RESPONSE_SANITIZATION feature-flag defaultValue fields are "false" and that data passes through all three application points (piiMasker, piiSanitizer, streamingPiiTransform) untouched when both flags are off, encoding Hard Rule #20 as a CI-enforced contract and fixing a misleading doc implication that PII masking was on by default. (#5159)
• docs(i18n): add Traditional Chinese (zh-TW) README + update zh-CN — adds a new Traditional Chinese translation (docs/i18n/zh-TW/README.md) and updates the Simplified Chinese README to the current English baseline; the language index (docs/i18n/README.md) and root README.md badge row are updated accordingly. (#5162 — thanks @lunkerchen)
• docs(i18n): full sync of zh-TW and zh-CN README to canonical English v3.8.39 — brings both translations to full parity, adding the complete What's New section, compression real-token examples, and all sections updated in the v3.8.38/39 English README. (#5171 — thanks @lunkerchen)
• docs(combo): sync combo/routing-strategy docs to current state + document test coverage — removes a stale ordinal from the Fusion bullet in README.md; adds a new Testing & Coverage section to docs/routing/AUTO-COMBO.md documenting the deterministic strategy matrix (npm run test:combo:matrix), quota-share DRR E2E coverage, and context-relay handoff tests delivered across the v3.8.39 cycle. (#5185)
• fix(docker): copy the open-sse workspace manifest before npm ci so workspace-only deps install — the Dockerfile copied only the root package*.json, so npm ci skipped safe-regex and @toon-format/toon (declared in open-sse/package.json, not hoisted to root), breaking the multi-arch image build with Module not found during npm run build. (thanks @diegosouzapw)

---

[3.8.38] — 2026-06-27

✨ New Features

• feat(sidebar): colored menu icons — sidebar menu icons now render with a per-item accent color: curated colors for known items (SIDEBAR_ICON_ACCENTS) plus a deterministic hash-based fallback (getSidebarIconAccent) so every item gets a stable, distinct color across sessions. (#3812 — thanks @rafacpti23)
• feat(providers): add Factory (factory.ai) as a subscription gateway provider — factory (Factory Droids' hosted gateway) is now a first-class routing provider on the OpenAI-compatible https://api.factory.ai/v1 endpoint with Bearer apikey auth; the key is supplied from the Dashboard connection (not env). (#5065 — thanks @KooshaPari)
• feat(providers): add Grok Build (xAI) provider with OAuth import-token flow — grok-cli (alias gc) routes through Grok's CLI chat proxy; users paste their ~/.grok/auth.json (or the JWT), with automatic refresh_token rotation. The public xAI client_id is embedded via resolvePublicCred("grok_id") (Hard Rule #11), never a literal. (#5020 — thanks @fulorgnas)
• feat(dashboard): click-to-edit model alias in the provider page — click an alias to edit it inline (Enter/blur saves, Escape cancels), instead of only being able to delete and re-add it. (#5119 — thanks @waguriagentic)
• feat(providers): add ZenMux Free (session-cookie free-tier) provider — zenmux-free (alias zmf) with a dedicated executor translating ZenMux's Anthropic-style SSE to OpenAI format; ships 12 free-tier models (DeepSeek V3.2, GLM 4.7 Flash Free, etc.). (#5105 — thanks @mrnasil)
• feat(providers): allow local/private provider URLs by default (Allow Local Provider URLs flag) — adding/validating an OpenAI-compatible provider on a loopback/LAN address (e.g. http://127.0.0.1:3264/api) was rejected by the SSRF guard with "Blocked private or local provider URL", even though OmniRoute is local-first. A new OMNIROUTE_ALLOW_LOCAL_PROVIDER_URLS feature flag (default ON, toggle in Settings → Feature Flags) now scopes the provider-validation guard to allow local/private hosts while still blocking cloud-metadata endpoints (169.254.169.254, metadata.google.internal). Disable it to restore strict public-only blocking. Webhook/remote-image SSRF defaults are unchanged. (#5066, thanks @daniij)
• feat(blackbox): refresh provider model catalog with latest models. (thanks @ptkelanatechsolutions)
• kiro: inline <thinking> stream splitter — when <thinking_mode>enabled</thinking_mode> is present, assistantResponseEvent content is now split into separate delta.content / delta.reasoning_content SSE chunks (new open-sse/executors/kiroThinking.ts module wired into KiroExecutor.transformEventStreamToSSE).
• feat(cursor): parse Cursor Composer DeepSeek-style inline tool calls — Composer cu/composer-2.5* models embed tool invocations in their visible text using <｜tool▁calls▁begin｜>…<｜tool▁calls▁end｜> markers instead of structured protobuf frames; a new streaming parser (composerToolCalls.ts) intercepts these in both streaming and non-streaming paths, suppresses the markers from the client-visible content, and emits proper OpenAI tool_calls deltas so downstream clients handle them natively. (thanks @noestelar)
• feat(proxy): support auth-less host:port batch import and surface proxy-test failures. (thanks @dimaslanjaka)
• feat(video): Alibaba DashScope video provider (wan2.7-t2v) — adds the alibaba video provider (DashScope async task → poll → MP4) wired through the standard apikey credential path, so text-to-video requests can route to Alibaba's wan2.7-t2v model. (thanks @josevictorferreira)
• feat(cc): per-connection "summarized thinking display" toggle for Claude-Code-compatible providers — exposes a connection-level toggle that drives the existing Copilot summarized-thinking marker, so operators can opt a CC-compatible connection into summarized reasoning display from the UI (schema + request defaults + provider modals, with i18n). (thanks @rdself)
• feat(compression): compression playground in the studio (Play + Compare tabs) — /dashboard/compression/studio gains a synthetic playground: paste text → per-engine lanes (each deterministic engine run alone via /api/compression/preview) plus a combined waterfall ordered by stackPriority, and a free A/B Compare grid with on-demand, USD-capped fidelity verdicts (/api/compression/compare + compare/verify). The preview route now uses the real cl100k tokenizer, returns engineBreakdown, and accepts an ordered pipeline[]; new compare / compare/verify / retrieve routes; the live WS feed moved to /dashboard/compression/live. Management-only. (#5080)
• feat(dashboard): expose Fusion judgeModel + fusionTuning in the combo editor — the Fusion strategy editor now surfaces the judge model (synthesizes the panel answers; defaults to the first panel model) plus the quorum-grace tuning fields (minPanel, stragglerGraceMs, panelHardTimeoutMs) that open-sse/services/fusion.ts already reads. Schema-validated + bounded; empty tuning is never persisted. (#5074)
• feat(compression): opt-in per-step fidelity gate for the stacked pipeline — each compression step can now be guarded by a pure fidelity checker (4 invariants, fail-open) so a lossy engine that would degrade the prompt past a threshold is rejected and its lane skipped instead of silently shipping. Configurable via fidelityGate (advanced thresholds intentionally API-omitted), with a per-lane rejection breakdown surfaced in the studio playground toggle. (#5143)
• feat(compression): fuzzy near-duplicate dedup (session-dedup 2nd pass) — the session-dedup engine gains a second fuzzy pass that collapses near-duplicate (not just byte-identical) segments, with a playground toggle to compare on/off. (#5143)
• feat(quota): opt-in Codex/Claude auto-ping keepalive — an opt-in background keepalive can periodically ping Codex/Claude connections to keep their session/quota state warm, reducing cold-start failures on the first real request. (#5102)
• feat(ops): SRE playbooks + ops helper scripts — salvaged from a closed stale PR; adds operator runbooks and ops helper scripts. (#5138 — thanks @KooshaPari / @diegosouzapw)
• feat(mcp): web-session robustness — cookie dedup + browser-pool observability — the MCP web-session path now de-duplicates cookies when (re)hydrating a session (avoiding conflicting duplicate Cookie headers) and exposes browser-pool observability (pool size / in-use / acquisition metrics) for the headless web providers. (#5121, builds on #3368)
• feat(compression): Ionizer engine — lossy JSON-array sampling reversible via CCR — a new compression engine that down-samples large JSON arrays to a representative subset and records a Compact Change Representation (CCR) so the omitted rows can be reconstructed, trading exactness for a large token reduction on tabular/array-heavy payloads. (#5148)

🔧 Bug Fixes

• fix(proxy): make the SOCKS5 handshake timeout operator-tunable (SOCKS_HANDSHAKE_TIMEOUT_MS) — under high concurrency against a single residential gateway host, the SOCKS5 connect handshake could exceed the hardcoded 10s even though the proxy was reachable, surfacing as a false [Proxy Fast-Fail] Proxy unreachable (the pool size is already tunable via OMNIROUTE_PROXY_DISPATCHER_CONNECTIONS). The handshake timeout now reads SOCKS_HANDSHAKE_TIMEOUT_MS (default unchanged at 10000, capped at 120000) so a concurrency-heavy deployment can raise it without a code change. Mitigation for #5109 (the full concurrency-100 collapse still needs the reporter's live load-test confirmation). (#5109)
• fix(api): resolve GET /v1/models/{id} case-insensitively — clients that normalise the model id (e.g. OpenCode requesting minimax/minimax-m3 for the canonical catalog entry minimax/MiniMax-M3) missed the single-model lookup, which is case-sensitive, and fell back to advertising context_length: 0. findModelById now prefers an exact-case match and falls back to a case-insensitive match, so the real entry (and its context window) is returned regardless of casing. (#5082)
• fix(services): embed WS proxy honours LIVE_WS_HOST; reject empty messages early — two headless/Docker deployment fixes (#5110). The embed WebSocket proxy (:20131) only read EMBED_WS_PROXY_HOST, so behind a reverse proxy/tunnel it stayed bound to 127.0.0.1 even with LIVE_WS_HOST=0.0.0.0 set and the Live dashboard showed "WebSocket disconnected"; it now falls back to LIVE_WS_HOST (default still loopback). Separately, a request with an explicitly empty messages: [] array was forwarded upstream and bounced back as a confusing raw 400/502; handleChat now rejects it up front with a clear messages: at least one message is required (Responses-API input requests are unaffected). (#5110)
• fix(proxy): repair one-click Deno & Cloudflare relay deployments — the /api/settings/proxy/test endpoint only recognized the vercel relay type, so testing a deployed Deno or Cloudflare relay returned proxy.type must be http, https, or socks5 and never reached the relay; it now routes all relay types through isRelayType(). On installs with STORAGE_ENCRYPTION_KEY the relay-auth token is read via extractRelayAuth (encrypted relayAuthEnc form), fixing the silent 401 that left publicIp null. The Cloudflare Worker upload now sends the script part as application/javascript (the API rejects application/javascript+module; ES-module semantics come from main_module), and the proxy-registry schema accepts the deno/cloudflare types + deno-relay/cloudflare-relay sources so editing a deployed relay no longer 400s. (#5128)
• fix(kiro): retire claude-sonnet-4.5 from the Kiro catalog + pin the exact Kiro 400 error — claude-sonnet-4.5 left the Kiro free-tier lineup (current active models: Opus 4.8/4.7/4.6, Sonnet 4.6, Haiku 4.5), so it is removed from the Kiro registry entry and the free-model catalog. A regression test now pins Kiro's verbatim [400] Invalid model. Please select a different model to continue. to the isModelUnavailableError model-unavailable classification. A 400 on every model (including current ones) points to a server-side Kiro tier/region gate, not an OmniRoute catalog bug. (#5140, closes #4484)
• fix(dashboard): preserve every rendered field when loading/saving Resilience settings — ResilienceTab renders comboCooldownWait and quotaShareConcurrencyLimit, but both the initial-load and save paths rewrote component state without those fields, so after a successful /api/resilience response the cards received undefined and the page fell back to the generic "failed to load" state. A shared toResilienceResponse() mapper now keeps all rendered fields, and PATCH /api/resilience returns quotaShareConcurrencyLimit to match GET and the UI contract. (#5139 — thanks @rdself)
• fix(quota): hydrate the in-memory quota cache from snapshots + scope auto-combo candidates — after a restart the quota cache was empty, so a known-exhausted connection looked healthy until re-queried; isAccountQuotaExhausted now lazily hydrates from persisted quota_snapshots. Auto-combo candidate expansion is also scoped to the connections each combo target actually allows, instead of pulling in every connection for the provider. (#5015 — thanks @JxnLexn)
• fix(resilience): harden quota cutoff, Gemini audio MIME, and model-lockout cooldown — stored quota hard-cutoff values are no longer coerced to enabled=true from arbitrary strings; Gemini audio input parts have their MIME type validated/normalized before forwarding; and model lockout now honours the configured maxCooldownMs ceiling. (#5093 — thanks @KooshaPari)
• fix(streaming): harden long OpenAI-compatible SSE streams — a late pipeline-wind-down error can no longer overwrite an already-recorded successful stream (streamCompletionRecorded guard), client disconnects finalize as 499 client_disconnected instead of poisoning provider/account failure state, JSON bodies that are actually SSE (wrong application/json content-type) are sniffed and re-streamed, and reasoning fields (reasoning/reasoning_content + OpenRouter/Gemini encrypted reasoning_details) are preserved through the JSON-as-SSE fallback. (#5124 — thanks @rdself)
• fix(usage): dedupe request-usage logging and debounce stats events — saveRequestUsage now guards against duplicate inserts (natural key: timestamp + provider + model + connection + api-key + token counts), back-fills a missing endpoint, and only emits usageRecorded when a row was actually inserted; stats update/pending event bursts are collapsed into a single debounced notification to reduce churn. (#4940 — thanks @nguyenxvotanminh3)
• fix(sse): convert the native Gemini request body to OpenAI format in the Antigravity MITM handler — contents / systemInstruction / generationConfig / thinkingConfig are now translated to OpenAI chat-completions format before forwarding to /v1/chat/completions, so thinking-capable models (e.g. ag/claude-opus-4-6-thinking) no longer fail with provider-side 400 "invalid argument" errors. (#4845 — thanks @anuragg-saxenaa)
• fix(db): translate the two pt-BR SQLite driver-fallback log lines to English — [DB] Pré-inicializando sql.js WASM… and [DB] Drivers síncronos indisponíveis… were the only non-English server log strings, mixing languages in the logs. Now [DB] Pre-initializing sql.js WASM (synchronous drivers unavailable)… / [DB] Synchronous drivers unavailable — falling back to sql.js (WASM), guarded by a test that scans the driver path for accented log strings. (#5103)
• fix(diagnostics): non-streaming Claude responses no longer false-502 as empty_choices — the v3.8.37 malformed-200 detector (#4942) only understood OpenAI choices and Responses-API output shapes, so a /v1/messages response that stays in Claude shape ({type:"message", content:[…]}) fell through to empty_choices → 502 (cascading to "All models failed" in a combo). Most visibly, an extended-thinking turn whose buffered body is a single empty thinking block with a valid signature (Claude Code's non-streaming Bash classifier) 502'd on every call. detectMalformedNonStream now understands the Claude shape: text/tool_use blocks and thinking blocks carrying a signature count as valid output, while a genuinely empty content:[] is still flagged. (#5108, thanks @insoln)
• fix(combo): empty-content 502 now fails over within the same request instead of exhausting the provider — a leg that answers HTTP 200 with no usable completion is rewritten to 502 "Provider returned empty content", but the combo exhaustion classifier treated that synthetic 502 as a connection-level failure (#1731v2) and marked the whole provider/connection exhausted, skipping every remaining same-provider leg in that request. The connection is actually healthy (it just returned an empty body), so empty-content 502s are now classified as model-level transient failures: the request advances to the next leg and the rest of that provider's legs stay eligible. Genuine gateway 502s still trip connection exhaustion. (#5085, thanks @andrea-kingautomation)
• fix(dashboard): surface the detailed credential-validation error instead of a bare "invalid" badge — the inline "Check" in the Add-Connection modal discarded the error message returned by /api/providers/validate and showed only an invalid badge. For web providers (claude-web / chatgpt-web) the real cause is often an environment error the backend already reports (e.g. TLS impersonation client failed to start: EACCES … mkdir tls-client-node/bin), so users were left guessing. The modal now renders the full reason next to the badge. (#5088, thanks @tkhs101)
• fix(executors): strip client_metadata from forwarded body for Cerebras and Mistral — Cerebras returns 400 (wrong_api_format) and Mistral returns 422 (extra_forbidden) when the passthrough body carries client_metadata (an OpenAI Codex / Claude CLI field with no equivalent on these upstreams). The default executor now drops it for these two providers before sending downstream; other providers (notably openai/codex) keep it. (thanks @saurabh321gupta)
• fix(codebuddy): only send reasoning params when the client requests reasoning. (thanks @anki1kr)
• fix(sse): keep streaming for forceStream providers when a JSON client requests it. Providers marked forceStream:true reject stream:false upstream (HTTP 400); resolveStreamFlag now guards against this so stream-only providers keep streaming even when the client sends Accept: application/json or stream:false. (thanks @anki1kr)
• fix(sse): prevent non-JSON SSE lines and duplicate [DONE] from breaking clients. (thanks @qianze0628)
• fix(sse): dedupe case-variant Anthropic headers in the executor buildHeaders path — Node/undici's fetch merges anthropic-version and Anthropic-Version into a single "v, v" value that the Anthropic API rejects, so both case variants are now collapsed to one canonical lowercase header (same for anthropic-beta). (thanks @Delcado19)
• oauth(kiro): support Kiro IDC (organization) token import — when the ~/.aws/sso/cache token carries a clientIdHash, auto-import now reads the linked client registration file to obtain clientId/clientSecret, probes the Kiro IDE profile.json for profileArn (ARN region normalized to us-east-1 for the runtime gateway), and refreshes via the regional AWS OIDC endpoint instead of the social path; the import schema and modal forward these credentials so manual imports also work for IDC tokens. (thanks @enjoyer-hub)
• fix(translator): preserve client cache_control breakpoints when routing Claude-format requests (e.g. Claude Code) to Alibaba DashScope's OpenAI-compatible providers (alibaba / alibaba-cn). The Claude→OpenAI translation previously stripped the markers from the system and message text blocks, so DashScope's explicit caching never engaged and every request was a cache miss. Cache hints now survive when preservation is requested for caching-capable OpenAI-format providers. (thanks @sacrtap)
• fix(tts): resolve Gemini TTS models from catalog and add gemini-3.1-flash-tts-preview as the new default Vertex TTS model. (thanks @nguyenha935)
• fix(sse): don't cool down a healthy connection on a self-inflicted upstream timeout (504) — when OmniRoute's own deadline elapses (surfaced as TimeoutError/BodyTimeoutError → 504), the connection is no longer disabled/failed-over, so a slow-but-healthy provider isn't penalised for our timeout. Genuine upstream 5xx/429 still trigger cooldown; antigravity keeps its own policy. (thanks @costaeder)
• fix(translator): forward image tool_result blocks as image_url instead of stringifying base64. (thanks @alican532)
• fix(sse): robust Anthropic /v1/messages streaming — real ping keepalive + client-disconnect guard — slow first tokens on reasoning models could trip strict clients' idle-read watchdog; the route now keeps the stream warm with a real event: ping (Anthropic clients ignore SSE comments) from the very first frame, and a client disconnect (AbortError / controller-closed) no longer counts as a provider failure (no failover/cooldown). (thanks @costaeder)
• fix: preserve model hidden flags (isHidden) across model sync — replaceCustomModels pruned the compat-override list to the new custom-model ids, silently wiping the isHidden flag of eye-hidden SYNCED models on every periodic sync / import (all hidden models turned back on). The redundant cleanup is removed (per-model removal already handles its own compat cleanup), so eye-hidden models stay hidden across re-sync. (#5086 — thanks @herjarsa)
• fix(models): derive model-discovery config from the registry modelsUrl — providers absent from the hardcoded PROVIDER_MODELS_CONFIG but carrying a registry modelsUrl (e.g. MiniMax) now get an auto-derived Bearer /v1/models discovery config, so "discover models" works instead of returning nothing. (thanks @herjarsa)
• fix(compression): resolve worker + rule/filter assets via runtime anchors (standalone bundle) — the LLMLingua worker and the RTK rule/filter loaders relied on fileURLToPath(import.meta.url), which the standalone bundle freezes to the build-machine path, so the worker never spawned and rule/filter packs failed to resolve. They now anchor on process.cwd()/argv[1] (with pathToFileURL for the worker URL). (thanks @fulorgnas)
• fix(api): sanitize error responses on seven management routes (Rule #12 hardening) — cli-tools/backups, cli-tools/guide-settings/[toolId], logs/export, models/catalog, providers/test-batch, settings/import-json and usage/proxy-logs no longer return raw error.message; they wrap caught errors in sanitizeErrorMessage(...), and the routes are removed from the check-error-helper allowlist. (thanks @JxnLexn)
• fix(sse): keep output_text-only Responses bodies from being dropped/false-502'd — some upstreams return a shorthand Responses body whose answer is only in output_text with an empty output[]. sanitizeResponsesApiResponse discarded the text, so the response then tripped the malformed-200 guard. The sanitizer now synthesizes an output[] message item from a non-empty output_text (complements the Claude-native fix in #5108; both stem from #4942).
• fix(executors): preserve a lone caller-supplied Anthropic-Version header casing — the case-variant dedupe (#4846) unconditionally rewrote Anthropic-Version/Anthropic-Beta to lowercase even when only one variant was present, clobbering the caller's header. Dedupe now runs only when both case variants coexist (the actual undici-merge collision it was meant to fix).
• fix(responses): default text.format to { type: "text" } for openai-compatible responses providers — some Responses-compatible upstreams (e.g. LM Studio) reject a text object missing text.format with a 400 missing_required_parameter; the default executor now fills the Responses-API default before forwarding (guarded to openai-compatible-*responses*, never overwriting an existing format). (thanks @StevanusPangau)
• fix(translator): stop stripping client-provided reasoning_content for reasoning-replay providers — the #4849 agentic-context strip (which drops reasoning_content from tool-call assistant turns to avoid O(n²) token growth) ran unconditionally, so replay providers (DeepSeek V4, Kimi K2, Qwen-Thinking, etc.) lost the client's reasoning and the reasoning-replay cache then overwrote it with a stale cached value (and such upstreams 400 without the original reasoning). The strip now skips reasoning-replay targets while non-reasoning providers keep the O(n²) protection. (#5122)
• fix(providers): add MiniMax M3 & Nemotron 3 Ultra to the Cline catalog — the two models were missing from Cline's provider catalog and could not be selected; both are now registered. (#5136, closes #3321)
• fix(dashboard): key model-visibility toggle on the canonical providerId — the per-model visibility toggle keyed off a display id, so toggling a model on one provider alias could mis-target another; it now keys on the canonical providerId. (#5091 — thanks @Theadd)
• fix(diagnostics): recognize the Claude API format in detectMalformedNonStream — salvaged null-guard so a Claude-shaped non-streaming body is no longer misclassified. (#5141 — thanks @herjarsa / @diegosouzapw)
• fix(logging): track the final connection IDs in failover logs — failover log lines now record the connection that actually served (or last failed) the request, instead of only the first attempt. (#5016 — thanks @JxnLexn)
• fix(sse): ignore disconnect races during in-band stream error handling — a client disconnect that races with in-band upstream error handling no longer surfaces as a spurious provider failure. (#5007 — thanks @JxnLexn)
• fix(dashboard): surface the server error on handleToggleCombo failure — a failed combo toggle now shows the backend error instead of silently no-op'ing. (#5138 — thanks @KooshaPari / @diegosouzapw)
• fix(quota): track provider quota reset windows + enrich the Codex playground — observed quota reset windows are tracked and surfaced, and the Codex playground gains the enriched quota metadata. (#5141 — thanks @Witroch4 / @diegosouzapw)
• fix(sidebar): drop the orphan settings accent color — removed a dangling accent-color entry that broke typecheck:core. (#5142)
• fix(sse): preserve non-stream reasoning fields for compatible clients — non-streaming responses now keep the upstream reasoning fields (reasoning / reasoning_content and OpenRouter/Gemini reasoning_details) instead of stripping them in responseSanitizer, so clients that render reasoning on buffered responses no longer lose it. (#5155 — thanks @rdself)
• fix(i18n): add missing English UI labels — fills in untranslated English strings that were surfacing as raw keys in the dashboard. (#5153 — thanks @rdself)

🔒 Security

• fix(security): exact-host Anthropic baseUrl check — the Anthropic base-URL guard used a substring match that a crafted host could partially satisfy; it now requires an exact host match (resolves CodeQL js/incomplete-url-substring-sanitization alert #674). (#5130)

📝 Maintenance

• refactor(store): remove dead legacy store modules — salvaged cleanup of unused legacy store code. (#5138 — thanks @JxnLexn / @diegosouzapw)
• test(combo): deterministic routing-decision matrix for all 17 strategies — a deterministic E2E matrix pins the routing decision of every combo strategy. (#5146)
• chore: baseline reconciliations (complexity / file-size / cognitive), golden-snapshot + apikey-count alignment for new providers, orphan-test relocation, release base-red repairs, CHANGELOG i18n mirror sync, and an actions/cache 5→6 bump. (#5145, #5144, #5125, #5126, #5120, #5117, #5112)
• test: gated live smoke for combo strategies (in-process + VPS HTTP) and refreshed release expectations to match current code. (#5151, #5150 — thanks @KooshaPari / @diegosouzapw)

---

[3.8.37] — 2026-06-26

✨ New Features

• feat(providers): add DGrid AI gateway provider — OpenAI-compatible gateway at api.dgrid.ai/v1 (alias dgrid, API-key auth, passthrough models). Free router tier (10 RPM / 100 RPD); a $5 lifetime top-up raises limits to 20 RPM / 1,000 RPD. (#4931 — thanks @dgridOP)

• feat(providers): add Pioneer AI (Fastino Labs) provider — OpenAI-compatible chat completions at api.pioneer.ai/v1. Registered with alias pn, X-API-Key auth, and a catalog of 10 open-tier serverless models (Qwen3, Llama 3.1/3.2, Gemma 3, SmolLM3). Free $75 credits, no credit card required. Gated enterprise models (Claude/GPT/Gemini) require prior fine-tuning on the Pioneer platform and are intentionally excluded from the catalog. (#4909 — thanks @HikiNarou)

• feat(providers): add xAI Grok inbound translators and a thinking patcher — Grok requests are now translated on the inbound path and reasoning is normalized so Grok modes behave consistently across clients. (#4910 — thanks @mugnimaestra)

• feat(oauth): Codex bulk-import endpoint — POST /api/oauth/codex/import accepts multiple Codex OAuth credentials in one call for fast multi-account onboarding. (#4914 — thanks @beaaan)

• feat(embeddings): add a dimensions override field to embedding combos so an embedding combo can pin the output vector size per target. (#4913 — thanks @wenzetan)

• feat(sse): auto-promote successful combo model — a new opt-in comboAutoPromoteEnabled setting reorders a combo's persisted model list so that, when a combo model responds successfully, it is moved to position #1 for future requests. (#4852 — thanks @arssnndr)

• feat(sse): add toggleable tool-source diagnostics — an opt-in switch surfaces where each tool definition originated when debugging tool-routing issues. (#4856 — thanks @DuyPrX)

• feat(headroom): proxy lifecycle management + dashboard UI — start/stop/monitor a Headroom compression proxy from the dashboard, with Docker sidecar support. (#4649 — thanks @diegosouzapw / @carmelogunsroses)

• feat(sse): x-omniroute-strip-reasoning request header to drop reasoning_content from upstream responses (opt-in, preserving reasoning-aware clients). (#4678 — thanks @anuragg-saxenaa / @diegosouzapw)

• feat(cli): multi-model support for the Factory Droid CLI integration. (#4682 — thanks @anuragg-saxenaa / @diegosouzapw)

• feat(sse): parse Gemini CLI 429 retryDelay from the structured RetryInfo payload so cooldowns honor the upstream-provided backoff. (#4738 — thanks @NoxzRCW)

• feat(sse): add GPT-4 and GPT-4o mini to the GitHub Copilot provider catalog. (#4798, #4797 — thanks @decolua)

• feat(api): add the MiniMax-M3 pricing row (canonical + lowercase alias) so the new MiniMax default model gets accurate per-request cost accounting instead of falling back to a zero/default rate. (#4814 — thanks @octo-patch)

🔧 Bug Fixes

• fix(sse): dense, deterministic response.output ordering in response.completed — items are now sorted by their actual output_index (via a recorded-as-emitted accumulator + stable sort) instead of being rebuilt from unordered state dicts; normalizeOutputIndex replaces fragile parseInt calls for robust index coercion; superseded tool calls (replaced at the same index mid-stream) are excluded from the final output array. (#4906 — thanks @Marco9113)

• fix(sse): normalize Codex custom/freeform tools (apply_patch, type:"custom" with no parameters) to a { input: string } function schema instead of an empty schema — the empty schema made models invoke apply_patch with {}, breaking the Codex runtime which expects { input: string }. Also maps custom_tool_call / custom_tool_call_output input items and streams apply_patch tool calls via custom_tool_call_input.delta/.done events. (#4862 — thanks @nstung463)

• fix(sse): preserve the required array when translating Draft 2020-12 antigravity tool schemas (e.g. from OpenCode), stripping unsupported JSON Schema meta keywords while keeping mandatory arguments required so the model no longer calls tools without them. (#4843 — thanks @anuragg-saxenaa)

• fix(sse): Kiro tool-schema sanitizer — strip unsupported JSON-Schema keywords (anyOf/$ref/if-then, etc.) and hash-truncate tool names >64 chars before dispatch, mapping the streamed tool-call name back for the client, so Kiro no longer rejects tool calls with 400 "Improperly formed request". (#4847 — thanks @smarthomeblack)

• fix(sse): make the anthropic-version default-guard case-insensitive for anthropic-compatible-* providers, so a caller/operator-supplied Anthropic-Version (any casing) is no longer clobbered by a second lowercase anthropic-version: 2023-06-01 header. (#4823 — thanks @zakirkun)

• fix(db): validate HuggingFace API tokens via the whoami-v2 endpoint as a pure auth probe so fine-grained Inference-Provider tokens (valid even when model/task endpoints reject them) are no longer falsely marked invalid; only 401/403 means an invalid key, other non-OK statuses surface as transient upstream errors. (#4819 — thanks @Delcado19)

• fix(sse): reject the Anthropic-only [1m] context-1m suffix in buildKiroPayload before it reaches AWS Bedrock — Kiro is Bedrock-backed and cannot honor the beta, so a forwarded kr/*[1m] model id was malformed upstream; callers now get a clear error pointing them at a direct-Anthropic provider for 1M-context routing. (#4816 — thanks @Delcado19)

• fix(dashboard): align the Engine Combos editor engines with the API schema — the named-combos pipeline dropdown offered four engines (headroom, session-dedup, ccr, llmlingua) that PUT /api/context/combos/[id] rejects, so selecting one made the save return 400 while the UI swallowed the error. The dropdown is now sourced from a single canonical engine map shared with stackedPipelineStepSchema (parity guarded by a unit test), and the editor surfaces save errors plus empty-name/empty-pipeline validation instead of failing quietly. (#5062 — closes #4955)

• fix(sse): surface malformed HTTP-200 upstream responses instead of treating them as success, so combo fallback can trigger. (#4942 — thanks @haipham22)

• fix(antigravity): retry transient upstream failures rather than failing the request outright. (#4941 — thanks @Jordannst)

• fix(sse): exclude WS-bridge controller-closed errors from the provider circuit breaker so a client disconnect no longer trips the whole provider. (#4870 — closes #4602, thanks @huohua-dev)

• fix(sse): resolve custom combos by id and case-insensitive name. (#4869 — closes #4446, thanks @herjarsa)

• fix(sse): forward AI SDK image parts in the Responses translator. (#4859 — thanks @mugnimaestra)

• fix(sse): emit valid concatenable Kiro tool_calls.arguments deltas. (#4855 — thanks @wahyuzero)

• fix(sse): strip temperature for Claude models with extended thinking enabled (the upstream rejects it). (#4853 — thanks @noestelar)

• fix(sse): unwrap the Qoder HTTP-200 SSE error envelope so combo fallback can trigger. (#4850 — thanks @vianlearns)

• fix(sse): strip reasoning blobs from agentic context to prevent O(n²) token growth across multi-turn agent loops. (#4849 — thanks @GodrezJr2)

• fix(sse): close the reasoning block before message content in the Responses stream so clients render reasoning and answer in the right order. (#4848 — thanks @kwanLeeFrmVi)

• fix(config): sync the full SiliconFlow model list into the registry. (#4844 — thanks @letanphuc)

• fix(sse): strip Composer <｜final｜> sentinel markers that leaked after Composer reasoning. (#4842 — thanks @noestelar)

• fix(build): trace-include sql.js's sql-wasm.wasm in the standalone bundle so SQLite-WASM works in the packaged build. (#4839 — thanks @Delcado19)

• fix(cli): persist lazily-installed native runtime deps (better-sqlite3, systray2) to the shared runtime package.json with --save-exact instead of --no-save, so installing one no longer prunes the other as "extraneous" — fixing a "No SQLite driver available" failure after a --tray install. (#4841 — thanks @omartuhintvs)

• fix(sse): resolve bare model names to a connection's defaultModel before upstream calls. (#4825 — thanks @anuragg-saxenaa)

• fix(api): surface a Docker-localhost hint on provider-node validation connection errors. (#4822 — thanks @anuragg-saxenaa)

• fix(sse): strip Gemini built-in tools when functionDeclarations are present in the Antigravity envelope (the two are mutually exclusive upstream). (#4821 — thanks @vanszs)

• fix(sse): strip X-Stainless-* headers and normalize the SDK User-Agent for OpenAI-compatible endpoints. (#4820 — thanks @anuragg-saxenaa)

• fix(oauth): allow a per-connection refresh lead-time override via providerSpecificData.refreshLeadMs. (#4818 — thanks @anuragg-saxenaa)

• fix(dashboard): resolve passthrough model aliases by providerId in ModelSelectModal. (#4815 — thanks @anuragg-saxenaa)

• fix(sse): strip enumDescriptions from Antigravity tool schemas. (#4813, #4740 — thanks @anuragg-saxenaa)

• fix(dashboard): keep the desktop sidebar visible via an explicit CSS class. (#4812 — thanks @Delcado19)

• fix(sse): filter nameless hosted tools when converting Responses API to Chat format. (#4789 — upstream, thanks Владимир Акимов)

• fix(sse): stream-writer mock abort() now returns a Promise (test-stability fix). (#4788 — thanks @decolua)

• fix(sse): use the WorkOS auth-token shape for Cline. (#4787 — thanks @apeltekci)

• fix(api): fall back to the existing access token for any OAuth provider when a refresh fails. (#4786 — thanks @decolua)

• fix(sse): read Antigravity usage from the response.usageMetadata envelope. (#4785 — thanks @decolua)

• fix(oauth): verify Cursor installation on Linux before auto-import. (#4770 — upstream, thanks Ibrahim Ryan)

• fix(cli): fall back to the default data dir when DATA_DIR is not writable. (#4767 — upstream, thanks Thiên Toán)

• fix(sse): json_schema fallback for OpenAI-compatible providers that don't support structured outputs. (#4766 — thanks @mustafabozkaya)

• fix(cli): verify launchd registration and skip self-SIGTERM in macOS autostart. (#4765 — thanks @ntdung6868)

• fix(sse): finalize the tool_calls finish_reason on early stream end in the OpenAI Responses translator. (#4764 — thanks @decolua)

• fix(sse): gate Kiro image attachments behind a Claude-capability check. (#4763 — thanks @decolua)

• fix(sse): track Ollama streaming usage from raw NDJSON chunks. (#4754 — thanks @fresent)

• fix(sse): include low-level cause details in formatProviderError. (#4741 — thanks @decolua)

• fix(executors): anthropic-compatible-* gateways now get a Bearer token alongside x-api-key. (#4729 — thanks @hodtien)

• fix(translator): strip the x-anthropic-billing-header in the claude-to-openai path. (#4728 — thanks @weimaozhen)

• fix(translator): preserve reasoning_effort for non-Copilot Responses clients. (#4688 — thanks @ryanngit / @diegosouzapw)

• fix(codex): treat an OAuth 401 as an unrecoverable refresh failure (stop retrying a dead token). (#4686 — thanks @sacwooky / @diegosouzapw)

• fix(translator): coerce tool descriptions to strings in OpenAI normalization. (#4675 — thanks @East-rayyy / @diegosouzapw)

• fix(dashboard): stop double-masking an already-masked API key in the list view (E2E 3/9 regression). (#4671 — thanks @diegosouzapw)

• fix(combo): flatten Anthropic tool messages + tool history to prevent an upstream 503. (#4648 — thanks @warelik / @diegosouzapw)

• fix(providers): require a Default Model in the compatible-provider API-key setup flow. (#4641 — thanks @arden1601)

🔒 Security

• fix(auth): only trust forwarding headers (X-Forwarded-For / X-Real-IP) from loopback TCP peers, so a non-loopback client can't spoof its origin to bypass local-only route guards. (#4689 — thanks @Jordannst / @diegosouzapw)

• fix(sse): redact the API key from the AUTH debug log in the chat handler. (#4858 — thanks @sacwooky)

• fix(oauth): classify /api/oauth/cursor/auto-import as a local-only route in the route guard, so the loopback-enforced process-spawning endpoint can't be reached through a tunneled/leaked JWT (Hard Rule #17). (#5070 — thanks @diegosouzapw)

📝 Maintenance

• chore(ci): harden the release flow — decouple the Quality Ratchet from coverage-shard flakes (if: !cancelled() + --allow-missing), add fast-path drift gates (check:complexity, check:cognitive-complexity, check:pack-policy, check:build-scope), and raise the default build heap to 8 GB. (#5054 — thanks @diegosouzapw)

• docs(routing): sync the combo strategy docs for Fusion (17 strategies). (#5067 — thanks @diegosouzapw)

• test(sse): golden-lock the provider.ts translate-path across all providers. (#4734 — thanks @diegosouzapw / @decolua)

• docs(env): document HEADROOM_URL in .env.example + ENVIRONMENT.md. (thanks @diegosouzapw)

• chore(quality): rebaseline the file-size ratchet across the rc17 PR-batch levas (leva2/leva3/leva4) to absorb cycle drift. (thanks @diegosouzapw)

---

[3.8.36] — 2026-06-25

✨ New Features

Quota-Share system

• feat(quota): introduce a dedicated quota-share combo strategy (Fase 3 #9) — Deficit Round Robin scheduling with per-model in-flight gating (P2C), automatic DB migration that promotes existing qtSd/* combos, and per-policy gating so invalid allocations cannot bleed allow to unintended connections. (#4939, #4901)
• feat(quota): multi-window usage buckets, per-(key,model) caps, and session stickiness — connections now track consumption across 5 h, 7 d, and per-model windows; quota_allocation_model_caps enforces per-key/model limits; session stickiness preserves prompt-cache integrity across turns. (#4928, #4927, #4929)
• feat(quota): headroom strategy + proactive saturation — new headroom combo strategy selects connections by available quota margin; universal proactive saturation via upstream token-usage response headers; real Claude quota saturation sourced from /api/oauth/usage. (#4908, #4907, #4885)
• feat(quota): concurrency control + cooldown-wait (Fase 2.1) — max_concurrent is enforced at dispatch time; quota-share combos queue concurrent requests with a short cooldown-wait and re-dispatch on slot availability (Variant A); a cron heal proactively restores connections after their window resets. (#4965, #4970, #4967, #4900)

Combo routing

• feat(combo): task-aware routing strategy — routes requests to the best-fit connection based on task-type metadata, enabling per-task provider specialization within a combo. (#4945)
• feat(combo): Fusion strategy (16th strategy) — fan out to a configurable panel of models in parallel, then synthesize results through a judge model. (#4652)
• feat(combos): add an editable per-combo description field. The routing-combo form now has a Description input, persisted in the combo data blob via /api/combos (POST/PUT) and round-tripped through GET — no new DB column. (#5005)
• feat(routing): honor X-Route-Model request header to override body.model, enabling per-request model switching without modifying the request body. (#4863 — thanks @costaeder)

Providers & models

• feat(providers): update volcengine-ark model list, adding DeepSeek-V4-Flash and DeepSeek-V4-Pro. (#4905 — thanks @kenlin8827)
• feat(provider): add CodeBuddy CN (copilot.tencent.com) — full OAuth + executor + model catalog stack. (#4664)
• feat(opencode-go): advertise glm-5.2 and kimi-k2.7-code to align with official Go endpoints. (#4711)
• feat(sse): add Google Flow video-generation provider. (#4769)
• feat(api/v1): include alias-backed models in the /v1/models listing. (#4630)

Proxy pool

• feat(proxy-pool): Cloudflare Workers proxy deployer + pool integration — deploy Cloudflare Workers relays directly from the dashboard and register them in the proxy pool. (#4640)
• feat(proxy-pool): Deno Deploy relays + group action buttons — deploy Deno Deploy relay workers and manage proxy groups with new bulk-action controls. (#4643)

Compression & infrastructure

• feat(compression): Kiro/CodeWhisperer tool-result compression engine — dedicated compressor for Kiro/CodeWhisperer tool outputs integrated into the streaming pipeline. (#4635)
• feat(endpoint): per-endpoint custom system prompt injection. A toggle + text field in the Endpoint settings card lets users inject a custom system prompt into every model request, applied via the existing system-prompt engine. Stored in settings DB. (#5022 — thanks @whale9820)
• feat(live-ws): allow non-loopback clients via LIVE_WS_ALLOWED_HOSTS env var, enabling multi-host setups to access the live WebSocket API. (#4877 — thanks @KooshaPari)
• feat(db): track API endpoint dimension on usage_history for per-endpoint cost and usage analytics. (#4676)

---

🔧 Bug Fixes

Translator

• fix(translator): regroup parallel tool results to be adjacent to their originating assistant turn, fixing tool-message ordering for providers that require strict interleaving. (#4882)
• fix(translator): preserve literal empty-string tool arguments in OpenAI-to-Claude streaming — they were previously dropped, causing tool calls to arrive with missing parameters. (#4959)
• fix(translator): normalize tools to Anthropic-native shape for non-Anthropic providers, ensuring tool definitions pass validation regardless of the format at the call site. (#4650)
• fix(translator): provider thinking compatibility — correct thinking-block serialization for DeepSeek and Gemini providers. (#4946)
• fix(translator): emit </think> close marker for Anthropic thinking blocks, fixing truncated reasoning output in streamed responses. (#4633)
• fix(translator): normalize developer role to system for OpenAI-format providers. (#4625)
• fix(translator): strip top-level client_metadata on the OpenAI passthrough path (port from 9router#1157). (#4624)
• fix(translator): replay reasoning_content on plain Xiaomi MiMo turns (port from 9router#1321). (#4639)

Copilot / GitHub executor

• fix(copilot): never route Gemini/Claude model variants to the /responses endpoint — these models require the chat-completions path only. (#4627)
• fix(github): route Copilot Codex models to /responses (port from 9router#102). (#4626)
• fix(copilot,antigravity): cap maxOutputTokens at 16384 to stop "Invalid Argument" 400 errors on high-token requests. (#4636)
• fix(codex): drop non-standard codex.* streaming events that break responses.stream consumers. (#4715 — thanks @jeffer1312)

Claude / Anthropic

• fix(claude): omit adaptive_thinking and output_config.effort for Haiku model variants, which reject those parameters. (#4661)
• fix(claude): skip mcp__ tool-name cloaking and guard against missing connectionId to prevent crashes on Claude-native MCP tool calls. (#4861 — thanks @costaeder)
• fix(claude-oauth): respect 429 backoff headers on the Claude OAuth usage endpoint to reduce polling spam during quota checks. (#4655)

Routing & SSE

• fix(sse): fail over on 400 responses that carry rate-limit text in the body, not just on canonical 429 status codes. (#4986)
• fix(sse): honor per-account proxy and fingerprint-rotation settings in the opencode executor. (#4989)
• fix(sse): soft-penalize exhausted providers in auto-combo scoring instead of hard-excluding them, improving fallback resilience. (#4990)
• fix(sse): drop the CCP pin when the pinned provider is durably unhealthy, with anti-flap logic to prevent oscillation. (#4864 — thanks @costaeder)
• fix(combo): fetch models dynamically from custom provider endpoints instead of relying on a static list. (#4860)
• fix(combo): propagate the selected connection ID to fallback error responses so model lockout applies to the correct connection rather than the wrong fallback target. (#4809 — thanks @Chewji9875)
• fix(sse): skip third-party tool-name cloaking for Anthropic-native server tools to prevent naming conflicts. (#4808 — thanks @NomenAK)

Quota

• fix(quota): quota-exclusive qtSd/* connections now appear in /v1/models listings; EPSILON-threshold check no longer falsely blocks under-budget allocations. (#4830)
• fix(quota): migration 107 correctly activates the quota-share strategy on existing qtSd/* combos. (#4962)

API / responses

• fix(api): parse the /v1/responses request body once instead of 3–4 times on the hot path, reducing per-request overhead. (#4958)
• fix(api): evict stale in-memory rate-limit windows to stop a slow heap leak on long-running instances. (#4957)
• fix(api): require authentication on the compression run-telemetry endpoint; document OMNIROUTE_EVAL_CREDENTIALS env var. (#4796)
• fix(api): stop GET /api/system/env/repair returning HTTP 500 on packaged installs (it broke the onboarding wizard). createRequire(import.meta.url) ran at module top-level; once webpack bundles the route into the standalone build, import.meta.url is frozen to the build-machine path and createRequire throws during evaluation, so the whole route failed to load. createRequire is now resolved lazily inside the guarded better-sqlite3 block, root-dir resolution falls back to process.cwd(), and the route passes an explicit rootDir. (#5028)

Dashboard

• fix(dashboard): show custom provider given-name instead of internal id across dashboard pages — cache, combo health, compression analytics, cost overview, health/autopilot, provider stats, route explainability, provider utilization, runtime. Adds shared resolveProviderName resolver and useProviderNodeMap hook. (#4603)
• fix(dashboard): on OAuth providers (e.g. GLM Coding), "Test all models" with auto-hide-failed now switches the model list to the "visible" filter after the run, so just-hidden failed models actually disappear on-screen — parity with the passthrough-provider path (#3610). Previously they were hidden in the DB but stayed visible under the "All" filter, so it looked like nothing was hidden. (#4887)
• fix(dashboard): restore the home-page provider topology card that was hidden by a default state change in #4596. (#4963)
• fix(dashboard): proxy-pool success gating, sync-timestamp persistence, and opt-in Redis backend. (#4988)
• fix(dashboard): show custom vision models in the LLM selector dropdown. (#4653)

Providers

• fix(pollinations): stop forcing jsonMode on every request. Pollinations treats jsonMode=true as "the model MUST return JSON" and rejects (HTTP 400 "messages must contain the word 'json'") any normal chat request whose messages don't mention "json", so all non-JSON chat was broken. jsonMode is now only enabled when the caller actually requests JSON output (response_format.type of json_object or json_schema). (#3981)
• fix(antigravity): default safetySettings to all-OFF for parity with the native Gemini paths. The Antigravity (Google Cloud Code) request builder set safetySettings: undefined, which JSON.stringify drops — so no safety settings reached Google and its server-side defaults false-flagged benign technical prompts as prohibited_content (HTTP 200 + blocked body, which combo failover treats as terminal). Now honors a caller-supplied value and otherwise defaults to DEFAULT_SAFETY_SETTINGS, matching the claude-to-gemini / openai-to-gemini paths. (#5003)
• fix(antigravity): exclude the standard Gemini rate-limit message from quota-exhaustion keyword matching to prevent false-positive saturation flags. (#4810 — thanks @Chewji9875)
• fix(chatgpt-web): map the advertised gpt-5.5, gpt-5.5-pro, gpt-5.4-pro and gpt-5.2-pro catalog ids to their dash-form ChatGPT backend slugs. They were missing from MODEL_MAP, so the executor sent the dot-form id verbatim, which the ChatGPT backend silently ignored and served the default Plus model instead of the requested one. Adds a drift guard asserting no advertised dot-form id reaches the backend verbatim. (#4665)
• fix(gemini): preserve the pattern field in the Antigravity tool schema sanitizer to avoid stripping valid regex constraints from tool definitions. (#4651)
• fix(opencode): preserve DeepSeek reasoning content in streamed responses. (#4631)
• fix(perplexity): validate API keys via the /v1/models endpoint instead of issuing a full chat request. (#4654)
• fix(qoder): exchange PAT for jt-* job token before initiating Cosy chat, fixing auth failures after the Qoder credential format change. (#4884)
• fix(executors): strip parameters unsupported by the target provider/model to prevent 400 Invalid parameter errors on strict endpoints. (#4658)
• fix(executors): preserve literal reasoning_effort: "max" for Ollama Cloud instead of normalizing to xhigh. Ollama Cloud accepts high|medium|low|max|none and rejects xhigh (invalid reasoning value: 'xhigh'); OpenRouter DeepSeek max→xhigh normalization is unaffected. (#4993 — thanks @Thinkscape)
• fix(headroom): translate openai-responses input through OpenAI for external compression. adaptBodyForCompression now serialises function_call_output items whose output field is a JSON object (not a string) so compression engines can process the content — previously those items were excluded from compression because hasTextContent() returned false for object values. (#5023 — thanks @anki1kr)
• fix(proxy): fan out direct dispatcher streams to all registered proxy endpoints. (#4803 — thanks @makcimbx)

Compression

• fix(compression): eliminate ReDoS in the math_inline preservation regex — the previous pattern could catastrophically backtrack on untrusted input. (#4838)
• fix(compression): stop RTK over-truncating file-read tool results — RTK now respects the full content length for file-read outputs. (#4987)

Build / CLI / infrastructure

• fix(build): drop @omniroute/open-sse from optimizePackageImports to fix the Next.js build OOM crash. (#4968)
• fix(cli): SIGKILL the systray child PID before closing the IPC channel to prevent macOS NSStatusItem orphan processes. (#4732)
• fix(cli): bump better-sqlite3 runtime pin to 12.10.1 for Node 26 compatibility. (#4685)
• fix(cli): harden the systray2 tray runtime (port from 9router#1080). (#4628)
• fix(cli-tools): tolerate JSONC (comments and trailing commas) in tool settings files. (#4659)
• fix(install): make the transformers dependency optional so CUDA-host installs that lack Python bindings succeed. (#4807 — thanks @megamen32)
• fix(db): correct storage tuning settings to prevent WAL runaway on high-write workloads. (#4834 — thanks @rdself)
• fix(image): prevent compatible nodes from shadowing provider aliases in the image routing table. (#4656)

Plugin

• fix(plugin): opencode auth.json dual-key fallback for the auto-prefix migration. The config hook now looks up both the prefixed (opencode-omniroute) and bare (omniroute) keys, so users who authenticated before the opencode- prefix landed no longer need to re-auth. (#5027 — thanks @herjarsa)

---

🔒 Security

• fix(security): block SSRF allowlist bypass via x-relay-path header manipulation on Deno/Vercel relays. (#4899)
• fix(security): pin image-fetch DNS resolution to prevent SSRF DNS-rebinding attacks (GHSA-cmhj-wh2f-9cgx). (#4634)
• fix(security): do not trust the loopback socket as local-only when the server is behind a reverse proxy, closing a potential auth bypass path. (#4632)
• fix(security): validate the Kiro region parameter to prevent SSRF via crafted region strings (GHSA-6mwv-4mrm-5p3m). (#4629)
• fix(copilot): replace execSync shell interpolation with execFile in the runOmniRouteCli tool to prevent command injection. The user-supplied command is now split into an argv array and passed to execFile (no shell), so shell metacharacters are treated as literal text; error output is routed through sanitizeErrorMessage(). (#5024 — thanks @hamsa0x7)

---

📝 Maintenance

God-file decomposition (continued, #3501)

• refactor(chatCore): extracted 12 focused helpers from chatCore.ts covering the streaming pipeline (assembleStreamingPipeline), cache-store logic (storeStreamingSemanticCacheResponse, storeSemanticCacheResponse), response headers (assembleStreamingResponseHeaders, buildNonStreamingResponseHeaders), JSON→SSE bridge (maybeConvertJsonBodyToSse), guardrail context (buildPostCallGuardrailContext), usage buffer (applyClientUsageBuffer), plugin hook (runPluginOnRequestHook), analytics (writeCompressionAnalytics, emitOutputStyleTelemetry), and compression predicates/settings (resolveCompressionSettings, et al.). (#4811–#4837)
• refactor(sse/db/api): continued decomposition of services/usage.ts (extracted quota-core, scalar/format helpers, Antigravity/GLM/MiniMax usage families), db/core.ts (schema-column reconciliation, snake↔camel column-mapping), db/apiKeys.ts (row-parsers, model-permission matching), and validation.ts (URL/headers/transport leaf layer, web-cookie/Meta-AI validators, enterprise-cloud + probe, audio/speech/apikey, search/embedding/rerank, and OpenAI/Anthropic format validators). (#4921–#4956)
• refactor(pricing/providers): decomposed pricing.ts into shared tiers + partitioned DEFAULT_PRICING modules, and split the providers.ts catalog into semantic data modules organized by provider family. (#4917, #4918)
• refactor(open-sse): extract safeParseJSON utility and dedup tryParseJSON call sites; extract and dedup the fallback tool_call ID generation helper. (#4735, #4736)

Quality & CI

• chore(quality): release base-red reconciliation + ratchet rebaselines — file-size, env-doc, and catalog baseline updates across multiple gates. (#4630, #4879, #4886, #4915, #4961, #4973)
• ci(quality): shift heavy validation gates to the PR→release merge fast-path to accelerate the release cycle. (#4857)
• fix(ci): include coverage/lcov.info in the coverage-report artifact so SonarQube can consume it. (#4670)
• fix(test): validate Anthropic-compatible connections via POST /v1/messages for accurate connectivity testing. (#4657)

Docs

• docs(resilience): document Quota-Share Concurrency Control — max_concurrent enforcement, serialization behavior, and cooldown-wait semantics. (#4980)
• docs(perf): add per-endpoint p50/p95/p99 latency and cost budgets reference. (#4867 — thanks @KooshaPari)
• docs(ops): add canonical incident response runbook. (#4868 — thanks @KooshaPari)
• docs(ops): document the release-green family — green-prs, check:release-green, babysit, and nightly gate workflows. (#4679)
• docs(agentbridge): document Electron NODE_EXTRA_CA_CERTS, real model IDs, and identity caveat for agent bridge integrations. (#4718)
• docs: clarify Kiro provides ~50 credits/month per account, not unlimited. (#4690)

Misc

• chore(claude,codex): bump pinned CLI identities — Claude 2.1.158 → 2.1.187, Codex 0.132.0 → 0.142.0. (#4883)
• chore(dashboard): rename Qoder display label from "Qoder AI" to "Qoder". (#4733)

---

[3.8.35] — 2026-06-23

✨ New Features

• Adaptive context compression (Phase 4): a four-layer compression upgrade landed across stacked PRs — an Output Styles registry (terse-prose / less-code / terse-cjk) (#4694 — thanks @diegosouzapw), an opt-in SLM ultra tier (two-tier LLMLingua with heuristic fallback) (#4707 — thanks @diegosouzapw), a context-budget adaptive dial (reserve-output ladder + floor) (#4716 — thanks @diegosouzapw), and an offline evaluation harness (PII-gated corpus, self-test judge, gold-grader, real-pipeline runner behind a ModelClient seam) (#4720 — thanks @diegosouzapw). All four layers share a single CompressionRunTelemetry contract.
• Redoc-rendered API docs: a consolidated OpenAPI spec now lives at docs/openapi.yaml and is served as interactive Redoc documentation at /api/docs. (#4781 — thanks @KooshaPari / @diegosouzapw)

🔧 Bug Fixes

• db-backups: make the database-import size cap configurable via OMNIROUTE_DB_IMPORT_MAX_MB (default 100 MB, 4 GB ceiling) so backups larger than 100 MB can be restored; error message now points to the env var and to VACUUM (#4757 — closes #4719, thanks @diegosouzapw).
• Onboarding: add the missing onboarding.tiers step-title translation so the setup wizard no longer crashes with MISSING_MESSAGE: onboarding.tiers (#4755 — closes #4698, thanks @diegosouzapw).
• deepseek-web: fold role:"tool" results into the single-prompt transcript (messagesToPrompt) so tool outputs reach the model instead of being silently dropped when a follow-up turn omits the tools[] array (#4756 — closes #4712, thanks @diegosouzapw).
• Dashboard: remove the dead, unconditional useLiveRequests() call from HomePageClient.tsx — it crashed the /home page in production builds with ReferenceError: useLiveRequests is not defined (#4759, #4745) and opened the live-dashboard WebSocket even when Provider Topology was hidden (#4596). The live feed remains owned by the settings-gated HomeProviderTopologySection (#4761 — thanks @diegosouzapw).
• Providers dashboard: dedupe provider nodes by id when adding a compatible provider (upsertProviderNodeById) so the same provider can no longer appear twice and no-op adds don't invalidate the compatible-provider memo (#4768 — closes #4746, thanks @diegosouzapw).
• Storage VACUUM: the scheduled VACUUM job now follows the Storage page settings (scheduledVacuum / vacuumHour) as the single source of truth; the legacy env-flag control path was removed (#4726 — thanks @rdself).
• Tiers: no-auth providers are now counted as free, and the free-tier filter returns an empty set instead of falling through to every provider (#4753 — thanks @megamen32 / @diegosouzapw).
• Combos: auto-promote zeroLatencyOptimizationsEnabled so legacy configs (pre-3.8.33 fallbackCompressionMode="lite") round-trip cleanly on the first GUI edit (#4774 — thanks @KooshaPari / @diegosouzapw).

📝 Maintenance

• chatCore (#3501): continued the incremental decomposition of executeProviderRequest and the streaming/non-streaming hooks into pure leaf modules — top-level helpers + 6 pure leaves (#4571), resolveExecutorWithProxy + getExecutionCredentials (#4646), Claude message transforms (#4708), persistAttemptLogs (#4717), stageTrace + compressionUsageReceipt (#4721), prepareUpstreamBody (#4730), parse + non-streaming usage-stats (#4762), recordContextEditingTelemetryHook (#4779), scheduleQuotaShareConsumption (#4780), emitRequestGamificationEvent (#4776), runPluginOnResponseHook (#4782), scheduleStreamingQuotaShareConsumption (#4784), recordCompressionCacheStats (#4792), writeCavemanOutputAnalytics (#4794), recordStreamingUsageStats (#4791), and recordStreamingCost (#4790). (thanks @diegosouzapw)
• Quality: expand check:release-green to reproduce the full release-PR gate set locally (#4758 — thanks @diegosouzapw).
• db: re-export compressionRunTelemetry from localDb to satisfy the db-rules gate (#4775 — thanks @diegosouzapw).
• Security docs: add a canonical STRIDE-based threat model (#4783 — thanks @KooshaPari).
• Tests: add a smoke test for the home-client dashboard (#4793 — thanks @JxnLexn).
• Docs: credit ponytail and OmniCompress in the README inspiring-projects list and restore the check:env-doc-sync release-green by exempting the harness-only OMNIROUTE_EVAL_CREDENTIALS var (#4799 — thanks @diegosouzapw); declare the Phase 4 compression layers in the README + GUIDE (#4801 — thanks @diegosouzapw).
• Quality: trim combo-config.test.ts comments back under the file-size cap (follow-up to #4774) (#4800 — thanks @diegosouzapw).

---

[3.8.34] — 2026-06-23

✨ New Features

• feat(executors): Microsoft 365 Copilot pure framing + connection helpers — adds the request/response framing and connection helpers to support m365.cloud.microsoft/chat for individual M365 plans. (#4696 — thanks @skyzea1 / @diegosouzapw)
• feat(compression): per-request x-omniroute-compression header (Phase 3) — a request header now overrides the compression plan with the highest precedence (request-header > routing > profile > auto-trigger > Default > off), accepting off / default / engine:<id> / <combo>. The response echoes X-OmniRoute-Compression: <mode>; source=<source>. (#4645 — thanks @diegosouzapw)
• feat(audio): MiniMax T2A v2 TTS dispatch in audioSpeech — adds MiniMax text-to-speech dispatch (port of upstream #1043). (#4553 — thanks @diegosouzapw)
• feat(opencode): OpenCode Go DeepSeek reasoning variants — registers the Go DeepSeek reasoning model variants. (#4647 — thanks @DevEstacion)
• feat(quota): quota scraping for OpenCode Go and Ollama Cloud — surfaces quota windows for the OpenCode Go and Ollama Cloud providers. (#4642 — thanks @JxnLexn)
• feat(settings): expose stream recovery feature flags — surfaces the stream-recovery toggles in settings. (#4586 — thanks @rdself)
• feat(providers): optional model ID for custom API-key validation — custom API-key connection tests can now specify the model ID used to validate the key. (#4555 — thanks @diegosouzapw)

🐛 Fixed

• fix(db): scheduled cleanup actually runs + queries target the real tables (DB-bloat / OOM) — runAutoCleanup was never scheduled, so retention cleanup never executed and tables (compression_analytics, usage_history, …) grew unbounded into multi-GB SQLite files driving high RSS. Worse, several cleanup queries referenced wrong table/column names (call_logs.created_at→timestamp, compression_analytics.created_at→timestamp, mcp_audit_log→mcp_tool_audit, a2a_events→a2a_task_events, memory_entries→memories), so even a manual run silently no-op'd or errored. Fixed the five queries to match the real schema, added cleanupProxyLogs, and wired a startCleanupScheduler (startup + every 6h, VACUUM after deletes) into server-init alongside the existing budget-reset and reasoning-cache jobs. (#4691, extracted from #4428 — thanks @oyi77 / @diegosouzapw)
• fix(routing): include all noAuth models in auto-combos + add reka-flash + best-free template — noAuth provider models are no longer skipped when building auto-combos, reka-flash is registered, and a best-free combo template is added. (#4621 — thanks @oyi77)
• fix: noAuth provider validation + Kimi executor routing — corrects noAuth provider membership checks and removes a mis-routed Kimi alias. (closes #4620) (#4699 — thanks @oyi77)
• fix(executors): Firecrawl web_fetch 500 with include_metadata=true — fixes a crash when Firecrawl web_fetch is invoked with metadata extraction enabled. (#4692 — thanks @ponkcore)
• fix(proxy): apply pipelining:0 + connections cap to the direct dispatcher — same-provider concurrent requests no longer serialize behind a long/streaming request on the direct path. (#4684 — thanks @jeffer1312 / @diegosouzapw)
• fix(telemetry): back off live-WS event forwarding when the sidecar is unreachable — stops repeatedly attempting to connect to LIVE_WS_PORT when live monitoring is not configured. (#4687 — thanks @FikFikk / @diegosouzapw)
• fix(api): serve GET /v1/models/{model} as JSON, not the HTML dashboard — the per-model endpoint (IDs with slashes via a catch-all route) now returns JSON, unbreaking Claude Code. (#4677 — thanks @papajo / @diegosouzapw)
• fix(executors): robust deepseek-web tool-call parsing and agentic context retention — hardens DeepSeek-web tool-call parsing and preserves agentic context across turns. (#4644 — thanks @BugsBag)
• fix(cli): authenticate omniroute logs and honor the active context — the logs command now authenticates and respects the active context. (#4638 — thanks @Rahulsharma0810)
• fix(stream): estimate input tokens when upstream reports prompt_tokens=0 — input token usage is estimated when the upstream omits it. (#4615 — thanks @adivekar-utexas)
• fix(plugin): auto-prefix providerId with opencode- for OpenCode 1.17.8+ native gate — adapts provider IDs to the OpenCode 1.17.8+ native provider gate. (#4527 — thanks @herjarsa)
• fix(catalog): shorten no-thinking gateway prefix to no-think/ — renames the no-thinking gateway prefix. (#4525 — thanks @Rahulsharma0810)
• fix(models): unknown max output limits no longer default to 8192 — models without synced/registry/static maxOutputTokens resolve the limit as unknown instead of a generic 8192 cap; clamping/injection only happens when a real cap is known. (#4584 — thanks @rdself)
• fix(resilience): respect upstream retry-hint toggle — honors the configured toggle for upstream retry hints. (#4585 — thanks @rdself)
• fix(providers): show revealed connection API keys — fixes revealing stored connection API keys in the UI. (#4583 — thanks @rdself)
• fix(logs): make active-request stale sweep configurable — exposes the stale-request sweep interval as a setting. (#4599 — thanks @rdself)
• fix(resilience): retain provider cooldowns for the configured max window — cooldowns persist for the configured maximum window. (#4588 — thanks @KooshaPari)
• fix(resilience): reject invalid provider cooldown bounds — validates cooldown bound configuration. (#4589 — thanks @KooshaPari)
• fix(combo): preserve production combo metrics on shadow eviction — shadow eviction no longer drops production combo metrics. (#4590 — thanks @KooshaPari)
• fix(combo): exclude exhausted connections from auto scoring — exhausted connections are no longer scored as auto-combo candidates. (#4592 — thanks @KooshaPari)
• fix(relay): apply IP rate limit to the Bifrost sidecar — extends IP rate limiting to the Bifrost relay sidecar. (#4593 — thanks @KooshaPari)
• fix(bifrost): finalize SSE relay usage after stream — finalizes relay usage accounting once the SSE stream completes. (#4612 — thanks @KooshaPari)
• fix(quota): expose Bailian quota windows — surfaces Bailian provider quota windows. (#4610 — thanks @KooshaPari)
• fix(dashboard): gate home topology live-WS networking behind widget visibility — the home dashboard no longer starts topology polling / live sockets when topology is hidden. (#4618, #4606 — thanks @KooshaPari)
• fix(dashboard): isolate the quota widget refresh clock — the quota widget refresh no longer drives unrelated re-renders. (#4611 — thanks @KooshaPari)
• fix(dashboard): memoize compatible provider groups — avoids recomputing compatible provider groups on every render. (#4613 — thanks @KooshaPari)
• fix(cli): align omniroute data dir and env loading with the runtime — the CLI's data-dir/env loading no longer drifts from the server runtime configuration. (#4619, #4607 — thanks @KooshaPari)
• fix(api/settings): prevent cached /api/settings responses — disables caching on the settings endpoint (port from 9router#951). (#4566 — thanks @diegosouzapw)
• fix(executors): strip temperature for the GitHub Copilot gpt-5.4 family — removes the unsupported temperature param for Copilot gpt-5.4 models (port from 9router#612). (#4564 — thanks @diegosouzapw)
• fix(dashboard): keep play_arrow spinning on provider "Test All" buttons — fixes the spinner state on the provider test buttons (port from 9router#715). (#4563 — thanks @diegosouzapw)
• fix(dashboard): surface manual config CTA when Open Claw CLI auto-detect fails — shows a manual-config call-to-action on the Open Claw CLI card when auto-detection fails. (#4562 — thanks @diegosouzapw)
• fix(oauth): update Qwen OAuth URLs from chat.qwen.ai to qwen.ai — refreshes the Qwen OAuth endpoints (port of decolua/9router#683). (#4561 — thanks @diegosouzapw)

📝 Maintenance

• refactor(imageGeneration): extract 8 provider families to co-located files — splits the image-generation module into eight co-located per-provider files with no behavioral change. (#4609 — thanks @KooshaPari)
• deps: bump production + development groups; migrate js-yaml to v5 (ESM) — dependency bumps plus a js-yaml v4→v5 migration to the ESM-only namespace import. (#4697 — thanks @diegosouzapw)
• chore(quality): release-green pre-flight validator + nightly signal — new npm run check:release-green (scripts/quality/validate-release-green.mjs) reproduces the release-equivalent validation (full unit + vitest + ratchets + typecheck + lint, optional --with-build package-artifact) against the current working tree and classifies each red as HARD (real defect) vs DRIFT (ratchet, rebaselined at release) — purely diagnostic, never blocking contributors. A new nightly-release-green workflow runs it on the active release branch and opens/updates a tracking issue on hard failures. Closes the gap where the full gate (ci.yml) only ran on the release PR, so reds accrued silently on release/** and surfaced in layers at release time. (#4622 — thanks @diegosouzapw)
• chore(quality): reconcile file-size baseline for #4644 (deepseek-web.ts 1117→1125) — rebaselines the file-size gate after the deepseek-web hardening. (#4695 — thanks @diegosouzapw)

---

[3.8.33] — TBD

See English CHANGELOG for v3.8.33 details.

[3.8.32] — TBD

See English CHANGELOG for v3.8.32 details.

---

[3.8.31] — 2026-06-20

✨ New Features

• perf(dashboard): combos UI leaf-split, Next.js config tuning, 1-click Redis & Bifrost sidecar — delivers four of the five performance/UX tracks from the #3932 thread: the combos dashboard page is split into focused leaf components (smaller bundles, faster reloads), next.config is tuned for the standalone build, Redis can be provisioned in one click, and a Bifrost sidecar option is wired in. (The fifth track — chatLogHelpers extraction — was already covered upstream and dropped.) (#4381 — thanks @KooshaPari)

🐛 Fixed

• fix(embeddings): NVIDIA NIM asymmetric embedding models inject the required input_type — NVIDIA NIM asymmetric embedders (e.g. nvidia/nv-embedqa-e5-v5) reject requests without an input_type parameter with 400 "'input_type' parameter is required", but OmniRoute only forwarded input_type when the client supplied it — so callers (and OpenAI-style SDKs that don't emit the field) got a hard failure. The embedding registry now carries a model-level default (input_type: "query") for the asymmetric NVIDIA model, and the embeddings handler injects a model's default params into the upstream body only when the client didn't already send them — a client-supplied input_type (e.g. "passage") is respected unchanged, and symmetric models that carry no default are unaffected. (#4341 — thanks @hydraromania)
• fix(api): migrate the deprecated Codex [features].codex_hooks flag to [features].hooks — Codex renamed the codex_hooks feature flag to hooks; recent Codex CLI versions ignore the old key and print a deprecation notice. When OmniRoute rewrites an existing ~/.codex/config.toml (configuring/resetting the Codex provider) it now carries the user's intent forward by renaming [features].codex_hooks → [features].hooks (preserving its value, never clobbering an already-present hooks) and dropping the deprecated key. No-op when the flag is absent. (#4342 — thanks @Bian-Sh)
• fix(translator): same-format response path no longer leaks a data: null SSE event — the streaming response translator's same-format fast path returned [chunk] unconditionally, so the end-of-stream null/flush signal (chunk === null) propagated as a literal [null]. Downstream this surfaced as an empty data: null SSE event between chunks and crashed strict clients (e.g. Factory Droid BYOK on /v1/responses). The fast path now drops the null flush (returns []) while still passing real chunks through unchanged. (#4344 — thanks @thaitryhand)
• fix(translator): strip client-only assistant echo fields on the OpenAI target path (Mistral 422) — strict OpenAI-compatible upstreams (e.g. mistral/codestral-latest) reject client-only assistant "echo" fields sent back as input history with 422 extra_forbidden (the report hit messages[].assistant.reasoning_content via Codex /responses). Only reasoning_content was being stripped on the OpenAI target path; the sibling echo fields reasoning, refusal, annotations and cache_control leaked through and tripped the 422. They are now all dropped on the non-reasoner OpenAI target path. audio is deliberately preserved (OpenAI audio models reference a prior assistant audio response by id on multi-turn; Mistral never emits audio, so nothing is lost there). (#4350 — thanks @xxy9468615)
• fix(translator): accept AI SDK-style { type: "image", image: "data:…" } content parts — several OpenAI-input translators only recognized images shaped as image_url.url (or an object with .source/.url), so an AI SDK-style part where image is a bare data-URL string was silently dropped before reaching a vision provider (OpenCode is one affected client; the gap is generic). The OpenAI→Claude, OpenAI→Kiro and OpenAI→Gemini/Antigravity translators now parse a string image data URL into each provider's native image shape (Claude {source:{type:"base64"}}, Kiro images[].source.bytes, Gemini inlineData). (#4345 — thanks @mugnimaestra)
• fix(translator): Gemini accepts HTTP/HTTPS image URLs instead of silently dropping them — the OpenAI→Gemini request helper (convertOpenAIContentToParts) discarded remote image_url parts (emitting only a console.warn) because Gemini's inlineData needs base64 and the synchronous helper can't fetch+encode upstream. It now uses Gemini's native fileData: { fileUri } part for HTTP/HTTPS URLs (the model fetches the asset itself), so vision requests carrying a URL — not a data: URI — reach Gemini intact. (#4373 — ported from 9router#344, thanks @diegosouzapw)
• fix(executors): strip stream_options for qwen non-streaming / thinking Claude-Code requests — Claude-Code-compatible providers force the executor-level stream flag on while the outgoing body keeps the caller's original stream: false, so DefaultExecutor.transformRequest injected stream_options: { include_usage: true } onto a body that still said stream: false, and qwen rejected it with 400 "'stream_options' only set this when you set stream: true". The executor now strips stream_options whenever the body's effective stream is false. (#4374 — ported from 9router#663, thanks @anuragg-saxenaa / @diegosouzapw)
• fix(executors): don't inject thinking when tool_choice forces a tool (native Claude) — the Claude-Code wire-image emulation injects thinking: { type: "adaptive" } for non-Haiku Claude models, but Anthropic rejects thinking when tool_choice forces a specific tool ({type:"any"|"tool"}) with 400 "Thinking may not be enabled when tool_choice forces tool use.". Any Opus/Sonnet call that pins a tool (e.g. Claude Code's message_user, or agent harnesses that force a tool) hit a hard 400; the injection is now suppressed when tool_choice forces a tool. (#4389 — thanks @NomenAK)
• fix(codex): request reasoning summaries on Codex Responses requests — Codex/OpenAI Responses can return reasoning-token accounting and empty reasoning items unless visible reasoning summaries are requested, so Codex CLI / pi.dev paths missed visible thinking text. OmniRoute now requests reasoning.summary: "auto" (and includes reasoning.encrypted_content) when reasoning is enabled — preserving an explicit client reasoning.summary and existing include entries, and skipping it for reasoning.effort: "none". (#4359 — thanks @xz-dev)
• fix(sse): default the combo per-target timeout to 120s for fast failover — a combo's per-target timeout inherited the full FETCH_TIMEOUT_MS (600s default) when the combo didn't set targetTimeoutMs, so a single hung/slow target (e.g. an openai-compatible upstream returning 524/504) could stall the whole combo for up to 10 minutes before failing over. A new DEFAULT_COMBO_TARGET_TIMEOUT_MS = 120_000 is used as the default-when-unset in resolveComboTargetTimeoutMs (backward-compatible 3rd arg, wired in phaseComboSetup); an explicit ceiling/opt-out is preserved. (#4365 — thanks @diegosouzapw)
• fix(cli): Tailscale login honors TAILSCALE_AUTHKEY for non-interactive sign-in — startTailscaleLogin built tailscale up without ever reading process.env.TAILSCALE_AUTHKEY, so on a pre-authenticated / headless daemon the login waited for an interactive auth URL and timed out (~15s). When TAILSCALE_AUTHKEY is set it is now passed via --auth-key= (as a spawn argv element — no shell interpolation) so the daemon authenticates non-interactively; when unset, behavior is unchanged. (#4343 — thanks @ipeterpetrus)
• fix(dashboard): OAuth modal shows the real error on a non-JSON server response — the OAuth connect/reauth modal called await res.json() unconditionally, so when a build/OAuth endpoint returned a plain-text error (e.g. a 500 Internal Server Error page) the modal threw Unexpected token 'I'… and hid the real failure. Two shared helpers (parseResponseBody / getErrorMessage in src/shared/utils/api.ts) now read the body safely (JSON when it is JSON, raw text otherwise) and surface a clean message either way; all modal fetch sites use them. (#4351 — thanks @DNNYF)
• fix(dashboard): a disabled connection's last error is now visible — the provider card's error badge counts a disabled connection (isActive === false) that has an error (its effective status is still error/expired/unavailable), but the connection row hid the lastError text for disabled rows — so the operator saw the error count without being able to see what failed. The row now shows the error text whenever there is one, regardless of the active toggle. (#4352 — thanks @ntdung6868)
• fix(providers): the "Test Connection One-by-One" OAuth probe can no longer hang the queue forever — the OAuth connection-test path called bare fetch(url, { method, headers }) with no AbortController/signal/timeout, so when a provider's probe endpoint accepted the socket but never responded, the awaited fetch never settled and the one-by-one test queue stalled indefinitely (the API-key path was already bounded via validateProviderApiKey's timeoutMs). Both the initial probe and the post-refresh retry are now bounded with AbortSignal.timeout(30s) — matching the API-key path's 30s budget — and a timed-out probe resolves as a failure with a clear Test timed out after 30s message in the same shape as every other test error. (#4347 — thanks @ntdung6868)
• fix(providers): a deactivated account is labeled distinctly from a revoked token — a Codex connection whose OAuth refresh is fully healthy but whose ChatGPT account has been deactivated by the provider gets a 401 from the upstream API. The connection test labeled that the same as a bad credential (Token invalid or revoked → upstream_auth_error), so the operator couldn't tell a deactivated account from a revoked token. The test now reads the 401/403 body and, when it indicates account deactivation, classifies it as account_deactivated — which the dashboard already renders as "Account Deactivated". A plain auth 401 is unchanged. (#4353 — thanks @ntdung6868)
• fix(db): cascade-delete orphaned model aliases when a provider is removed — deleting a custom provider removed its connections and node but left behind the imported model-alias rows (stored as key=<alias>, value="<providerId>/<model>"). Those stale aliases then blocked re-importing the same provider — the import dedup treated them as "already exists", so no new models appeared. A new deleteModelAliasesForProvider(providerId) DB helper drops every alias whose stored value begins with <providerId>/ (leaving other providers and user-defined settings aliases untouched), and the provider-node DELETE handler now calls it after removing the connections and node, so a fresh import is unblocked. (#4348 — thanks @nguyenvanhuy0612)
• fix(api): persist max_input_tokens / max_output_tokens when adding a custom model — POST /api/provider-models silently dropped the per-model token limits set in the "add custom model" form: the handler destructured the rest of the body but never read max_input_tokens / max_output_tokens, and addCustomModel() had no parameter for them, so the values were thrown away on write. The DB layer (inputTokenLimit / outputTokenLimit) and the /v1/models catalog already round-trip these fields — only the write path was missing. The validation schema now accepts the two optional limits, the handler forwards them, and addCustomModel() persists them so a custom model's context/output window survives into the catalog. (#4349 — thanks @codename-zen)
• fix(plugin): the OpenCode static-catalog plugin prefixes combo/raw model keys with the provider id — OpenCode's static-catalog reader misdetected the omniroute provider: combo keys emitted as combo/MASTER were parsed as provider combo ("No credentials for provider: omniroute"), while a bare-MASTER form was misread as a model with no resolvable provider, and mixed omniroute/MASTER + bare-raw keys were rejected by OpenCode's schema. The plugin now emits every combo and raw model key prefixed with the omniroute provider id, emits the provider id explicitly, and drops the legacy combo/ prefix — so the static-catalog reader detects the provider and the auth loader returns the right credentials (the catalog-fetch timeout was also raised so a cold-start server doesn't publish an empty stub). (#4384 — thanks @herjarsa)

🔒 Security

• fix(security): scope the OAuth callback postMessage to a trusted-origin allowlist — the OAuth callback at /callback previously posted { code, state, … } to window.opener.postMessage(…, "*") whenever the opener was cross-origin, so a hostile page that opened the well-known redirect URI in a popup could receive the OAuth code/state and complete the flow as the user. The wildcard fallback is replaced with iteration over a fixed allowlist (same-origin + Codex's localhost:1455 / 127.0.0.1:1455 loopback helper); the browser silently drops postMessage to any opener whose origin isn't listed. (#4372 — ported from 9router#998, thanks @aeonframework / @diegosouzapw)
• fix(mitm): exact host membership in the MITM hosts test (CodeQL false positive) — tests/unit/mitm-tool-hosts.test.ts checked host membership with Array.includes(host), which CodeQL's js/incomplete-url-substring-sanitization heuristic misreads as a String.includes() URL-substring sanitization test (HIGH false positive). Switched to .some((h) => h === host) — identical semantics, no flagged pattern. (#4386)

📝 Maintenance

• docs: one-time feature-documentation catch-up (v3.8.20 → v3.8.30) — reconciled the docs with every user-facing feature shipped since v3.8.20: a new README ✨ What's New section; new guides for CLI integrations, MITM TPROXY transparent decrypt and delegated Anthropic Context Editing; refreshed AUTO-COMBO (auto/<category>:<tier> + Arena-ELO), API_REFERENCE (x-omniroute-no-memory), MEMORY (int8 quantization, off-by-default), RESILIENCE (model-lockout success-decay), RTK, AGENTBRIDGE, TRAFFIC_INSPECTOR, GUARDRAILS, CLOUD_AGENT, ENVIRONMENT; regenerated PROVIDER_REFERENCE (231 providers) and synced the provider count in README/CLAUDE/AGENTS. Going forward this runs every release (generate-release step 6b). (#4391)
• refactor(chatCore): extract the checkHeapPressureGuard leaf (god-file decomposition start) — first increment of decomposing chatCore.ts (~5127 LOC, the hottest path — every chat request flows through handleChatCore). The V8 heap-pressure guard at the top of handleChatCore (rejects with 503 when heapUsed exceeds the shed threshold) is moved to a self-contained, co-located utils/heapPressure.ts::checkHeapPressureGuard(...) with no behavior change. (#4371 — thanks @diegosouzapw)
• refactor(combo): de-dup the exhausted-target skip predicate across both dispatchers — the byte-identical #1731/#1731v2 pre-check (skip a target already exhausted on the provider/connection within a request) lived in both combo dispatchers; extracted to a shared combo/comboPredicates.ts helper. (#4362 — thanks @diegosouzapw)
• refactor(combo): de-dup the upstream-error exhaustion classification across both dispatchers — both dispatchers ran a near-identical post-error block classifying the upstream error and updating the exhaustion Sets (#1731 provider exhausted / #1731v2 connection error / transient rate-limited); extracted to a shared combo/targetExhaustion.ts::applyComboTargetExhaustion(...). (#4366 — thanks @diegosouzapw)
• chore(cli): localize CLI / scraping copy and stabilize fetch, memory & coverage handling — localizes CLI and scraping UX copy plus the Adapta onboarding tutorial (and corrects the CLI Code page title), makes fetch retries honor the start timeout, tightens SSE/response typing, respects configured memory token limits during search, and reduces CI coverage-merge memory by merging V8 data incrementally. (#4383 — thanks @JxnLexn)
• test(combo): reset circuit breakers between stream-readiness cases (restore green) — a stream-readiness fallback case failed on the release branch since the cycle-open tip due to test isolation: earlier combo-dispatch cases in the same file deliberately fail glm (tripping the module-level provider circuit breaker), and that OPEN state leaked into the next test so combo.ts skipped the model. The test now resets the circuit breakers between cases. (#4396 — thanks @diegosouzapw)
• chore(quality): reconcile the complexity ratchet baseline (1896 → 1900) — absorbs the small complexity-metric increase from the v3.8.31 /review-prs merge batch into quality-baseline.json so the ratchet reflects the shipped code (no production change). (#4410 — thanks @diegosouzapw)
• test/gate: reconcile release-time drift surfaced by the full CI gate — three already-merged changes left the release branch's full-CI gate red (the per-PR fast gates don't run it): the Gemini convertOpenAIContentToParts tests were realigned to the #4373 HTTP/HTTPS-URL fileData pass-through (they still asserted the old warn-and-drop behavior), the t11 any-budget for open-sse/executors/base.ts was raised to 2 with a justification (#4389 compares tool_choice against the string literal "any", not a TS any type), and the #4384 opencode-plugin combos test's net-assert reduction (dropping the obsolete combo/ namespace) was allowlisted. No production behavior change. (thanks @diegosouzapw)

---

[3.8.30] — 2026-06-20

✨ New Features

• feat(dashboard): category (media serviceKind) filter on the providers page — /dashboard/providers gains a media-category filter row (Image / Video / Music / Text→Speech / Speech→Text / Embedding) that composes with the existing search, free-only and "show configured only" filters. Membership is derived from the backend media registries (a provider that serves a kind is surfaced even if it never declared serviceKinds), keeping the UI in lockstep with the backend. (#4240)
• feat(combo): per-step account allowlist — scope a round-robin/weighted step to a subset of a provider's connections — a combo model step can now carry a first-class account allowlist so a round-robin (or weighted) strategy is scoped to a chosen subset of a provider's connections (e.g. only foo1+foo2 out of foo1..foo4) without hand-pinning one step per account. Empty = the whole active pool (unchanged). When a step both has an allowlist and is tag-routed, the two intersect (most-restrictive wins); a single pinned account still takes precedence. The combo builder's Precision step editor gains an optional "Restrict to accounts" picker. (#3266)
• feat(providers): add OpenAdapter, dit.ai and TokenRouter as OpenAI-compatible providers — three community-requested OpenAI-compatible aggregators now register as standard named OpenAI-style providers with live /v1/models discovery (the zenmux pattern), falling back to a seeded catalog when the upstream list is unavailable: OpenAdapter (https://api.openadapter.in/v1, free tier, 70+ open-source models — #4239), dit.ai (https://api.dit.ai/v1, dynamic-pricing router/gateway — #4155), and TokenRouter (https://api.tokenrouter.com/v1, free MiniMax model — #3841, thanks @FerLuisxd). No custom executor/translator — default OpenAI passthrough.
• feat(api): x-omniroute-no-memory request header — per-request opt-out of memory/skills injection — clients that manage their own context (e.g. their own RAG/memory) can send x-omniroute-no-memory: true (mirrors the existing x-omniroute-no-cache convention) to skip the gateway injecting up to memorySettings.maxTokens (~2k) tokens of memory and skills context into that chat request — avoiding the token/cost inflation it otherwise adds on every call. Absent the header, behavior is unchanged. (PRD-2026-06-19-no-memory-header)
• feat(dashboard): MITM tool card lists the exact hosts-file entries to add manually — the CLI-tools MITM card's "How it works" section now lists the full set of 127.0.0.1 <host> lines for the selected tool (sourced from the canonical MITM target registry) instead of a single example domain. Users on locked-down machines — where the automatic, sudo-gated hosts-file edit isn't available — can now copy every required entry by hand. (thanks @mrcyclo)
• feat(cli): omniroute launch-codex + setup-codex — run/configure the Codex CLI against OmniRoute — a launcher and setup command that point the Codex CLI at an OmniRoute endpoint (remote-mode aware). (#4270)
• feat(cli): Claude Code launcher + setup — remote mode + profiles — omniroute launch/setup for Claude Code with remote-mode support and named connection profiles. (#4274)
• feat(cli): OpenCode setup — OpenAI-compatible provider + remote-aware plugin — setup-opencode registers OmniRoute as an OpenAI-compatible provider for OpenCode and installs a remote-aware plugin. (#4277)
• feat(cli): one-command setup for popular AI coding tools — new setup-* commands that configure each tool to talk to OmniRoute: Cline (#4280), Kilo Code (#4284), Continue (#4289), Cursor (#4291), Roo Code (#4292), Crush (#4298), Goose (#4300), Qwen Code (#4301), Aider (#4302) and the Gemini CLI (native /v1beta) (#4303).
• feat(providers): provider model sweep — live discovery, refreshed catalogs, dead-provider cleanup — a broad sweep that enables live /v1/models discovery for more OpenAI-style providers (the zenmux pattern), refreshes the seeded catalogs with current models, and marks dead providers deprecated. (#4324)
• feat(mitm): translate Antigravity cloudcode end-to-end (Gap B) — the MITM decrypt path now translates Antigravity cloudcode traffic end-to-end. (#4299)
• feat(keys): per-key USD usage quota controls — an API key can now carry a USD spend quota that caps its usage once the threshold is reached. (#4327 — thanks @Witroch4)

🔧 Changed

• change(memory): memory is now OFF by default — DEFAULT_MEMORY_SETTINGS.enabled now defaults to false. Enabling memory injects up to ~2,000 tokens of retrieved context into every chat request (and that context is billed), which was a surprising default for new installs and for clients with their own context. Memory is now an explicit opt-in: installs that already enabled it keep it on; installs that never configured it default to off. The Settings → Memory panel now shows a token-cost warning when memory is enabled. (PRD-2026-06-19-no-memory-header)

🐛 Fixed

• fix(translator): Gemini accepts HTTP/HTTPS image URLs (no longer silently dropped) — OpenAI-style image_url parts whose URL was http://… or https://… reached convertOpenAIContentToParts (the OpenAI→Gemini request helper) and were dropped with only a console.warn, because Gemini's inlineData requires base64 and the helper is synchronous (it cannot fetch + encode). Gemini's Part schema, however, natively accepts fileData: { fileUri } for remote URIs — the model fetches the asset itself. The helper now emits a fileData part (mimeType: "image/*", inferred upstream on fetch) instead of dropping, so vision requests that pass a URL — not a data: URI — now reach Gemini intact. data: URIs still go through inlineData unchanged; unsupported schemes (e.g. ftp:) are still skipped. (thanks @East-rayyy)
• fix(security): OAuth callback page no longer relays code/state to a wildcard postMessage target — the OAuth callback at /callback posted { code, state, ... } to window.opener.postMessage(..., "*") whenever the opener was cross-origin (a fallback for the legitimate remote-dashboard + local-loopback callback scenario). A hostile page that opened the callback URL in a popup against the well-known redirect URI would therefore receive the OAuth code+state and could complete the OAuth flow as the user. The wildcard fallback is replaced with an iteration over a fixed allowlist of trusted target origins (same-origin + Codex's loopback helper at localhost:1455 / 127.0.0.1:1455); the browser silently drops the message for any opener whose origin is not in the list. Methods 2 (BroadcastChannel) and 3 (localStorage) — already in the page — still cover same-origin parents when the opener was severed by COOP. (thanks @aeonframework)
• fix(compliance): startup cleanup honors the dashboard data-retention setting instead of always trimming to 7 days — on every restart, cleanupExpiredLogs() (run at startup) read retention only from the CALL_LOG_RETENTION_DAYS / APP_LOG_RETENTION_DAYS env vars, which default to 7 days when unset, and trimmed usage_history (the Usage Analysis data) before the dashboard-based runAutoCleanup() — which respects the configured retention — ever ran. So a dashboard "Data Retention" of 90 days was silently overridden and the Usage Analysis page only ever showed the last 7 days after a restart. Retention now follows the precedence explicit env var → dashboard DB setting → 7-day default, per table (usage_history→usageHistory, call_logs/proxy_logs/request_detail_logs→callLogs, mcp_tool_audit→mcpAudit); an operator who sets the env var still wins, and non-DB deployments still fall back to it. (#4354 — thanks @akbardwi)
• fix(providers): bailian-coding-plan static fallback catalog matches the registry (10 models) — the provider-model sweep (#4324) added four current Model Studio coding-plan models (qwen3.7-plus, qwen3-coder-plus, qwen3-coder-next, glm-4.7) to the bailian-coding-plan registry entry but missed the static fallback mirror in staticModels.ts, which still listed only the older six. The static catalog (served when live discovery is unavailable) therefore diverged from the registry, and the existing static↔registry parity test went red on the release branch (only surfacing when test-impact analysis happened to select it). The static mirror now carries all ten models in registry order, restoring parity. (#4324)
• fix(executors): ArenaLLM accepts LMArena's split Supabase SSR auth cookie — LMArena migrated to @supabase/ssr chunked auth cookies: the single arena-auth-prod-v1 cookie is now empty and the real session is split across arena-auth-prod-v1.0, arena-auth-prod-v1.1, … (ascending). A user who pasted the (now-empty) single cookie therefore sent an empty session and upstream rejected it as "invalid cookie". The LMArena executor now reconstructs the single cookie from its chunks — reading .0, .1, … in ascending numeric order until one is missing and concatenating their raw values (@supabase/ssr's combineChunks rule: plain join(""), no base64-decode, no JSON-parse, the base64- prefix kept verbatim) — while preserving the rest of the pasted jar. A non-empty single cookie is still forwarded unchanged (back-compat). The credential UX now instructs pasting the full Cookie header and tracks the .0/.1 storage keys. (#4271 — thanks @caussao)
• fix(compression): preserve the cacheable prefix for automatic-cache providers — OpenAI / Codex (and Azure-OpenAI) use automatic prefix caching: the upstream caches the longest matching prefix of a request (system prompt + earliest messages) without any explicit cache_control markers in the body. The cache-aware compression guard only protected that prefix when the request carried explicit cache_control, so for automatic-cache providers the guard was skipped — and with compression enabled and preserveSystemPrompt: false (or a prefix-compressing mode like aggressive/ultra) it rewrote the system prompt / earliest messages, guaranteeing a cache miss and higher token spend through OmniRoute than going direct. The guard now treats a caching provider as sufficient on its own (isCachingProvider alone, independent of cache_control) to skip the system prompt and downgrade prefix-compressing modes, and OpenAI/Codex/Azure are now recognized as caching providers. Compression is still off by default — this only affects operators who enabled it with prefix preservation turned off. (#3955)
• fix(executors): DuckDuckGo AI Chat uses duckduckgo.com (fixes 400) — the DuckDuckGo AI Chat executor fetched status/chat and set Origin/Referer against https://duck.ai while still sending Sec-Fetch-Site: same-origin, so the request's same-origin triplet (host + Origin + Referer) was inconsistent and the backend rejected it with HTTP 400. All current DDG reverse-engineering references — and the provider registry's own baseUrl — use https://duckduckgo.com; the executor now uses it consistently for the status URL, chat URL, Origin, and Referer (the same-origin header is now coherent). The x-fe-version scrape regex also required a 40-hex tail but the real served token has a 20-hex tail (e.g. serp_20250401_100419_ET-19d438eb199b2bf7c300), so it silently fell back to a hardcoded default; the pattern is relaxed to a bounded {20,40} tail (still ReDoS-safe). This addresses the DuckDuckGo half of the report; the separate Chipotle/chipotle upstream breakage is tracked independently. (#4037 — thanks @daniij)
• fix(security): bound the prompt-injection scan to the first 16 KB (hot-path perf) — the prompt-injection guard joined every message/system string into one buffer and ran several regexes over the whole thing on every chat request, with no size cap — so a 300 KB body (pasted code, RAG context) meant O(body) CPU scanning on the hot path, a self-inflicted latency/GC source under concurrency. Both detection call sites (detectInjection in inputSanitizer.ts and the custom-pattern scan in promptInjection.ts) now slice the joined text to the first 16 KB (MAX_INJECTION_SCAN_BYTES) before the regex loop. Injection directives sit near the top of a prompt, so the generous cap preserves real detection while scanning only a bounded prefix; the existing 10 MB body-size cap (which protects ingestion) is unchanged. (#3932 — thanks @KooshaPari)
• fix(sse): retry direct-connection socket failures on a fresh socket (fewer 502 bursts) — the default direct-connection undici dispatcher pools keep-alive sockets for up to 4 s, but some edges (e.g. nvidia, opencode-zen) silently close idle keep-alive sockets within that window, so the next request reusing a pooled socket fails with UND_ERR_SOCKET ("other side closed") — in bursts. proxyFetch already retried once on such transient errors, but the retry reused the same pooled dispatcher and could grab another stale socket, then fell through to native fetch (which also pools) → the job sat in the rate-limit queue until the 30 s timeout → 502 + circuit-breaker open. The retry now uses a dedicated no-keep-alive / no-pipelining dispatcher so it opens a brand-new socket that can't be a dead pooled one; the first attempt still uses the pooled dispatcher (healthy keep-alive reuse is preserved). Complements the v3.8.29 diagnostics (describeFetchCause, #4281). (#4252 — thanks @klimadev)
• fix(sse): combo now stops at the first body-specific 400 instead of trying every target — the #2101 guard that detects a body-specific 400 (context overflow / malformed / model-access-denied, e.g. "model is not supported when using Codex with a ChatGPT account") logged "stopping combo" but executed a bare break, which only exited the inner retry loop; executeTarget then returned null and the outer target loop treated that as "this target produced nothing" and advanced to the next model. A combo of N targets that all reject the same request body therefore marched through all N (the report shows a 143-model Codex combo iterating every target), wasting upstream calls and per-attempt work. The guard now surfaces the 400 via the { ok, response } contract (mirroring the 499 client-disconnect path) so the combo resolves and stops immediately. (#4279)
• fix(sse): non-streaming combo over a Responses-API target no longer returns empty content — a Responses-API target (codex/cx) streams from upstream even on stream:false, and its terminal response.completed snapshot can carry a non-empty output that lacks the assistant message item (e.g. only a reasoning item) while the streamed output_text deltas had reconstructed the full message. The SSE→JSON aggregator preferred the terminal output wholesale, dropping the reconstructed text → HTTP 200 with empty content (hit notably via n8n, which defaults to stream:false). The aggregator now falls back to the reconstructed delta output when the terminal output has no message item but the reconstruction does; the terminal snapshot still wins whenever it already carries the message. (#3948)
• fix(executors): preserve tool-name casing on native Claude OAuth (read no longer leaks back as Read) — native Claude OAuth traffic runs through an anti-fingerprint tool-name cloak that renames a tool literally named read to Read on the wire and records the reverse alias on a non-enumerable _toolNameMap, which the response side uses to restore the client's original casing. Since v3.8.27 the executor returned a JSON-round-tripped copy of the body as transformedBody, and that round-trip dropped the non-enumerable map — so the restore saw an empty map and the cloaked Read streamed verbatim to the client, corrupting the tool name. The executor now re-attaches the cloak map onto the serialized body (mirroring the Antigravity executor), so tool-name casing round-trips correctly. (#4307 — thanks @dev-cj)
• fix(api): cache-HIT X-OmniRoute-Response-Cost now reports the incremental cost (≈0), not the original — on a semantic-cache HIT the gateway serves the stored response without an upstream call, but X-OmniRoute-Response-Cost was reporting the original call's full cost (recomputed from the cached usage). A consumer summing response-cost for billing was therefore charging for responses that cost ≈$0 to serve (and stale entries could inflate it). Cache hits now bill X-OmniRoute-Response-Cost: 0.0000000000 (the real incremental cost), and the avoided cost is surfaced in a new X-OmniRoute-Cost-Saved header for cache analytics — mirroring the existing tokens_saved concept. The MISS path is unchanged. (PRD-2026-06-19-cache-hit-cost-reporting)
• fix(models): imported vision-capable models keep their vision capability — after importing a provider key, vision-capable models (e.g. OpenRouter models whose architecture declares image input, and other synced providers) were listed as text-only in /v1/models and the dashboard — even though image requests actually worked. Synced model records never captured the vision flag, and the catalog's OpenRouter live-enrichment (which derives vision from architecture.input_modalities) is skipped once a provider has synced models. Discovery now captures supportsVision at sync time (from architecture.input_modalities, the string architecture.modality, or a top-level input_modalities), mirroring the existing supportsThinking capture, and the catalog surfaces capabilities.vision for synced models. (#4264 — thanks @FerLuisxd)
• fix(providers): Cloudflare Workers AI model discovery shows model names, not UUIDs — importing a Cloudflare Workers AI key listed models with internal UUID identifiers (e.g. 429b9e8b-d99e-…) instead of their usable slugs (@cf/meta/llama-3.1-8b-instruct). Cloudflare's /ai/models/search returns { id: "<uuid>", name: "@cf/…" }, and discovery was passing the raw objects through — so the UUID id became the callable model id. The cloudflare-ai discovery now maps each result's name → id, surfacing the real @cf/… model ids. (#4259 — thanks @FerLuisxd)
• fix(translator): clamp Responses API call_id to 64 characters — the OpenAI Responses API rejects call_id values longer than 64 characters with a 400. Long upstream tool-call ids (some clients emit ids well over the limit) are now clamped deterministically on both the function_call item and its matching function_call_output, so the pair stays matched through the orphaned-output filter and the request is accepted. (thanks @anuragg-saxenaa, @ngapngap)
• fix(oauth): GitHub Copilot token refresh now sends the public client_id — the github provider config never carried a clientId, so GitHub OAuth refresh_token exchanges either omitted client_id or sent the literal string undefined (and a bogus client_secret=undefined), which GitHub rejects — leaving a Copilot connection stuck once its short-lived token expired and the long-lived refresh path was needed. The provider now resolves its public device-flow client_id from the embedded public credential and omits client_secret entirely (GitHub's Copilot app is a public client with no secret). (thanks @baslr)
• fix(translator): a tool property named pattern survives Gemini/Antigravity schema sanitization — the Gemini schema sanitizer strips JSON-Schema constraint keywords Gemini rejects (pattern, minLength, …) at every nesting level, but it also deleted any tool property literally named one of those keywords. glob/grep tools declare a property called pattern, so on ag/* (Antigravity) backends that argument (and its required entry) was silently dropped, breaking the tools. Keyword stripping is now position-aware: it only removes constraint keywords at the schema-node level and never against the user-defined names inside a properties map. A genuine string-level pattern constraint is still stripped. (thanks @youthanh)
• fix(translator): MCP namespace tools flatten to individual functions on the Responses→Chat path — when a Codex CLI client routes a Responses-API request to a non-Codex backend (e.g. kr/claude-opus-4.7), each MCP server is declared as a namespace tool ({ type:"namespace", name, tools:[…] }). The Responses→Chat translator had no namespace branch, so the whole group collapsed into a single empty-schema function named mcp__<server>__ and every MCP call returned unsupported call: mcp__<server>__, breaking all MCP-based workflows (context7, codegraph, custom MCPs) for that combination. The translator now expands a namespace into one Chat function per sub-tool (preserving each sub-tool's name and parameters); an empty namespace yields no tools instead of a broken placeholder. The native Codex passthrough path was already correct. (thanks @V13t4nh)
• fix(cli): the active remote-context credential wins over an ambient OMNIROUTE_API_KEY — when a remote context is selected, its scoped access token now takes precedence over an OMNIROUTE_API_KEY present in the environment, so the connected remote is targeted as expected. (#4364)
• fix(cli): wire the contexts command into the CLI program — the omniroute contexts command (list/switch saved remote contexts) was implemented but never registered, so it was unreachable; it is now wired into the CLI program. (#4369)
• fix(mitm): mask bare Bearer <token> header values in the Traffic Inspector — the inspector now redacts bare Authorization: Bearer … values so tokens don't leak into captured traffic. (#4358)
• fix(pricing): price the gpt-5.x-pro OpenAI models + align the opencode-go discovery test — adds pricing for the gpt-5.x-pro models so cost telemetry reports a real cost instead of zero. (#4355)
• fix(sse): release the reader and cancel the stream on abort/error (no more Undici pool socket leak) — on abort or a mid-stream error the response reader is released and the stream cancelled, preventing leaked pooled sockets that degraded later requests. (#4309 — thanks @Ardem2025)
• fix(kiro): emit an early role-only start chunk to release the stream-readiness gate — Kiro streams now send an initial role-only chunk so the stream-readiness gate releases promptly instead of stalling. (#4311 — thanks @artickc)
• fix(dashboard): the proxy modal stops pre-filling new scopes with an unrelated proxy — adding a new scope assignment no longer inherits a previously-selected proxy's configuration. (#4312)
• fix(open-sse): inner-ai stops silently rerouting unmatched models to models[0] — an unmatched model id is no longer silently served by the first available model; the lookup now returns null and the request is handled explicitly. (#4310)
• fix(pollinations): handle auth-required premium models (claude, gemini, midjourney) — premium Pollinations models that require authentication are now handled correctly instead of failing. (#4266 — thanks @oyi77)
• fix(codex): isolate the Spark quota scope — Codex Spark usage is tracked under its own quota scope so it no longer bleeds into other Codex quotas. (#4293 — thanks @xz-dev)
• fix(dashboard): improve the API "try it" functionality — fixes the request path used by the dashboard's API "try it" panel. (#4296 — thanks @edrickrenan)
• fix: polyfill crypto.randomUUID for non-secure contexts — restores UUID generation when the dashboard is served over a non-secure (plain-HTTP) origin where crypto.randomUUID is unavailable. (#4287 — thanks @pizzav-xyz)
• fix(proxy): allow concurrent proxy dispatcher streams — the proxy dispatcher no longer serializes streams, so concurrent requests through a proxied connection run in parallel. (#4288 — thanks @wilsonicdev)
• fix(build): co-locate llmlingua SLM optionals into dist/node_modules (postinstall) — the optional llmlingua SLM packages are co-located into the standalone build so the compression worker can actually spawn in production. (#4286)
• fix(mitm): surface AgentBridge traffic in the Traffic Inspector (D4 ingest) — AgentBridge requests now appear in the Traffic Inspector. (#4285)
• fix(sse): surface undici err.cause on dispatcher failure — dispatcher failures now flatten the cause chain (and AggregateErrors) into the error detail for diagnosability. (#4281)
• fix(cli): harden launch/launch-codex with free-claude-code patterns — the launchers adopt the hardened launch patterns ported from free-claude-code. (#4278)
• fix(compression): end-to-end audit — fixes across the whole compression flow — a sweep of the compression pipeline fixing ultra/aggressive/lossless edge cases, accessibility-anchor handling, language detection, and mode decoupling. (#4323)

🧪 Tests

• test: align two tests left red by merged PRs — re-aligns the db-rules classification count (#4335) and the LMArena split-cookie metadata test (#4271) after concurrent merges. (#4346)
• test(ci): reconcile the release/v3.8.30 baseline + test drift — reconciles quality baselines and drifted tests accumulated on the release branch. (#4276)

📝 Maintenance

• refactor(combo): ComboContext + extract phaseComboSetup (god-file split, phase 1) — begins decomposing the combo god-file by extracting combo setup into a context object, without touching dispatch/semaphore logic. (#4326)
• feat(quality): cap test-file size — anti-reinflation Layer 1 — freezes the existing god-tests and caps new test files at 800 lines to stop re-inflation. (#4273)
• feat(quality): seed per-module mutationScore floors + a blocking aggregation ratchet (T3) — adds per-module mutation-score floors with a blocking aggregate gate. (#4305)
• feat(quality): make the a11y gate real (@axe-core/playwright in nightly) — wires the previously-phantom accessibility gate into the nightly run with real baselines. (#4321)
• feat(quality): unblock R1 — test-redundancy measurement via disableBail — enables the test-redundancy measurement that was previously blocked by fail-fast. (#4322)
• fix(quality): the complexity gate now covers bin/ + electron/, and tracked-artifacts runs in pre-commit — extends the complexity gate's scope and moves the tracked-artifacts check into the pre-commit hook. (#4318)
• fix(quality): restore release/v3.8.30 green — 3 latent reds from concurrent merges — fixes three latent test reds surfaced by concurrent merges into the release branch. (#4335)
• fix(combo): keep phaseComboSetup under the complexity ceiling — extracts a helper so the new combo setup phase stays under the complexity gate. (#4338)
• ci(mutation): split over-budget batches by range/pair so every batch fits the job cap — re-splits the mutation batches so each fits the CI job budget. (#4272)
• chore(ci): align the electron audit gate to the root advisory policy — the electron-workspace audit gate now follows the same advisory policy as the root. (#4275)
• chore(quality): reconcile the complexity/quality baselines across concurrent-merge drift — rolls up the cycle's baseline reconciliations driven by concurrent merges into the release branch. (#4330, #4336, #4370)
• docs: ban AI-generation footers in commits/PRs/CHANGELOG (Hard Rule #16) — codifies the prohibition on AI-generation footers and bot co-author trailers. (#4328)
• docs(design): add the OmniRoute design system and visual identity specification — adds the design-system / visual-identity specification document. (thanks @diegosouzapw)

🔒 Security

• fix(sse): harden the DuckDuckGo lite scraper sanitization (CodeQL) — closes four HIGH CodeQL alerts in the no-key web-search scraper: decodeEntities now resolves & last so an already-escaped entity (e.g. <) survives as literal text instead of being double-unescaped (js/double-escaping); stripTags decodes entities first, then strips tags in a loop to a fixpoint and drops any trailing unclosed <…, so entity-encoded markup like <script> can never reach the LLM/client as a live tag (js/incomplete-multi-character-sanitization); and the host checks in the search tests use new URL().hostname equality instead of substring .includes (js/incomplete-url-substring-sanitization). (#4356)

🔧 Dependencies

• fix(deps): bump undici to 7.28.0 and dompurify to 3.4.11 (security) — addresses the undici SOCKS5-TLS / cache advisories and the dompurify advisory. (#4306)
• chore(deps): bump actions/checkout from 4 to 7 — CI checkout-action update. (#4297)
• fix(executors): strip stream_options for qwen non-streaming / thinking-mode Claude Code requests — Claude-Code-compatible providers force the executor-level stream flag on via upstreamStream = stream || isClaudeCodeCompatible (open-sse/handlers/chatCore.ts), but the outgoing body keeps the caller's original stream: false. The shared stream && targetFormat === "openai" branch in DefaultExecutor.transformRequest then injected stream_options: { include_usage: true } onto a body that still said stream: false, and qwen upstream rejected it with 400 "'stream_options' only set this when you set stream: true". Same rejection when the body carries thinking / enable_thinking. The qwen branch now skips the injection (and strips any client-sent stream_options) when the body explicitly says stream: false or requests thinking, leaving regular qwen streaming requests with the usage injection intact. (thanks @anuragg-saxenaa)

---

[3.8.29] — 2026-06-19

✨ New Features

• feat(cloud-agent): Cursor Cloud Agent via the official API-key REST API (no IDE-OAuth ban risk) — adds a cursor-cloud cloud agent that drives Cursor's Background / Cloud Agents through the official REST API (api.cursor.com) authenticated with a user or service-account API key — the safer, first-party alternative to re-using the Cursor IDE's OAuth session (the existing cursor provider, which carries a ban-risk warning). Implemented as a plain REST adapter mirroring the Devin/Jules agents (createTask/getStatus/sendMessage/listSources), so it does not pull in the @cursor/sdk package and its per-platform native binaries (Cursor's SDK is itself a thin wrapper over this REST API). Cursor's UPPERCASE status enums (CREATING/RUNNING/FINISHED/ERROR) are mapped explicitly to the shared CloudAgentStatus, and baseUrl is overridable per-credential. Credentials are stored encrypted via the existing cloud_agent_credentials table; no schema change. (#4227 — thanks @MRDGH2821)
• feat(routing): OpenRouter-style auto/<category>:<tier> combos — auto-routing now understands suffixed combos that separate the category (what kind of route) from the tier (how to optimize): auto/coding:fast, auto/coding:cheap (alias :floor), auto/coding:free, auto/coding:pro, auto/coding:reliable, plus the new category roots auto/reasoning, auto/vision, auto/multimodal. The tier picks the scoring weights — :fast → ship-fast, :cheap/:floor → cost-saver, :reliable → a new reliability-first pack (circuit-breaker health + latency stability) — while :free/:pro filter the candidate pool by model tier (classifyTier: free-tier vs. premium models). The category filters the pool by capability (vision/multimodal → vision-capable models, reasoning → reasoning/thinking models). Any valid auto/<category>:<tier> resolves on demand; a curated set is advertised in /v1/models and the dashboard. Filtering is fail-open — if a constraint matches no connected models the full pool is used so routing never breaks. All composition lives in the new open-sse/services/autoCombo/suffixComposition.ts; the core combo scorer (combo.ts) is untouched. Second slice of #4235 (premium account-tier weighting is a later follow-up). (#4235 — thanks @MRDGH2821)
• feat(routing): advertise the auto/cheap, auto/offline, auto/smart combos (catalog ↔ README sync) — the README lists auto/cheap (cheapest-per-token first), auto/offline (most quota/rate-limit headroom first) and auto/smart (quality-first + 10% exploration), and they already resolved at request time via parseAutoPrefix → createVirtualAutoCombo. But they were missing from AUTO_TEMPLATE_VARIANTS, so /v1/models and the dashboard combos list (which iterate that catalog) never showed them — the catalog drifted from the docs (visible in the issue's screenshots). Added the three entries so they're advertised everywhere alongside the other built-in auto/* combos. First slice of #4235 (OpenRouter-style auto/<category>:<tier> suffixes + new categories follow). (#4235 — thanks @MRDGH2821)
• feat(cli): remote mode — drive a remote OmniRoute with scoped access tokens — a new CLI mode that connects to a remote OmniRoute instance using scoped access tokens, so a local CLI can drive a server you don't own a session on. (#4256)
• feat(api): cost-telemetry parity — X-OmniRoute-* headers on every endpoint + a non-token cost engine — every endpoint now emits the X-OmniRoute-* cost/usage headers, backed by a cost engine that also prices non-token (media/request-based) usage. (#4247)
• feat(api): register Kimi K2.7 Code models (kimi-k2.7-code + -highspeed) — the new Moonshot thinking-only coding models are registered (fixed sampling; temperature/top_p marked unsupported). (#4183)
• feat(catalog): add kimi-k2.7-code to the kmca catalog + qwen-web models discovery — surfaces the new Kimi coding model in the kmca catalog and wires qwen-web into model discovery. (#4185)
• feat(api): expand the zai provider catalog with GLM-5.2 / GLM-4.7 — adds the real GLM-5.2, GLM-4.7 and GLM-4.7-flash model ids to the Anthropic-direct zai provider. (#4201)
• feat(api): no-thinking gateway model IDs (FCC port, Fase 8.1) — gateway model id variants that force thinking off, ported from free-claude-code. (#4145)
• feat(sse): mid-stream continuation for truncated streams (FCC port, Task 4.4) — when a stream is cut short, OmniRoute can transparently continue it, ported from free-claude-code. (#4147)
• feat(sse): per-provider sliding-window rate-limit fallback (FCC port, Fase 8.2) — a per-provider sliding-window rate limiter as a fallback path, ported from free-claude-code. (#4146)
• feat(sse): transparent stream recovery (FCC port, Fase 4, opt-in) — opt-in transparent recovery of interrupted upstream streams, ported from free-claude-code. (#4131)
• feat(search): free DuckDuckGo web search as a last-resort provider (FCC port, Fase 6) — adds a no-key DuckDuckGo web-search provider used as a last resort, ported from free-claude-code. (#4136)
• feat(logging): credential-redaction safety net in the pino logger (FCC port, Fase 8.3) — a logger-level redaction pass that scrubs credentials from log output, ported from free-claude-code. (#4140)
• feat(memory): opt-in Qdrant scalar int8 quantization (F4.4 Q1) — opt-in int8 scalar quantization for Qdrant-backed memory vectors. (#4187)
• feat(memory): opt-in sqlite-vec int8 vector quantization (F4.4 Q2) — opt-in int8 quantization for the sqlite-vec memory backend. (#4190)
• feat(deploy): keep optional deps on update (--include=optional) — the in-place update path now passes --include=optional so native/optional packages aren't dropped on update. (#4260)
• feat(dashboard): unified visual identity — grid, primitives, tables, form controls (design phases 1-4) — a sweeping design pass aligning the dashboard with the site: grid wallpaper, button/card/input primitives, theme-aware tables and form controls. (#4122)
• feat(dashboard): grid wallpaper on all standalone screens + fluid 4K layout — the identity grid now backs every standalone screen and the layout scales fluidly to 4K. (#4158)
• feat(dashboard): make the identity grid visible + unify the focus ring on accent — design follow-up making the grid actually visible and standardizing focus rings on the accent color. (#4141)
• feat(dashboard): import only free models + free-model list controls — the model-import page can import just the free models, with controls to manage the free-model list. (#4176 — thanks @felipesartori)
• feat(dashboard): compact grid layout for no-auth provider accounts — a denser grid layout for provider accounts when auth is disabled. (#4137 — thanks @felipesartori)
• feat(dashboard): derive media serviceKinds from the registries (surface MiniMax + the media catalog) — /media-providers/[kind] now derives its service kinds from the registries instead of a hand-maintained list, surfacing ~48 previously-invisible media providers (incl. MiniMax TTS/video/music). (#4212)
• feat(traffic-inspector): live (in-flight) request filter (Gap 5) — the Traffic Inspector can filter to in-flight requests as they happen. (#4130)
• feat(agent-bridge): maintenance & diagnostics dashboard controls — adds maintenance and diagnostics controls for the Agent Bridge to the dashboard. (#4127)
• feat(mitm): TPROXY IP_TRANSPARENT native addon + conditional loader (Epic A) — a native IP_TRANSPARENT addon with a conditional loader, the foundation for TPROXY capture. (#4148)
• feat(mitm): Fase 3 Epic A spike — TPROXY command builder — a transactional builder for the iptables/TPROXY command set. (#4139)
• feat(mitm): TPROXY setup layer — transactional apply/revert (Epic A) — applies and reverts the TPROXY routing setup transactionally. (#4144)
• feat(mitm): add setSocketMark to the TPROXY addon (anti-loop primitive) — exposes setSocketMark so OmniRoute's own egress can be marked and skipped (anti-loop). (#4160)
• feat(mitm): TPROXY capture-mode listener + connectMarked (Epic A) — the capture-mode listener plus a marked-connect primitive. (#4169)
• feat(mitm): dynamic per-SNI cert authority for TPROXY (TLS decrypt 1/N) — a per-SNI on-the-fly certificate authority, the first slice of TLS decrypt. (#4173)
• feat(mitm): TLS-terminating capture for TPROXY (decrypt 2/N) — terminates TLS to capture decrypted traffic. (#4179)
• feat(mitm): wire the TLS decrypt engine into TPROXY capture mode (decrypt 3/N) — connects the decrypt engine to the capture-mode pipeline. (#4200)
• feat(mitm): TPROXY capture-mode manager (decrypt 4a/N) — a manager coordinating the TPROXY capture lifecycle. (#4208)
• feat(mitm): local-only route + trust-store installer for TPROXY decrypt (4b/N) — a loopback-only management route plus a CA trust-store installer for the decrypt CA. (#4211)
• feat(dashboard): TPROXY decrypt capture toggle in the Traffic Inspector (4c/N) — a UI toggle to enable/disable decrypted capture. (#4216)
• feat(compression): replace the headroom tabular encoder with a vendored GCF — swaps the tabular encoder for a vendored GCF implementation. (#4167 — thanks @blackwell-systems)
• feat(compression): live per-engine streaming via compression.step (F3.3) — streams per-engine compression progress through a compression.step event. (#4217)
• feat(compression): show an engine node for single-engine runs in the studio — the Compression Studio now renders an engine node even when only one engine runs. (#4210)
• feat(compression): expose the WaterfallInspector via a Canvas/Waterfall toggle — adds a Canvas/Waterfall view toggle that surfaces the WaterfallInspector. (#4238)
• feat(compression): make mcpAccessibility config reachable via a settings sub-route — exposes the mcpAccessibility config under a dedicated settings sub-route. (#4237)
• feat(compression): runnable A/B benchmark CLI (F2.4) — a CLI to run A/B compression benchmarks. (#4220)
• feat(compression): add a transcript loader to the replay harness — the replay harness can now load real transcripts. (#4246)
• feat(compression): wire MCP tool-cardinality reduction (F4.3, opt-in) — opt-in reduction of MCP tool-set cardinality to shrink prompts. (#4221)
• feat(compression): wire RTK comment-stripping config + honor preserveDocstrings — RTK comment-stripping is now configurable and honors a preserveDocstrings flag. (#4242)
• feat(compression): honor the per-filter RTK deduplicate flag — RTK filters now respect a per-filter deduplicate flag. (#4231)
• feat(compression): honor the registry enabled flag in the stacked loop — the stacked compression loop now skips engines disabled in the registry. (#4244)
• feat(compression): persist RTK grouping config (unlock R5 enableGrouping) — persists the RTK grouping configuration, unlocking the R5 enableGrouping rule. (#4207)
• feat(compression): wire ultra's modelPath/slmFallbackToAggressive to the LLMLingua SLM tier — connects the ultra tier's small-language-model knobs to the LLMLingua SLM path. (#4257)
• feat(quality): Onda 2 mutation-gate tooling — radiography classifier (T1) + mutationScore ratchet (T3) — new mutation-testing tooling: a survivor-radiography classifier and a mutationScore ratchet. (#4234)
• feat(ci): wire the F2.4 compression budget-gate ratchet — adds a CI ratchet that gates compression budget regressions. (#4232)

🐛 Fixed

• fix(providers): qwen-web model discovery now lists the live catalog instead of nothing — the qwen-web cookie provider had no entry in PROVIDER_MODELS_CONFIG, so its model-discovery page returned an empty/stale local catalog (the OAuth fallback at the top of the route only fires for provider === "qwen", leaving qwen-web to fall through to the no-config branch). Added a qwen-web entry that fetches the public https://chat.qwen.ai/api/v2/models endpoint (no auth header) and parses the { data: { data: [{ id, name, owned_by }] } } shape (with a flatter { data: [] } fallback). This is Problem #3 of #3931 (diagnosed by @thezukiru); Problem #1 — validator bare-token false-positive — shipped earlier in #3958, and Problem #2 — empty stream from Qwen WAF bot-detection on the streaming endpoint — remains a separate upstream/stealth concern. (#3931 — thanks @thezukiru)
• fix(providers): ZenMux model discovery now lists the live catalog (incl. the free models) instead of the stale 9-entry hardcoded list — adding a ZenMux key validated fine, but the connection then showed API unavailable — using local catalog and was missing the free models ZenMux advertises (z-ai/glm-5.2-free, moonshotai/kimi-k2.7-code-free). Root cause: zenmux carries a correct modelsUrl in the registry, but — like llm7/byteplus before #3976 — it was not classified by any live-fetch branch of the model-import route (not openai-compatible-*, not self-hosted, not in NAMED_OPENAI_STYLE_PROVIDERS), so the route never probed the upstream /models and fell through to the registry's hardcoded models[]. Added zenmux to NAMED_OPENAI_STYLE_PROVIDERS, so the route probes https://zenmux.ai/api/v1/models (the /chat/completions-stripped <baseUrl>/models candidate) and serves the live list, falling back to the local catalog only when the upstream fetch fails — import never breaks. (#4202 — thanks @mikmaneggahommie)
• fix(providers): Vercel AI Gateway "import models" now loads the live catalog instead of nothing — adding a Vercel AI Gateway key worked, but clicking import on the models page loaded nothing usable (manually adding the same models worked). Same class as #4202 (zenmux) / #3976 (llm7/byteplus): vercel-ai-gateway carries a real baseUrl (https://ai-gateway.vercel.sh/v1/chat/completions, format openai) in the registry, but was not classified by any live-fetch branch of the model-import route (not openai-compatible-*, not self-hosted, not in NAMED_OPENAI_STYLE_PROVIDERS), so the route never probed the upstream /models and fell through to the registry's tiny 5-entry hardcoded models[]. Added vercel-ai-gateway to NAMED_OPENAI_STYLE_PROVIDERS, so the route probes https://ai-gateway.vercel.sh/v1/models (the /chat/completions-stripped <baseUrl>/models candidate) and serves the live list, falling back to the local catalog only when the upstream fetch fails — import never breaks. (#4249 — thanks @FerLuisxd)
• fix(sse): clear error when the request queue drops a job (no more fake-upstream "This job timed out after Nms") — under concurrent load, requests that exceed the per-connection rate-limit queue budget (resilienceSettings.requestQueue.maxWaitMs) were dropped by Bottleneck with its raw This job timed out after <maxWaitMs> ms. message. That string is indistinguishable from an upstream gateway timeout, so the 502 body and call-log last_error looked like a provider outage across unrelated providers (TI:0|TO:0) — an operator spent ~3h misdiagnosing local queue saturation as upstream failures. withRateLimit now rewrites that specific Bottleneck error into a clear, OmniRoute-owned message that names the knob (requestQueue.maxWaitMs, tunable in Settings → Resilience), explicitly disclaims an upstream timeout, preserves the original as cause, and tags code: "RATE_LIMIT_QUEUE_TIMEOUT". Behavior is unchanged — the job is still dropped so combo falls back to the next target. (#4165 — thanks @KooshaPari)
• fix(api): advertise the built-in auto/* combos in /v1/models — OmniRoute ships a zero-setup auto/* catalog (auto/best-coding, auto/pro-reasoning, …, 16 variants) that the dashboard advertises and that resolve on demand, but the /v1/models listing only emitted persisted DB combos + provider models. Clients that build their model picker from /v1/models (e.g. Hermes Agent) never saw any auto/* option. The catalog now emits every AUTO_TEMPLATE_VARIANTS id (as owned_by: "combo") at the top of the list, deduped against persisted combos. (Showing each auto/*'s dynamically-selected members is a separate enhancement.) (#4164 — thanks @MRDGH2821)
• fix(sse): restore MCP / third-party tool names on the native Claude path (MCP dispatch broken in Claude Code) — since 3.8.27, every MCP tool call routed through OmniRoute to a native Claude OAuth provider failed client-side with Error: No such tool available: <PascalCaseName>: tool schemas arrived fine but the streamed tool_use.name reached Claude Code in its cloaked form (e.g. McpN8nMcpSearchWorkflows instead of the registered mcp__n8n-mcp__search_workflows). The native-Claude tool-name cloak stashes its per-request alias→original map as a non-enumerable _toolNameMap on the request body; the request-inspector capture added in 3.8.27 rebuilds the captured body from its serialized form (JSON.parse(JSON.stringify(...))), which drops non-enumerable properties, so finalBody._toolNameMap was empty and the response-side un-cloak silently fell back to the static built-in map — never restoring dynamic MCP / snake_case names. Built-in tools (Bash/Read/…) were unaffected (static map); cross-format paths were unaffected (they attach the map enumerably). The provider-request capture now re-attaches the per-request map (kept non-enumerable, so it still never re-serializes upstream) when the captured copy lost it, restoring MCP tool dispatch. (#4091 — thanks @pedrotecinf, @NakHalal)
• fix(dashboard): Logs auto-refresh self-heals in embedded/proxied hosts that pin or mis-fire visibility — a follow-up to #4054: the Request Logger still froze auto-refresh on some hosts (reported on 3.8.28 Docker, works on 3.8.24). #4054 made the initial visibility fail-open, but the pause is event-driven — a host that fires a one-shot visibilitychange → hidden and then keeps reporting "hidden" (or recovers without firing the event again) left the cached visibility flag stuck false, so the interval ticked but never polled (only the manual Refresh button worked). The poll tick now also re-checks the live document.visibilityState, and a window focus listener re-arms polling (a focused window is a reliable signal the page is actively viewed). A genuinely backgrounded browser tab still pauses (it reports "hidden" and never receives focus), preserving the #3109 network-saturation optimization. (#4133 — thanks @tjengbudi)
• fix(capabilities): unify vision model-id detection into one shared source — three code paths kept independent, drifting vision-model lists, so the same model id could get up to three different verdicts. Two concrete bugs: lite compression's gate was missing pixtral / llava / qwen-vl / glm-4v / kimi-vl / mistral-medium-3, so it stripped images for those real vision models and blinded them (same class as #4071 / #4012); and the /v1/models list was too broad, flagging text models (gemma, bare kimi like kimi-k2) as vision. All three (modelCapabilities routing fallback, /v1/models listing, lite image-strip gate) now delegate to a single conservative source src/shared/constants/visionModels.ts, which also restores glm-4v / gemini-3 coverage and keeps the #3328 MiniMax M3 carve-out. (#4072 — thanks @diego-anselmo)
• fix(sse): surface mid-stream Gemini errors instead of returning a truncated 200 — when an upstream Gemini SSE stream emitted some partial content and then a JSON error object ({"error":{"code":503,"message":"…high demand…","status":"UNAVAILABLE"}}) instead of a candidates payload, OmniRoute silently dropped it: the gemini→openai translator's no-candidate branch only handled promptFeedback (content-filter) and returned null for anything else, so the stream simply ended and the client got HTTP 200 with a truncated body and finish_reason: "stop" — masking the failure and skipping combo fallback. geminiToOpenAIResponse now detects an error object (optionally wrapped in response), records it as state.upstreamError (preserving the real status — 503/UNAVAILABLE, or 429 for RESOURCE_EXHAUSTED), and lets stream.ts error the stream out through the existing onFailure/buildErrorBody/controller.error path — the same mechanism the openai-responses translator already uses. (#4177 — thanks @hartmark)
• fix(capabilities): resolve models.dev-synced vision metadata for Mistral -latest aliases — root cause behind the #4071 heuristic: getResolvedModelCapabilities("mistral/pixtral-12b-latest").supportsVision resolved null (vision came only from the #4071 model-id heuristic, with attachment still null) even though models.dev exposes the model as multimodal. Confirmed against the live models.dev API: it catalogs Pixtral 12B under the short id pixtral-12b (with attachment: true, modalities.input: ["text","image"]), while requests use the Mistral API alias pixtral-12b-latest. The synced lookup tried the exact / raw / static-spec-canonical ids — all of which miss the short form — so it fell through to the heuristic. getSyncedCapabilityForResolved now adds a last-resort fallback that retries with a trailing -latest stripped, so synced metadata (attachment / image modalities) wins for these aliases; models whose -latest id is stored verbatim (e.g. pixtral-large-latest) keep resolving directly. Note: the models.dev sync is currently manual-only (Settings → models.dev) with no scheduled refresh, so a fresh instance still relies on the #4071 heuristic until that sync runs — a periodic-refresh cadence is left as a separate follow-up. (#4073 — thanks @diego-anselmo)
• fix(sse): map Xiaomi MiMo reasoning control to its native thinking:{type} shape — MiMo (api.xiaomimimo.com) controls chain-of-thought only via top-level thinking:{type:"enabled"|"disabled"} and does not understand OpenAI's reasoning_effort/reasoning, while its request validator is strict (400 Param Incorrect). OmniRoute's OpenAI path carried reasoning intent as reasoning_effort, and the claude→openai translator can leave a Claude-shaped thinking:{type, budget_tokens} — so the client's on/off choice was silently dropped and budget_tokens/reasoning_effort rode along as extra params the validator can reject. New open-sse/services/mimoThinking.ts::normalizeMimoThinking (wired in chatCore for provider==="xiaomi-mimo") reduces any thinking object to just {type} (disabled stays; enabled/adaptive/other → enabled) and drops reasoning_effort/reasoning. It deliberately does not synthesize thinking from a bare reasoning_effort — mimo-v2-omni is non-thinking, so that could turn a silently-ignored param into a hard error. (#4224)
• fix(capabilities): Xiaomi MiMo *-pro chat models are text-only (no vision) — only mimo-v2.5 and mimo-v2-omni accept images per Xiaomi's docs; mimo-v2.5-pro/mimo-v2-pro are text-only, but modelSpecs marked them vision-capable and models.dev mislabels them (hermes-agent#18884). Since resolveVisionCapability lets a synced attachment:true win first, an image request could be routed to a blind model (the #4071 failure mode). Corrected the specs and added a hard override in resolveVisionCapability (checked before the synced branch, anchored so mimo-v2.5-pro never matches the multimodal mimo-v2.5) that beats the wrong synced attachment. Also registered the missing native mimo-v2-pro chat model and the missing mimo-v2-tts speech model. (#4224)
• fix(sse): Claude Opus 4.7+/Fable 5 use adaptive thinking only (no more manual-budget 400s) — Opus 4.7 and later (Opus 4.7/4.8, Fable 5) removed manual extended thinking: thinking.type:"enabled" or any thinking.budget_tokens now returns 400 ("Any request that tries to set a fixed thinking budget gets a 400" — Anthropic migration guide). Reasoning is adaptive-only, steered by output_config.effort. OmniRoute's OpenAI→Claude translator mapped reasoning_effort low/medium/high to a manual thinking:{type:"enabled", budget_tokens}, so those requests hard-400'd on the most-used provider (and a Claude-native passthrough client sending the legacy shape did too). A new adaptiveThinkingOnly model flag now drives two fixes: the translator maps reasoning_effort of every level to {type:"adaptive"} + output_config.effort (preserving the requested level, never a budget) for these models, and a normalizeClaudeAdaptiveThinking catch-all at the existing post-translation thinking-normalization chokepoint collapses any residual manual thinking (passthrough legacy shape, per-model defaults) to {type:"adaptive"}, keyed on the resolved upstream model so it covers every routing mode. Pre-4.7 models (Opus 4.6/4.5, Sonnet, Haiku) keep manual budgets unchanged. (#4230)
• fix(providers): strip non-default temperature/top_p/top_k for Claude Opus 4.7+/Fable 5 (fixed sampling → no 400) — Opus 4.7 and later reject non-default temperature/top_p/top_k with a 400 (sampling is fixed; reasoning moved to output_config.effort). The translator forwarded client-supplied temperature/top_p unconditionally and the Claude registry models carried no unsupportedParams, so a plain OpenAI-format request with temperature: 0.7 to claude-opus-4-8 hard-400'd. Added unsupportedParams: ["temperature","top_p","top_k"] to the Opus 4.7+/Fable 5 ids in both the claude (dashed claude-opus-4-8) and anthropic (dotted claude-opus-4.7) registries, so they're stripped at the existing getUnsupportedParams dispatch chokepoint. Pre-4.7 Claude models still accept sampling params. (#4230)
• fix(providers): conditionally strip temperature/top_p for GPT-5 reasoning on the openai Chat Completions path (no 400 when an effort is active) — GPT-5 reasoning models reject non-default temperature/top_p with a 400 whenever a reasoning effort is active, yet accept them again under reasoning_effort:"none" (the GPT-5.1+ default, i.e. non-reasoning mode). On the openai provider only o3 carried REASONING_UNSUPPORTED; gpt-5.5/gpt-5.4/gpt-5.4-mini/gpt-5.4-nano carried no sampling guard, so a temperature + active-effort request hard-400'd. A static unsupportedParams list can't express the none-mode carve-out (it would over-strip the legitimate case), so the new gpt5SamplingGuard drops temperature/top_p only when the resolved effort is active — wired at the existing getUnsupportedParams chokepoint and scoped to the openai Chat Completions surface (the codex Responses path is already covered by the CodexExecutor allowlist; other providers are untouched). (#4245)
• fix(codex): stop silently dropping GPT-5 output verbosity (verbosity / text.verbosity) — the GPT-5 series added an output-verbosity control: verbosity (low/medium/high) on Chat Completions, nested as text.verbosity on the Responses API. The CodexExecutor gates translated requests through an allowlist that had no text entry, so for the codex provider the hint was dropped before reaching upstream (the openai Chat path already forwarded it). normalizeCodexVerbosity now folds whichever shape arrived into a single validated text:{verbosity} before the allowlist (which now permits text), and the OpenAI Chat↔Responses request translators map verbosity across formats so the hint survives a format crossing for non-codex Responses backends too. Invalid/absent verbosity collapses to no text (status quo). (#4245)
• fix(sse): map reasoning_effort to DeepSeek V4's native {high, max} vocabulary — DeepSeek V4 only understands high/max reasoning levels, so other reasoning_effort values are mapped onto its native vocabulary instead of being rejected. (#4219)
• fix(glm): default max_tokens and an extended timeout for GLM-5.2+ thinking — GLM-5.2+ thinking responses are slow and need headroom, so OmniRoute now sets a sensible default max_tokens and a longer timeout for them. (#4255 — thanks @dhaern)
• fix(antigravity): default includeThoughts for modern Gemini models — modern Gemini models on the Antigravity path now default to including thoughts so reasoning isn't silently dropped. (#4180 — thanks @dhaern)
• fix(provider-registry): add correct contextLength to theoldllm models — fills in accurate context-window sizes for theoldllm's models. (#4184 — thanks @herjarsa)
• fix(models): expose combo model token limits — /v1/models now reports token limits for combo models. (#4189 — thanks @megamen32)
• fix(combo): keep the passthrough quota fallback scoped — prevents the passthrough quota fallback from leaking across unrelated targets. (#4194 — thanks @Svetznaniy33)
• fix(combo): opt proactive-fallback compression into the TV1 bail-out (no silent target drop) — proactive-fallback compression now participates in the TV1 bail-out so a target is never silently dropped. (#4228)
• fix(compression): show engine preview output — the Compression Studio preview now renders the engine's output. (#4128 — thanks @megamen32)
• fix(compression): harden engines against I/O failures and misconfig (F5.3) — compression engines degrade gracefully on I/O errors and bad configuration instead of throwing. (#4198)
• fix(compression): harden RTK raw-output redaction + ReDoS guard for custom filters (F5.3) — broadens RTK raw-output redaction and adds a ReDoS guard for user-supplied filter patterns. (#4203)
• fix(compression): bound mcpAccessibility maxTextChars on the live read path — the live read path now clamps maxTextChars so a small value can't make tools disappear. (#4206)
• fix(dashboard): data tables paint an opaque surface so the grid doesn't bleed through — data tables now render on an opaque surface, fixing the grid wallpaper showing through. (#4233)
• fix(dashboard): make the provider card hover visible (was ~1% opacity) — the provider-card hover state was effectively invisible; it now has a visible surface. (#4214)
• fix(vscode): sanitize implicit editor context — redacts sensitive filenames/keywords from the implicit VS Code editor context before it's sent upstream. (#4124 — thanks @zhiru)
• fix(build): raise the Node heap for the local next build to stop OOM/stall — bumps the build-time heap so the local production build no longer OOMs or stalls. (#4171)
• fix(mitm): TPROXY OUTPUT-based recipe for local traffic (validated e2e on VPS) — switches the TPROXY rules to an OUTPUT-chain recipe so locally-originated traffic is captured; validated end-to-end on the VPS. (#4156)
• fix(mitm): forward anti-loop — put the bypass-marked socket on the Agent (decrypt 4d) — places the bypass-marked socket on the HTTP Agent so OmniRoute's own forwarded traffic never re-enters the capture loop; VPS-validated. (#4229)
• fix(free-tiers): retire dead-tier hasFree, round the headline to ~1.6B, regenerate the per-provider table — drops dead free tiers from the headline math and regenerates the per-provider free-tier table. (#4142)
• fix(free-tiers): retire 4 re-verified-dead free tiers, flag iflytek/sparkdesk ToS, clarify monsterapi one-time — removes four confirmed-dead free tiers and annotates ToS/one-time caveats. (#4152)

🧪 Tests

• test(sse): guard the Antigravity _toolNameMap cloak map through the request-capture round-trip — follow-up to #4091: the generic capture fix in createPreparedRequestLogger().body() (#4153) re-attaches the non-enumerable _toolNameMap that the request-inspector drops when it rebuilds the upstream body via JSON.parse(JSON.stringify(...)), but the only regression test covered the native-Claude OAuth cloak (PascalCase aliases). The Antigravity cloak differs — cloakAntigravityToolPayload suffixes custom tools with _ide (workspace_read → workspace_read_ide), leaves native tools untouched, and returns the reverse map separately — so a refactor of providerRequestLogging.ts or the executor could silently re-break Antigravity tool dispatch without tripping the Claude test. Adds a dedicated regression test driving the real cloakAntigravityToolPayload through the capture round-trip and asserting the _ide reverse map survives, stays non-enumerable (never re-serializes upstream), and that all-native traffic produces no spurious map (verified failing with the #4153 re-attach removed). No production change. (#4181 — thanks @hertznsk)
• test(chatcore): dedicated unit tests for 6 leaves + wire into stryker mutate (QG v2 Fase 9 T5 Fase 3) — adds focused unit tests for 6 chatCore leaf helpers and enrolls them in mutation testing. (#4218)
• test(chatcore): telemetry / memory-skills / semantic-cache tests + wire 2 into stryker (QG v2 Fase 9 T5 Fase 3) — new tests for the telemetry, memory-skills and semantic-cache leaves, two of which are added to the mutation set. (#4222)
• test+ci(chatcore): semanticCache HIT-path fixture (15/15 mutate) + 350min budget headroom — closes the semantic-cache HIT path to a full 15/15 mutation score and gives the nightly auth/accountFallback batches more budget headroom. (#4225)
• test(compression): close F5.1 coverage gaps (replay reducer, live accumulator, StatusDot) — fills the remaining F5.1 compression coverage gaps. (#4192)
• test(db,sse): de-flake db-backup + chatcore streaming timing assertions — stabilizes two timing-sensitive tests (fire-and-forget backup completion + a streaming race). (#4132)
• test: align stale integration tests surfaced post-v3.8.28 on main — realigns integration tests that drifted after the v3.8.28 merge. (#4129)

📝 Maintenance

• refactor(sse): split chatCore.ts pure helpers into chatCore/ modules (−561 LOC) — extracts pure helpers out of the chatCore god-file into dedicated modules (Onda 3). (#4159)
• refactor(chatcore): extract passthrough/header/telemetry helpers (QG v2 Fase 9 T5 C2-C3-C5) — further chatCore decomposition. (#4188)
• refactor(chatcore): extract combo/proxy context cache + semaphore helpers (QG v2 Fase 9 T5 C6-C7) — continues the chatCore split. (#4193)
• refactor(combo): god-file split pilot — types + validateQuality + predicates (QG v2 Fase 9 T5 D1-D3) — first slice of the combo.ts decomposition. (#4162)
• refactor(combo): god-file split part 2 — shadow + sorters + structure (QG v2 Fase 9 T5 D4-D6) — continues the combo.ts split. (#4175)
• refactor(combo): god-file split part 3 — auto strategy (QG v2 Fase 9 T5 D8) — extracts the auto strategy from combo.ts. (#4186)
• refactor(combo): extract round-robin sticky state to combo/rrState.ts (D7a) — moves round-robin sticky state into its own module. (#4196)
• refactor(combo): extract the reset-aware quota block to combo/quotaStrategies.ts (D7b) — moves the reset-aware quota strategies into their own module. (#4204)
• refactor(compression): remove vestigial SLM seam + dead deprecated alias — drops dead compression code. (#4253)
• chore(compression): remove vestigial reconstructCcr/SessionDedup round-trip helpers — removes unused round-trip helpers. (#4226)
• chore(compression): remove dead exports + fix stale llmlingua docs — prunes dead exports and corrects stale LLMLingua docs. (#4223)
• chore(build): build + ship the TPROXY native addon in the standalone (prebuilds 4e) — bundles the native TPROXY addon prebuilds into the standalone build. (#4236)
• chore(ci): add quota + 6 covered chatCore leaves to stryker mutate (QG v2 Fase 9 T5 Fase 3 follow-up) — enrolls more covered leaves into mutation testing. (#4209)
• chore(ci): re-add 8 combo split leaves to stryker mutate + expand nightly batch-matrix 3→5 (QG v2 Fase 9 T5 Fase 3) — restores mutation coverage for the split combo leaves and widens the nightly matrix. (#4205)
• chore(quality): close v3.8.28 cycle gate drift (re-baseline + nightly-mutation scope) — reconciles quality-gate baselines after the v3.8.28 cycle. (#4135)
• ci(mutation): split nightly into 3 parallel batches to fit the 180min budget (QG v2 Fase 9 T0) — parallelizes the nightly mutation run. (#4150)
• ci(mutation): restore cold-seed timeout headroom (a/b lost in #4225 squash) + extend to c/d/g/h — restores and extends per-batch cold-seed timeouts. (#4258)
• ci(docs): harden the fabricated-docs checker + enforce --strict (QG v2 Fase 9 T9) — tightens the anti-hallucination docs checker. (#4149)
• ci: derive the oasdiff base-ref from the package version + flag the mutation-toolchain regression — fixes the OpenAPI-diff base-ref and surfaces a mutation-toolchain regression. (#4134)
• docs(ci): correct the mutation-gate note (no regression — stryker -c is --concurrency); record Task 12 GO — corrects a misread of the stryker flag and records the spike GO. (#4138)
• docs(api): document the /api/v1/ws chat WebSocket endpoint in openapi.yaml — adds the WebSocket chat endpoint to the OpenAPI spec. (#4215)
• docs(readme): expand Acknowledgments into a themed, star-counted credits hall — reworks the README acknowledgments section. (#4195)
• style(dashboard): shrink the identity grid cell 46px → 32px (~30% smaller) — tightens the identity grid density. (#4143)

🔧 Dependencies

• deps: bump the production group with 5 updates — routine production-dependency bumps. (#4121)
• chore(deps): bump github/codeql-action from 3 to 4 — CI action update. (#4120)
• chore(deps): bump actions/setup-python from 5 to 6 — CI action update. (#4119)

---

[3.8.28] — 2026-06-17

✨ New Features

• feat(providers): add OrcaRouter (OpenAI-compatible routing gateway) — OrcaRouter is now registered as an API-key provider. Its adaptive router is exposed as orcarouter/auto (smart routing across 150+ upstream models), alongside a curated flagship set (GPT-5.5, Gemini 3.5 Flash, Claude Opus 4.8, Grok 4.3, DeepSeek V4 Pro, MiniMax M2.7, Qwen3.7 Max). passthroughModels is enabled so any OrcaRouter model id works. OpenAI-compatible endpoint (https://api.orcarouter.ai/v1), Bearer (sk-orca-…) auth — no custom executor or translator required. (#4070 — thanks @jinhaosong-source)
• feat(providers): add Wafer AI (Anthropic-compatible, Bearer auth) — Wafer AI is now a built-in provider speaking the Anthropic Messages format with Bearer authentication, registered with its model catalog so it works out of the box. (#4098 — thanks @diegosouzapw)
• feat(cli): omniroute launch — zero-config Claude Code launcher — a new CLI subcommand that boots OmniRoute (if not already running) and launches Claude Code pre-wired to it, with no manual env/settings editing. (#4097 — thanks @diegosouzapw)
• feat(api): exact offline token counting for the count_tokens fallback via tiktoken — the local count_tokens fallback now uses a real tiktoken (BPE) tokenizer for exact offline counts instead of a heuristic estimate, so token budgeting is accurate even when no upstream count endpoint is reachable. (#4087 — thanks @diegosouzapw)
• feat(sse): Claude Code quota-probe bypass + command meta-request helpers — ported from free-claude-code: OmniRoute now recognizes Claude Code's quota-probe and command meta-requests and serves them locally instead of burning an upstream call, reducing wasted quota during CLI sessions. (#4083 — thanks @diegosouzapw)
• feat(sse): generic 400 field-downgrade retry + Groq field stripping — when an upstream rejects a request with 400 because of an unsupported field, OmniRoute now strips the offending field and retries (a generic downgrade path), with Groq-specific field stripping wired in. Aligned with the existing context_management retry handling. (#4096 — thanks @diegosouzapw)
• feat(sse): delegated Anthropic Context Editing — relay coverage + 400-fallback — extends the Claude server-side Context Editing delegation (#4021) with broader relay coverage and a 400-fallback so a request the upstream rejects for the context-management beta degrades gracefully instead of failing. (#4065 — thanks @diegosouzapw)
• feat(compression): record per-engine Context Editing telemetry — the compression pipeline now records a context-editing engine entry so the dashboard attributes server-side Context Editing savings alongside the local compression engines. (#4062 — thanks @diegosouzapw)
• feat(compression): RTK learn/discover (sample source + API + UI) — the rule-based RTK compression engine gains a learn/discover workflow: sample a source, surface candidate rules through a new API, and review/apply them in the dashboard. (#4088 — thanks @diegosouzapw)
• feat(dashboard): 2026-06-17 free-tier refresh — honest catalog, uncapped + boost tiers, Layout A budget table — the free-tier page was refreshed with an honest, deep-researched catalog (pooled/realistic figures rather than inflated 24/7 RPM math), new recurring-uncapped and boost tiers, new providers, and a KPI + budget table (Layout A). (#4089 — thanks @diegosouzapw)
• feat(dashboard): Combo Studio connection-cooldown badge (U1b Slice 2) — the Combo Live cascade now surfaces each connection's cooldown state as a badge, complementing the circuit-breaker badge shipped in 3.8.27. (#4068 — thanks @diegosouzapw)
• feat(mitm): attribute intercepted requests to the originating process (Gap 1) — the Traffic Inspector now resolves each intercepted connection back to the originating local process (via /proc), so captured traffic can be attributed to the app that produced it. (ProxyBridge-inspired hardening.) (#4085 — thanks @diegosouzapw)
• feat(mitm): capture-pipeline self-test route (Gap 12) — a diagnostic route that exercises the MITM capture pipeline end-to-end so operators can confirm interception is working without crafting a real upstream call. (#4093 — thanks @diegosouzapw)
• feat(mitm): loop-guard self-check + verbosity control in server.cjs (Gaps 14+15) — the MITM proxy gains a self-referential loop guard (so it never proxies its own traffic into an infinite loop) and a MITM_VERBOSE routing-decision log level. (#4101 — thanks @diegosouzapw)
• feat(agent-bridge): portable JSON import/export of config (Gap 4) — the Agent Bridge / MITM configuration can now be exported to and imported from a portable JSON file, so a working setup can be backed up or moved between machines. (#4094 — thanks @diegosouzapw)

🐛 Fixed

• fix(ws): start the LiveWS sidecar with cwd at the package root (global/systemd installs) — the standalone LiveWS launcher (scripts/start-ws-server.mjs) re-spawns itself with node --import tsx <self> but did not set cwd. When the WebSocket sidecar was launched from outside the package directory — a global npm/homebrew install, or a systemd/launchd unit started from $HOME — Node could not resolve the tsx package (ERR_MODULE_NOT_FOUND: Cannot find package 'tsx'), and even from the package directory tsx could not resolve the tsconfig @/* path aliases (e.g. @/types/databaseSettings), so the sidecar never booted. The spawn now pins cwd to the package root (the directory above scripts/, where package.json + tsconfig.json live), which resolves both tsx discovery and the @/* aliases regardless of launch directory. (#4055 — thanks @Rahulsharma0810)
• fix(dashboard): Logs page auto-refresh now works in embedded/proxied dashboards — the Request Logger gated each auto-refresh tick on a static document.visibilityState === "visible" read. Hosts that report a permanent non-"visible" state without ever firing a visibilitychange event (Docker dashboard wrappers, embedded webviews) froze auto-refresh entirely — only the manual Refresh button worked, a regression from 3.8.24's unconditional polling. The pause is now event-driven and fail-open: polling starts enabled and only pauses after a real visibilitychange → hidden transition (still preserving the backgrounded-tab optimization for normal browser tabs). (#4054 — thanks @tjengbudi)
• fix(docker): raise the build-stage Node heap to stop the production-build OOM — the Docker builder stage ran npm run build with V8's default heap ceiling (~2 GB). After #4052 forced the heavier webpack engine (Turbopack panics on this Next.js version), the production optimization pass exceeded that ceiling and the build died with FATAL ERROR: … JavaScript heap out of memory at [builder] npm run build. The builder stage now sets NODE_OPTIONS=--max-old-space-size (default 4096 MB, overridable via --build-arg OMNIROUTE_BUILD_MEMORY_MB=…) before the build; the value propagates to the spawned next build. Build-only — the runtime heap (OMNIROUTE_MEMORY_MB on the runner stage) is unchanged. (#4076 — thanks @kamenkadmitry)
• fix(dashboard): "Update Available" banner reappears reliably across Docker/npm/desktop installs — the home-page banner is gated on GET /api/system/version's updateAvailable, which derived the latest version ONLY from npm info omniroute version --json via the npm CLI binary. When that binary is absent from the runtime PATH (Docker/desktop/locked-down installs) or the registry is unreachable, the call returned null → updateAvailable=false → the banner silently never rendered even when a newer release existed. The route now resolves the latest version through resolveLatestVersion(): the fast npm CLI path first, then an npm-binary-free fallback over the registry HTTP API (registry.npmjs.org/omniroute/latest), and a logged warning instead of silent degradation when both fail. Version comparison was also hardened to tolerate v-prefixed and pre-release version strings. (#4100)
• fix(sse): route image requests only to confirmed-vision combo targets — a combo could route an image-bearing request to a member that doesn't actually support vision. Routing now requires supportsVision === true (plus a model-id heuristic) before sending images to a target, so multimodal requests land only on members that can handle them. (#4071 — thanks @diego-anselmo)
• fix(security): injection guard respects the INJECTION_GUARD_MODE DB feature flag — the prompt-injection guard ignored the database feature-flag, so operators couldn't change its mode at runtime; it now reads the flag and honors the configured mode. (#4077 — thanks @zhiru)
• fix(ws): proxy LAN /live-ws upgrades and warn on an unset JWT_SECRET — WebSocket upgrade requests arriving over the LAN proxy path were not forwarded to the LiveWS sidecar; they are now proxied correctly, and the server logs a clear warning when JWT_SECRET is unset. (#4079 — thanks @Rahulsharma0810)
• fix(dev): force webpack in the custom dev server (Turbopack 16.2.x panics) — the custom dev server now forces the webpack engine because Turbopack panics on this Next.js version, so npm run dev boots reliably. (#4092 — thanks @chirag127)
• fix(auto): resolve built-in auto/* catalog combos — referencing a built-in auto/* combo returned a premature 400 because the catalog entry wasn't resolved; the built-in auto catalog is now resolved before validation so those combos work. (#4058 — thanks @megamen32)
• fix(sse): friendly 413 message for ChatGPT-web payload-too-large — an oversized ChatGPT-web payload returned an opaque error; it now returns a clear 413 with a human-readable message. (#4080 — thanks @diegosouzapw)
• fix(ws): warm the SSE auth import on LiveWS startup; relocate the boot test to integration — the LiveWS sidecar now pre-imports the SSE auth module at startup to avoid a first-request stall, and its boot test was moved to the integration suite. (#4063 — thanks @diegosouzapw)
• fix(mitm): crash-safe system-state teardown + socket timeouts (ProxyBridge-inspired hardening) — the MITM proxy could leave the host's system proxy settings applied if it crashed mid-teardown, and long-lived tunnels could leak as half-open sockets. Teardown is now crash-safe (system state is always restored) and proxied sockets get an idle timeout (MITM_IDLE_TIMEOUT_MS, default 60s). (#4084 — thanks @diegosouzapw)
• fix(responses): clear the /v1/responses keep-alive timer on cancel/abort — a cancelled or aborted /v1/responses stream left its keep-alive timer running, leaking a timer and burning CPU; the timer is now cleared on cancel/abort. (#4105 — thanks @artickc)
• fix(usage): reap orphaned pending-request details (unbounded memory leak) — pending-request detail entries whose request never completed accumulated without bound; they are now reaped, closing a slow memory leak. (#4107 — thanks @artickc)
• fix(auth): prune expired entries from the login brute-force guard map (unbounded growth) — the login brute-force guard map grew without bound because expired entries were never removed; expired entries are now pruned. (#4111 — thanks @artickc)
• fix(logger): hard-cap the error-dedup map to bound memory under unique-message bursts — a burst of unique error messages could grow the dedup map without limit; it is now hard-capped. (#4113 — thanks @artickc)
• fix(circuit-breaker): enforce MAX_REGISTRY_SIZE (declared but never applied) — the circuit-breaker registry declared a maximum size that was never enforced, so it could grow unbounded; the cap is now applied. (#4114 — thanks @artickc)
• fix(webhook): clear the abort timer in finally to avoid dangling timers on fetch error — a webhook dispatch that threw before clearing its abort timer left the timer dangling; it is now cleared in a finally block. (#4115 — thanks @artickc)
• fix(combo): detach the per-target listener from the shared hedge abort signal — combo hedging attached a per-target listener to a shared abort signal without detaching it, leaking listeners across requests; the listener is now detached. (#4116 — thanks @artickc)
• fix(timers): unref background interval timers so they don't block clean shutdown — long-lived background interval timers kept the event loop alive and blocked a clean process exit; they are now unref'd. (#4117 — thanks @artickc)

⚡ Performance

• perf(registry): precompute the model→provider index in parseModelFromRegistry — model→provider lookups now use a precomputed index instead of scanning the registry on every call. (#4110 — thanks @artickc)
• perf(obfuscation): cache per-word regexes instead of recompiling every request — the obfuscation pass now caches its per-word regexes rather than recompiling them on each request. (#4109 — thanks @artickc)
• perf(stream): use structuredClone instead of a JSON round-trip for per-chunk reasoning split — the per-chunk reasoning split now clones with structuredClone rather than JSON.parse(JSON.stringify(...)). (#4108 — thanks @artickc)
• perf(gemini): cache the reasoning close-tag regex instead of recompiling per token — the Gemini reasoning close-tag regex is now compiled once and reused instead of per token. (#4106 — thanks @artickc)

📝 Maintenance

• ci(quality): flip the TIA impacted-unit-tests gate from advisory to blocking (Fase 9) — the test-impact-analysis gate that runs the unit tests impacted by a diff is now blocking on PRs. (#4069 — thanks @diegosouzapw)
• ci(quality): dedup the doubly-run check:docs-sync + record the validated ROI backlog (Fase 9) — check:docs-sync was running twice in CI; the duplicate was removed and the validated quality-gate ROI backlog recorded. (#4099 — thanks @diegosouzapw)
• docs(quality-gates): reconcile the gate inventory with ci.yml + add the ROI rationalization backlog — the quality-gate inventory doc was reconciled against the actual CI jobs and a rationalization backlog added. (#4095 — thanks @diegosouzapw)
• test(infra): isolate DATA_DIR per test process; raise Stryker concurrency 1→4 — test processes now get an isolated DATA_DIR (no shared-DB cross-talk) and the mutation runner's concurrency was raised. (#4078 — thanks @diegosouzapw)
• test(dashboard): smoke e2e for the Combo Live Studio page — adds a Playwright smoke test covering the Combo Live Studio page. (#4075 — thanks @diegosouzapw)
• docs(compression): document LLMLingua optional deps + on-demand install — documents the optional LLMLingua dependencies and how they are installed on demand. (#4061 — thanks @diegosouzapw)
• chore(deps): freeze @huggingface/transformers in dependabot (hard-pin) — the transformers dependency is hard-pinned and frozen in dependabot to protect the VPS-validated LLMLingua + memory-embeddings stack from a breaking major bump. (#4066 — thanks @diegosouzapw)
• chore(docs): update the Discord invite link to a non-expiring one — replaces the expiring Discord invite with a permanent link. (#4067 — thanks @diegosouzapw)
• chore(docs): document the new MITM env vars + reconcile the env-doc contract — documents MITM_IDLE_TIMEOUT_MS and MITM_VERBOSE in .env.example + ENVIRONMENT.md, allowlists the framework-internal TURBOPACK and the Claude Code ANTHROPIC_AUTH_TOKEN, and relocates/prunes stale provider/guide docs. (thanks @diegosouzapw)

🔧 Dependencies

• deps: bump the development group with 10 updates — routine dependabot dev-dependency bumps. (#4051)
• deps(electron): bump electron 42.4.0 → 42.4.1 — (#4049)
• ci(deps): bump actions/setup-node 4 → 6 — (#4048)
• ci(deps): bump actions/cache 4.3.0 → 5.0.5 — (#4047)
• ci(deps): bump actions/github-script 7 → 9 — (#4046)
• ci(deps): bump ossf/scorecard-action 2.4.0 → 2.4.3 — (#4045)
• ci(deps): bump actions/upload-artifact 4 → 7 — (#4044)

---

[3.8.27] — 2026-06-17

✨ New Features

• feat(combos): advertise combo capabilities (multimodal / reasoning / caching) on the import surfaces — importing a combo package into a client (LobeHub / OpenCode / VS Code, via /v1/combos and the VS Code combo catalog) no longer requires manually enabling multimodal/image-input, reasoning, and caching afterwards. projectCombo now attaches a registry-derived capabilities block, gated conservatively: multimodal/reasoning are advertised only when every concrete model step proves the capability (an unprovable nested combo-ref drops them, since the strategy may route to any member), and caching reflects the combo's explicit Context-Cache-Protection setting (no surprise prompt-cache cost). The public /v1/combos default projection (#2300) is unchanged unless the caller opts in. (#3979 — thanks @xenstar)
• feat(sse): delegated Anthropic Context Editing for Claude (clear_tool_uses) — Claude requests can now offload context trimming to Anthropic's server-side context-management API (beta context-management-2025-06-27, clear_tool_uses_20250919), pruning stale tool-use turns upstream instead of locally. Claude-only by nature (the edit runs server-side); multi-provider context trimming remains the job of the local compression engines. (#4021 — thanks @diegosouzapw)
• feat(sse): real LLMLingua-2 ONNX compression engine (stable) — the LLMLingua-2 prompt-compression engine is now a real local ONNX model (TinyBERT default, transformers.js + tfjs), promoted to stable after VPS validation, replacing the previous placeholder. (#4014 — thanks @diegosouzapw)
• feat(compression): capture per-engine analytics + Lite schema fix — the compression pipeline now persists a per-engine breakdown for historical analytics so the dashboard can attribute savings to each engine in a stacked pipeline, and a Lite-schema mismatch was corrected. (#4018 — thanks @diegosouzapw)
• feat(dashboard): real circuit-breaker state in the Combo Live cascade (U1b) — the Combo Live cascade view now surfaces each provider's real circuit-breaker state (CLOSED / OPEN / HALF_OPEN) as a badge, read live from /api/monitoring/health, instead of inferring health from request outcomes. (#4029 — thanks @diegosouzapw)
• feat(openai): honor a custom base URL in model discovery + complete openai/codex pricing — OpenAI-format providers configured with a custom base URL now have that URL honored during model discovery (not just inference), and the openai/codex pricing table was completed. Discovery is routed through the SSRF-guarded outbound fetch. (#4005 — thanks @artickc)
• feat(observability): capture actual upstream provider requests — the request inspector now records the exact payload sent to the upstream provider (post-translation), so you can see what OmniRoute actually dispatched rather than only the client's original request. (#3941 — thanks @rdself)
• feat(providers): provider auth visibility controls — adds controls to show/hide provider auth details in the dashboard so credentials can be revealed only when needed. (#3953 — thanks @rdself)
• feat(providers): model search filter on the provider dashboard — the provider dashboard gains a search filter to quickly narrow a provider's model list. (#3950 — thanks @felipesartori)
• feat(compression): Indonesian caveman rules + language pack — adds an Indonesian "caveman" rule set and language pack to the rule-based compression engine. (#3975 — thanks @Veier04)
• feat(dashboard): sidebar group separator toggles — the dashboard sidebar can now toggle group separators for a cleaner navigation layout. (#3971 — thanks @rdself)
• feat(api): local @@om-usage command for cached per-key usage — API clients can send a message that is exactly @@om-usage to retrieve cached Claude-style usage data locally, without forwarding the prompt to an upstream provider. Gated by a new per-key allowance flag. (#4034 — thanks @Witroch4)

🐛 Fixed

• fix(opencode): forward the OpenCode session id to the upstream regardless of how the user named the provider — the OpencodeExecutor forwarded the x-opencode-session/request/project/client headers, but the OpenCode CLI only emits those when the configured providerID starts with "opencode". A user who adds OmniRoute as a custom provider (e.g. "omniroute") makes the CLI send x-session-affinity / X-Session-Id instead (both carry the same session id), which the executor never read — so the session-metadata forwarding was effectively dead code for the realistic provider-naming case. The opencode-family executor now falls back to x-session-affinity / X-Session-Id and maps it onto x-opencode-session when the client didn't send the header directly, so session continuity to the opencode.ai upstream works for any provider name (a direct x-opencode-session still wins). Scoped to this executor only — the generic DefaultExecutor intentionally does not do this, to avoid leaking the client session id to arbitrary third-party upstreams. (#4022 — thanks @pizzav-xyz)
• fix(guardrails): Vision Bridge no longer drops the image when the describe call fails (Nvidia NIM "Image unavailable") — the Vision Bridge is enabled by default and engages for any model whose vision capability OmniRoute can't prove from the registry (supportsVision !== true, which includes uncatalogued models that resolve to null). When the per-image describe call failed (e.g. no vision model configured), it replaced the image with the literal text [Image N]: (unavailable) and dropped the original image_url — so a genuinely vision-capable upstream (Nvidia NIM) received text only and answered "Image unavailable. Cannot provide description without visual data." A describe failure is no longer destructive: replaceImageParts now receives null for failed images and preserves the original image part so the upstream can still see it (successful describes still replace the image with the text description; meta.descriptions observability is unchanged). (#4012 — thanks @daniij)
• fix(kiro): preserve finish_reason: "tool_calls" on the Kiro streaming path — streaming tool-call requests through the Kiro (Responses API) provider had their terminal finish_reason reported as "stop" instead of "tool_calls", so agent clients (Hermes) treated the tool-call turn as a finished turn, never ran the tool, and the next request failed with HTTP 400 on the incomplete tool state. convertKiroToOpenAI's terminal messageStopEvent/done branch hardcoded finish_reason: "stop" regardless of whether the stream had emitted toolUseEvents. The translator now records state.sawToolUse when a tool-use chunk is emitted and reports finish_reason: "tool_calls" on the terminal chunk (and in state.finishReason) whenever the stream produced tool calls. The non-streaming path was already correct. (#3980 — thanks @lordavadon2)
• fix(resilience): respect connection cooldown stored as a numeric epoch — the router kept dispatching to connections still inside their rate-limit cooldown because rate_limited_until (a TEXT column) was persisted as a raw epoch number, which SQLite coerced to a string like "1781696905131.0" that new Date(...) parsed as NaN, so the cooling connection was never skipped. The cooldown read predicates now normalize numeric-epoch strings via a shared cooldownUntilMs() helper; ISO behavior is unchanged. (#3995 — thanks @diegosouzapw)
• fix(providers): fetch the live /models catalog for LLM7 and BytePlus — importing an LLM7 or BytePlus key surfaced only a small, outdated hardcoded list because neither provider was classified by any live-fetch branch of the model-import route. Both are now in NAMED_OPENAI_STYLE_PROVIDERS, so the route probes <baseUrl>/models with the key and serves the live catalog, falling back to the local catalog only when the upstream fetch fails. (#3996 — thanks @FerLuisxd / @diegosouzapw)
• fix(dashboard): logs auto-refresh reads live visibility, not a stale mount ref — the Logs page never auto-refreshed when the tab loaded in the background because the auto-refresh interval gated each tick on a visibility ref seeded once at mount; the tick now reads the live document.visibilityState, so polling self-heals as soon as the tab is visible while still pausing when genuinely hidden. (#3997 — thanks @tjengbudi / @diegosouzapw)
• fix(combo): shuffle the strict-random fallback remainder to spread load — with the strict-random strategy a persistently-failing model was retried on essentially every request because only the deck-selected slot 0 was shuffled while the fallback remainder stayed in fixed priority order; the remainder is now shuffled too, so fallback load (and recovery from a failing target) spreads evenly across healthy peers. (#3998 — thanks @KeNJiKunG / @diegosouzapw)
• fix(claude): forward the client tool-search-tool-2025-10-19 anthropic-beta on the Claude OAuth path — with deferred tools active, Claude Code negotiates the tool-search-tool-2025-10-19 beta, but OmniRoute dropped it on both Claude code paths, so the claude.ai backend rejected every deferred-tool request with 400 Tool reference not found. A new allowlist-merge (mergeClientAnthropicBeta) now unions the client's negotiated beta into the outbound set on both paths, appending only allowlisted client betas (preserving the #3415 fix). (#3999 — thanks @huohua-dev / @diegosouzapw)
• fix(executor): strip stream_options on non-streaming requests (NVIDIA NIM 400) — clients that send stream_options: { include_usage: true } regardless of stream (e.g. the OpenAI Python SDK) had it passed through untouched on non-streaming calls, and NVIDIA NIM rejected it with 400 "Stream options can only be defined when stream=True". DefaultExecutor.transformRequest now strips stream_options whenever stream is false; the streaming injection path is unchanged. (#4000 — thanks @andrea-kingautomation / @daniij / @diegosouzapw)
• fix(sse): guard model-less registry entries in getUnsupportedParams (mimocode) — a registry entry without a model map (mimocode) threw when computing unsupported params; the lookup now guards the model-less case so request validation no longer crashes. (#4015 — thanks @diegosouzapw)
• fix(perplexity-web): parse the schematized diff_block stream so answers aren't empty — Perplexity web streamed its answer as RFC-6902 diff_block patches that OmniRoute didn't apply during the PENDING phase, so responses came back empty; the parser now applies the patches and materializes the text only on COMPLETED. (#4001 — thanks @artickc)
• fix(default-executor): honor a custom providerSpecificData.baseUrl for OpenAI-format providers — OpenAI-format providers configured with a custom base URL had it ignored on the inference path; the default executor now honors providerSpecificData.baseUrl so requests reach the configured endpoint. (#4002 — thanks @artickc)
• fix(live-ws): bridge LiveWS sidecar events to the dashboard — events emitted by the LiveWS sidecar were not reaching the dashboard; they are now bridged so live websocket activity is visible. (A cookie-auth regression in the sidecar's auth-token parsing was also corrected.) (#4004 — thanks @megamen32)
• fix(qwen-web): cookie validation false-positive — check the response body for a user object — Qwen web cookie validation reported a valid cookie as invalid; it now inspects the response body for the user object instead of relying on the status code alone. (#3958 — thanks @thezukiru)
• fix(vision-bridge): force the bridge for tokenrouter deepseek models — tokenrouter DeepSeek models are now forced through the Vision Bridge so image inputs are handled correctly. (#3946 — thanks @WormAlien)
• fix(api): return 400 (not 500) for malformed JSON on /api/auth/login — a malformed JSON body on the login endpoint returned an opaque 500; it now returns a proper 400. (#4031 — thanks @rdself)
• fix(dashboard): Playground Compare tab loading + HTTP method guard — the Playground Compare tab failed to load; the loading path was fixed and an HTTP method guard added. (#4024 — thanks @rdself)
• fix(proxy): gate the control-plane proxy direct fallback behind a feature flag (fail-closed) — the direct-connection fallback for control-plane ops when a pinned proxy is unreachable is now gated behind a feature flag and fails closed, so a pinned proxy is never silently bypassed unless explicitly allowed. (#3963 — thanks @rdself)
• fix(db): persist backup retention days — the backup retention-days setting was not persisted across restarts; it is now stored durably. (#3970 — thanks @rdself)
• fix(dashboard): refine the provider quota card display — the provider quota card layout was refined for clearer quota/usage presentation. (#3969 — thanks @rdself)
• fix(dashboard): refine compression settings, storage labels, and sidebar grouping — polishes the compression-settings UI, clarifies storage labels, and tidies the sidebar grouping. (#4033 — thanks @rdself)

🔒 Security & Hardening

• fix(security): eliminate a polynomial ReDoS in the combo <omniModel> tag regex — comboAgentMiddleware's cache-tag pattern wrapped the tag in an unbounded newline run ((?:\n|\r)*), making .test() / .replace() run in O(n²) on inputs with many newlines (CodeQL js/polynomial-redos). The detection pattern now matches only the core <omniModel>…</omniModel> and the global strip pattern bounds the surrounding newline runs, keeping it linear; detection / extraction / multi-tag stripping behavior is unchanged. (#3982 — thanks @diegosouzapw)
• ci(security): harden workflows — artipacked persist-credentials, cache-poisoning, SC2086 — GitHub Actions workflows were hardened against the artipacked persist-credentials leak and cache-poisoning, and shell-quoting (SC2086) issues were fixed. (#3965 — thanks @diegosouzapw)
• ci(quality): flip require-tighten + osv + Trivy to blocking (cycle-end) — the per-module require-tighten check and the OSV / Trivy scanners moved from advisory to blocking for the v3.8.27 cycle close, so new dependency or coverage regressions fail CI. (#3984 — thanks @diegosouzapw)
• chore(deps): dependabot security bumps + drop unused gray-matter — applies a batch of Dependabot security bumps and removes the unused gray-matter dependency from the tree. (#4036 — thanks @diegosouzapw)
• chore(deps): automated dependency bumps — Dependabot upgraded the production dependency group (13 updates), vite, form-data, and the npm_and_yarn group. (#3915, #3942, #3943, #3944 — thanks @dependabot)

🧹 Internal / Quality / Docs

• feat(ci): Quality Gate v2 — Onda 0 + Onda 1 — first two waves of the Quality Gate v2 program: gate flips, test-impact analysis (TIA), SAST, DAST-smoke, and mutation-testing infrastructure. (#4016 — thanks @diegosouzapw)
• refactor: modularize the provider registry into individual provider plugins — providerRegistry.ts was split into individual per-provider plugin modules (non-stacked). A forward-fix restored the byteplus + mimocode modules dropped by the move. (#3993 — thanks @oyi77 / @diegosouzapw)
• refactor: modularize schemas (non-stacked) — the request/response schema definitions were split into individual modules to reduce file size and improve maintainability. (#3988 — thanks @oyi77)
• fix: restore unit regressions dropped by the lossy schema/registry modularizations — the schema/registry modularizations (#3988, #3993) silently dropped internal logic covered by unit tests; this PR restores the regressed units. (#4030 — thanks @diegosouzapw)
• refactor(dashboard): settings UI layout + API Keys naming — the settings UI layout was reorganized and the "API Keys" naming clarified. (#4020 — thanks @rdself)
• 大量UI显示和i18n优化 (dashboard UI display + i18n improvements) — a batch of dashboard UI-display refinements and i18n string improvements. (#3973 — thanks @rdself)
• fix(ci): scope TIA to node:test unit files only — test-impact analysis was matching files the node:test runner doesn't execute, producing 99 false failures; the TIA glob now mirrors the test:unit glob exactly. (#4035 — thanks @diegosouzapw)
• fix(ci): electron-release publish-npm needs contents: write — the reusable npm-publish job invoked by the electron release lacked contents: write, causing a v3.8.26 startup_failure; the permission was granted. (#3966 — thanks @diegosouzapw)
• test(opencode-plugin): ESM default-export test (drop the stale CJS bundle test) — replaces the stale CJS bundle test with an ESM default-export test, following up the #3883 ESM-only migration. (#3967 — thanks @diegosouzapw)
• fix(ci): Fix promptfoo security-assertion parsing — the promptfoo (DAST/security eval) assertion parser was corrected so security assertions are read reliably. (#4032 — thanks @rdself)
• docs(troubleshooting): note that the MITM proxy cannot intercept Windows-host apps under WSL — documents that the MITM proxy running inside WSL cannot intercept traffic from apps on the Windows host. (#4003 — thanks @diegosouzapw)
• chore(quality): maintenance roll-up — assorted quality-gate hygiene that does not change runtime behavior: re-baseline validation.ts for the #3958 qwen body-check, allowlist the socks dependency declared by #4004, ignore jscpd major bumps (the v5 Rust rewrite breaks the pinned duplication gate), untrack an accidentally-committed root node_modules symlink (and gitignore it), rehome the #3972 logs auto-refresh test so a runner collects it, and open the v3.8.27 development cycle. (thanks @diegosouzapw)

---

[3.8.26] — 2026-06-15

✨ New Features

• feat(media): Vertex AI (Google) speech, transcription, music & video generation — Vertex AI's Google media models are now routable through dynamic discovery: speech synthesis, audio transcription, music generation, and video generation. (#3929 — thanks @artickc)
• feat(glm): add GLM-5.2 with effort-tier routing (high/max) — GLM-5.2 is registered with high/max effort-tier routing. (#3885 — thanks @dhaern)
• feat(combo): add a sticky round-robin target limit — round-robin combos can cap how many targets stay "sticky" within a session (stickyRoundRobinLimit), balancing stickiness against spread. (#3846 — thanks @adivekar-utexas)
• feat(openrouter): connection presets — OpenRouter connections support reusable presets (provider routing / sort / quantization preferences), selectable when adding a connection. (#3878 — thanks @rdself)

🐛 Fixed

• fix(compression/memory): stop memory + compression from poisoning the upstream prompt cache — with compression and/or memory enabled, requests to caching providers (Anthropic-family) missed the prompt cache on every turn, multiplying cost. Two root causes: (1) memory injection prepended the retrieved memories — which vary per user query — at index 0 of the message array, shifting the entire cacheable prefix every turn; memory is now inserted just before the last user message when the request carries cache_control breakpoints, keeping the cacheable prefix (system prompt + prior turns) byte-stable. (2) the cache-aware skipSystemPrompt flag computed by getCacheAwareStrategy() was dropped by selectCompressionStrategy() (which can only return a mode), so the system prompt could still be compressed under caching; a new resolveCacheAwareConfig() now forces preserveSystemPrompt on for caching requests. (#3936, closes #3890 — thanks @xenstar / @diegosouzapw)
• fix(providers): register BytePlus ModelArk so its API key can be added — adding a BytePlus (ark-…) key reported "invalid". byteplus was present in the provider catalog (APIKEY_PROVIDERS) but never registered in the routing registry, so key validation fell through to { unsupported: true } → HTTP 400 → the UI rendered every key as invalid (and the provider was unusable for inference). Added a registry entry modeled on the existing Volcengine Ark provider: OpenAI-compatible format, base https://ark.ap-southeast.bytepluses.com/api/v3 (region ap-southeast-1), Authorization: Bearer auth, seeded with the catalog's advertised models (Seed 2.0, Kimi K2 Thinking, GLM 4.7, GPT-OSS-120B). (#3935, closes #3877 — thanks @nikohd12 / @diegosouzapw)
• fix(providers): Nous Research key validation no longer fails on a stale probe model — adding a valid Nous Research API key reported "invalid" even though the same key worked via the portal's copy-shell curl. The validation probe sent model: "nousresearch/hermes-4-70b", which Nous does not serve, so the API returned 400 and the validator (which only treated 200/429 as success) reported the key invalid. The probe now uses the real Hermes-4-70B slug, and any non-auth 4xx (400/404/422) is treated as a valid key (the request shape was wrong, not the credentials) — mirroring the longcat/nvidia validators so a future model rename can't re-break key validation. (#3934, closes #3881 — thanks @FerLuisxd / @diegosouzapw)
• fix(stream): persist mid-stream upstream failures — when an upstream stream fails partway through, the partial response and incremental usage are now finalized and persisted instead of lost; extracts a shared streamFailureFinalization path and merges incremental Claude usage (follow-up to #3879). (#3937 — thanks @rdself)
• fix(perplexity-web): update the request payload to schema v2.18 (HTTP 400) — Perplexity web requests started returning HTTP 400; the request payload was updated to Perplexity's v2.18 schema. (#3938 — thanks @artickc)
• fix(stream): keep the in-flight request payload in sync — the pending-by-id request record is now updated in place (Object.assign) so the in-flight payload stays consistent with what was dispatched (coexists with #3937). (#3940 — thanks @rdself)
• fix: stabilize reasoning streams and request logs — reasoning-token streaming and the request-log capture path were stabilized to avoid dropped/duplicated reasoning frames and inconsistent log entries. (#3879 — thanks @rdself)
• fix(opencode-plugin): include nested combo-refs in the LCD context window — the OpenCode plugin now follows nested combo references when computing the least-common-denominator context window, so a combo nested inside another no longer reports an inflated window. (#3910 — thanks @herjarsa)
• fix(models): correct the failed-model auto-hide defaults — the defaults governing when a failed model is auto-hidden were corrected, and auto-hide is now opt-in so models are no longer dropped unexpectedly. (#3930 — thanks @rdself)
• fix(openrouter): show the preset field when editing a connection — the connection-preset field appeared only when creating a connection, not when editing one; it now appears in both (follow-up to #3878). (#3921 — thanks @rdself)
• fix(sse): announce the assistant role on the first delta (Responses→Chat) — the first SSE delta of a Responses-API→Chat-Completions stream now carries role: "assistant", which strict OpenAI-compatible clients expect before content deltas. (#3911 — thanks @diego-anselmo)
• fix(vertex): add the generative-language scope so SA-JSON model discovery works — Vertex service-account (SA-JSON) model discovery failed without the generative-language OAuth scope; the scope is now requested. (#3922 — thanks @artickc)
• fix(proxy): direct-connection fallback for control-plane ops when a pinned proxy is unreachable — control-plane operations (validation, discovery) now fall back to a direct connection when a connection's pinned proxy is unreachable, instead of failing outright. (#3906 — thanks @zhiru)
• fix(providers): prevent zombie-socket hangs for zai/glm and tighten the default keepAlive — zai/glm could hang on dead keep-alive sockets; the default keepAlive was tightened to evict zombie sockets. (#3907 — thanks @insoln)
• fix(setup): remove the stale CJS bundle check from setup-open-code — the OpenCode setup helper no longer checks for a CJS bundle that the now ESM-only plugin no longer ships. (#3908 — thanks @herjarsa)
• fix(opencode-plugin): drop the CJS bundle to fix the OpenCode plugin loader — the plugin is now ESM-only, fixing the OpenCode loader which failed on the dual CJS/ESM build. (#3883 — thanks @herjarsa)
• fix(mcp): fall back to node:sqlite when the better-sqlite3 binding is missing — the MCP server now falls back to Node's built-in node:sqlite when the native better-sqlite3 binding is unavailable, instead of crashing. (#3887 — thanks @megamen32)
• fix(models): correct the generate-models alias lookup — alias resolution during model generation was corrected so aliased model ids resolve to their canonical entry. (#3870 — thanks @YunyunZhai)
• fix(combo): guard the candidate pool against an empty array — combo candidate-pool selection no longer throws when the pool resolves to an empty array. (#3871 — thanks @YunyunZhai)

🔒 Security & Hardening

• fix(security): bump form-data + vite (2 HIGH), harden workflow template-injection & allowlist guarded workflow_run — two HIGH Dependabot advisories (form-data, vite) were upgraded; GitHub Actions workflows were hardened against ${{ }} template-injection (untrusted values now passed via env:); and the guarded workflow_run trigger was allowlisted. (#3949 — thanks @diegosouzapw)

🧹 Internal / Quality / Docs

• fix(ci): grant contents: write to the npm publish job for SBOM attach — the v3.8.25 TokenPermissions hardening set the npm-publish publish job to contents: read, but its "Attach SBOM to GitHub Release" step (gh release upload) needs contents: write and failed with HTTP 403 on the v3.8.25 release (npm / GitHub Packages / opencode-plugin / Docker / Electron all published fine; only the SBOM attach broke — the v3.8.25 SBOM was attached manually). (#3874 — thanks @diegosouzapw)
• ci(quality): make zizmor / gitleaks / osv scanners functional + freeze advisory baselines — the supply-chain scanners are now actually executed (correct install + invocation) with frozen advisory baselines so new findings surface as diffs. (#3947 — thanks @diegosouzapw)
• ci(quality): fix scanner install + size-limit preset, promote codeqlAlerts to blocking — corrected the scanner install and the size-limit preset, and promoted the codeqlAlerts ratchet from advisory to blocking. (#3945 — thanks @diegosouzapw)
• ci(quality): wire Stryker mutation testing as an advisory nightly — Stryker mutation testing runs nightly (advisory) — Quality Gates Fase 7 · Task 11. (#3898 — thanks @diegosouzapw)
• ci(quality): freeze per-module coverage floors + wire require-tighten (advisory) — per-module coverage floors are frozen with an advisory "require-tighten" check that flags modules drifting below their floor. (#3901 — thanks @diegosouzapw)
• ci(quality): enforce the stale-allowlist check on check-known-symbols — stale allowlist entries (suppressing a symbol that no longer exists) now fail the gate — Fase 6A.3 follow-up. (#3899 — thanks @diegosouzapw)
• test(ci): de-flake pipeline-payloads via per-test re-seed + honest reset — the pipeline-payloads suite now re-seeds per test and performs an honest cache reset, eliminating a cross-test ordering flake. (#3893 — thanks @diegosouzapw)
• fix(ci): drop the secrets-in-job-if from nightly-llm-security — referencing secrets in a job-level if caused a startup_failure on push; the gating was moved so the workflow starts cleanly. (#3892 — thanks @diegosouzapw)
• test: reconcile the runtime-timeouts keepAlive baseline to 4000 after the #3907 source revert — the keepAlive assertion was realigned to the source value (4000) after #3907's source-side revert. (#3933 — thanks @diegosouzapw)
• chore(repo): nest quality-gate state under config/quality, declutter the repo root — baselines / allowlists / metrics moved under config/quality/, trimming the tracked root file count. (#3896 — thanks @diegosouzapw)
• docs: refresh the provider count to 226 + regenerate PROVIDER_REFERENCE.md — the README advertised a stale 177 providers; the canonical generator (scripts/docs/gen-provider-reference.ts) now reports 226 unique provider IDs, so the README badges/anchors and the generated provider reference were brought in sync. Also adds a documentation audit/sync report. (thanks @diegosouzapw)
• docs: sync all documentation to v3.8.24 + count-guard & wiki/prose CI — a full documentation sync with a strict provider/locale count-guard plus Vale / markdownlint prose CI. (#3804 — thanks @diegosouzapw)
• docs: regenerate stale counts to canonical values — 226 providers / 87 MCP tools / 15 strategies / 42 locales. (#3904 — thanks @diegosouzapw)
• docs(quality): correct the stale gate count + add an opt-in agent-lsp scaffold — (#3902 — thanks @diegosouzapw)
• docs(mcp): correct the MCP tool-inventory diagram source + text to 87 tools — (#3909 — thanks @diegosouzapw)
• docs: update the compression section to the 9-engine multi-layer stack — (#3894 — thanks @diegosouzapw)
• ci(docs): automate GitHub wiki sync (add missing pages + cover counts) — (#3900 — thanks @diegosouzapw)
• docs: require a dedicated git worktree + branch per development task (Hard Rule #19) — codifies the worktree-isolation rule after the shared-checkout incidents. (#3939 — thanks @diegosouzapw)
• fix(docs): add MDX frontmatter to DOCUMENTATION_AUDIT_REPORT so the fumadocs build passes — the audit report lacked the title: frontmatter MDX pages require. (thanks @diegosouzapw)

---

[3.8.25] — 2026-06-14

✨ New Features

• feat(compression): pluggable compression engines + async pipeline + Compression Studios — a new prompt-compression subsystem with selectable engines (Lite / Aggressive / Ultra), an asynchronous compression pipeline wired into the chat core, and "Compression Studios" tooling for inspecting and tuning compression. (#3848)
• feat(compression-ui): unified compression configuration UI — a Compression Hub with per-engine pages (Lite / Aggressive / Ultra), a combos editor, a dedicated sidebar entry, and live-WS default-on. (#3860)
• feat(security): prompt-injection guard across every LLM route + red-team suite — the prompt-injection guard now runs on all LLM routes (chat, responses, embeddings, images, audio, rerank, search, moderations, videos, music) with a shared input sanitizer and a promptfoo-based red-team suite (Quality Gates Fase 8 · Bloco D). (#3857)
• feat(kiro): live per-account model discovery — Kiro now discovers each account/tier's entitled models via CodeWhisperer ListAvailableModels (region-matched, with a static-catalog fallback). (#3836 — thanks @artickc)
• feat(gemini/vertex): surface Veo video models in dynamic discovery — Veo video models (predictLongRunning) now appear in Gemini/Vertex dynamic model discovery. (#3839 — thanks @artickc)
• feat(mimocode): per-account proxy for multi-account round-robin — each mimocode account can route through its own proxy (resolved per account by fingerprint via runWithProxyContext), with a "Distribute proxies" UI helper. (#3837 — thanks @pizzav-xyz)
• feat(intelligence): expose Arena ELO sync as a feature flag — the LM Arena ELO leaderboard sync is now toggleable (ARENA_ELO_SYNC_ENABLED, DB-override + env fallback). (#3821 — thanks @rdself)

🐛 Fixed

• test(oauth): prove refresh_token preservation for the real gemini-cli / antigravity dispatch — the #3679/#3766 regression test used a synthetic provider that routes through the generic tokenUrl path, so the fix was never proven for the actual Google-family providers, which dispatch through refreshGoogleToken() against the hardcoded OAUTH_ENDPOINTS.google.token. Added a test that drives checkConnection through the real gemini-cli/antigravity path (redirecting the Google token endpoint to a local server returning invalid_grant) and asserts the refresh_token is preserved (not nulled) — confirming these connections are not spuriously destroyed on a failed refresh. (#3850 — thanks @3xa228148)
• fix(oauth): clear setup message for GitLab Duo instead of "Internal server error" — adding a GitLab Duo connection without a registered OAuth client returned an opaque Internal server error at the Add Connection step. buildAuthUrl threw when GITLAB_DUO_OAUTH_CLIENT_ID was missing, and the route swallowed it into a generic 500. It now returns null (mirroring the Qoder provider) and the authorize route surfaces an actionable message: register an OAuth app at https://gitlab.com/-/profile/applications with redirect URI http://localhost:20128/callback and scopes ai_features read_user, then set GITLAB_DUO_OAUTH_CLIENT_ID. (#3861 — thanks @sidinsearch)
• fix(db): persist the "Keep latest backups" retention setting — changing the backup-retention count in Settings → Database backup retention had no effect: it always snapped back to 20 on refresh (and editing .env post-start was ignored too, since process.env isn't reloaded). getDbBackupMaxFiles() only read the DB_BACKUP_MAX_FILES env var — there was no setter and no persisted value. The value now round-trips through a dedicated key_value store (getDbBackupMaxFiles precedence: env override → persisted UI value → default 20), and the "Clean old backups" action persists the chosen count. Existing installs keep the historical default of 20 until explicitly changed. (#3834 — thanks @netstratego)
• fix(sse): clamp Gemini thinking budget to the model's real cap (reasoning_effort/effort=high 400) — translating OpenAI reasoning_effort=high (and Claude-Code output_config.effort=high) to a Gemini target sent a hardcoded thinkingBudget: 32768, which exceeds Flash-tier Gemini's real max of 24576 → upstream HTTP 400 (the thinkingLevel=high path already used 24576 and worked on the same model). gemini-2.5-flash now declares its real thinkingBudgetCap (24576) so the existing capThinkingBudget() chokepoint actually clamps, and the Claude→Gemini output_config.effort path — which previously sent the raw value with no cap at all — now routes through the same clamp (pro-tier, real cap 32768, is left untouched). (#3842 — thanks @andrea-kingautomation)
• fix(intelligence): run pricing + models.dev sync from the live startup path — like the Arena ELO sync (v3.8.24), the external pricing sync (PRICING_SYNC_ENABLED) and the models.dev capability sync (Settings → AI toggle) were only initialized from server-init.ts, which the Next standalone runtime never executes — and models.dev had no caller at all. Their toggles were inert in production. Both are now initialized from instrumentation-node.ts (self-gated, opt-in preserved, non-blocking, never fatal). (thanks @diegosouzapw)
• test(proxy): guard the per-connection 'direct' bypass over a global proxy + clearer label — the per-connection "Proxy Off" toggle (proxyEnabled: false) already overrides a configured global proxy (resolveProxyForConnection short-circuits to level: "direct" before the global step). Added an explicit regression test proving the bypass beats a global assignment (and round-trips on re-enable), and relabeled the UI to "Direct (bypass proxy)" so operators recognize it. Closes the verification gap in #2996. (thanks @diegosouzapw)
• feat(connections): per-connection "disable cooldown" opt-out — a connection can now opt out of the transient cooldown (providerSpecificData.disableCooling, with a toggle in the Edit Connection modal). When set, a recoverable failure still records the error/backoff but does not take the connection out of rotation, so it stays eligible for selection — useful for a primary key you never want parked on a blip. Terminal states (banned / expired / credits_exhausted) still apply. (#2997 — thanks @diegosouzapw)
• fix(combo): restore sessionless combo stickiness + reasoning-aware readiness (504 / TPS regression after v3.8.14) — #3399 (v3.8.16) replaced the <omniModel>-tag combo pinning with a server-side context-cache pin gated on a client sessionId. Clients that send no session id (most OpenAI-compatible tools) lost combo stickiness, so combos re-ran strategy selection every turn → upstream prompt-cache misses → cold high-reasoning starts (~78s) → intermittent [504] Upstream request did not return response headers + TPS collapse (only on combos). The pin now falls back to a stable per-conversation fingerprint (extractSessionAffinityKey(body)) when no session id is present — only when context_cache_protection is on, so #3399's anti-leak behaviour is preserved. Separately, the stream-readiness window now grants the +30s reasoning budget unconditionally for high-reasoning Codex GPT-5.x (small high-reasoning prompts were 504-ing at the 80s base regardless of stickiness). (#3825 — thanks @bypanghu)
• test(combo): cover the skipProviderBreaker consumer gate — the producer was tested but the consumer (whether a failed combo target trips the whole-provider circuit breaker) was not; the breaker decision is now an exported pure predicate (shouldRecordProviderBreakerFailure, behaviour-identical) with direct tests asserting a connection_cooldown 503 does not trip the breaker while a plain 503 does. Closes another deferred test gap from #2743. (thanks @diegosouzapw)
• fix(providers): surface the real Devin error + correct the Windsurf auth instructions — Devin chat returned a generic 502 "Invalid SSE response for non-streaming request" that swallowed the real cause (e.g. "Devin CLI not found"): an error-only SSE chunk (no choices) is now propagated with its sanitized message. The Windsurf "Visit windsurf.com/show-auth-token" instruction (the bare URL shows no token without an IDE-supplied ?state=) now directs users to the Windsurf: Provide Auth Token command-palette flow. (#3324 — thanks @mikmaneggahommie)
• fix(grok-web): clearer 403 message for anti-bot / IP-reputation blocks — a Grok Web subscription validating from a flagged datacenter/VPS IP got a 403 that read like an invalid cookie, sending users to chase a cookie that was actually fine. A non-auth 403 (Cloudflare challenge / anti-bot body) now returns a message stating the cookie is likely OK and the block is IP-reputation-based — retry from a residential IP or configure a proxy (auth-shaped 403s keep the re-paste guidance). (#3474 — thanks @friedtofu1608)
• fix(db): make the mass-pending-migrations safety threshold env-overridable — restoring a backup DB from an older version could trip "Detected N pending migrations … threshold is 50" with no way to override the hardcoded 50. The threshold is now configurable via OMNIROUTE_MAX_PENDING_MIGRATIONS (resolved at startup; 0 disables the check). (#3416 — thanks @samuraiIT)
• test(proxy): cover the Vercel-relay proxyFetch path — net-new tests for buildVercelRelayHeaders and the vercel-type relay short-circuit (x-relay-target/-path/-auth, TCP-skip, missing-auth fail-closed), closing one of the deferred test gaps tracked in #2743. (thanks @diegosouzapw)
• fix(cli): surface omniroute runtime repair in the native-module error messages — after a Node major upgrade, better-sqlite3's prebuilt binary mismatches the ABI and the service can crash-loop; the error only mentioned npm rebuild better-sqlite3 (which fails for global / no-toolchain installs). The startup + SQLite error hints now also point to the existing self-heal command omniroute runtime repair (rebuilds into a user-writable runtime), and a top-level omniroute repair alias was added. (#3476 — thanks @Rahulsharma0810)
• fix(antigravity): per-request Pro-family upstream-id fallback chain (gemini-3.1-pro-high 400) — Antigravity silently renamed the Gemini 3.1 Pro-high upstream id, so gemini-3.1-pro-high started returning HTTP 400 (while -low still worked) and the live id can't be determined statically (competitor proxies disagree). The executor now retries alternative ids on a 400 (gemini-3.1-pro-high → gemini-pro-agent → gemini-3-pro-high, analogous for pro-low), bounded and only on a 400, with zero extra cost on the happy path; the 1:1 tier-passthrough invariant is preserved (the chain is request-time, not a static alias remap). (#3786 — thanks @aliaksandrsen)
• fix(sse): retry once on an early stream close (STREAM_EARLY_EOF) for single-model requests — flaky OpenAI-compatible upstreams (e.g. NVIDIA NIM with minimax-m3 / qwen3.5 / glm-5.1) intermittently send HTTP 200 then close the SSE with zero useful frames, surfacing as a 502 "Stream ended before producing useful content". Only Antigravity got an early-close retry; every other provider returned the 502 immediately on the non-combo single-model path. A bounded one-retry (early-close only — not readiness-timeout — and without marking the account unavailable) now generalizes it. (The separate qwen-web validation SSRF part of the same report was already fixed in v3.8.24, #3767.) (#3758 — thanks @Svatosalav)
• fix(models): preserve eye-hidden models across auto-sync / import — hiding models via the visibility (eye) toggle to keep only a combo's models was undone on every model import or auto-sync, which re-showed all of them. The sync re-import treated "hidden" identically to "deleted" and dropped both; a distinct isDeleted marker now separates the trash/delete path (still dropped on re-import, #3199) from the eye toggle (preserved as listed-but-hidden), and eye-hidden models are no longer re-aliased into the routable catalog on sync. (#3782 — thanks @xenstar)
• fix(providers): correct the lmarena cookie hint (session → arena-auth-prod-v1) — the lmarena credential hint asked for a cookie named session, but lmarena.ai's real auth cookie is arena-auth-prod-v1, so users who pasted only session=… hit validation failures. The credential name, placeholder and storage keys now use the correct name (the legacy session key is retained for back-compat with already-saved credentials). (#3810 — thanks @xspylol)
• fix(reasoning): normalize OpenAI-compatible max effort to xhigh by default — OpenAI-compatible providers do not accept literal max, but some upstreams (for example DeepSeek through OpenRouter) support xhigh; max now maps to xhigh unless the target model explicitly opts out of xhigh, with Claude alias variants still honoring the canonical Claude opt-out list. (#3826 — thanks @rdself)
• fix(combo): return the replay response on the round-robin streaming path — a round-robin combo with a streaming target returned a body already locked by the readiness peek, surfacing as a 500 "ReadableStream is locked"; the round-robin path now returns the replay clone like the priority path does. (#3811 — thanks @0xtbug)
• fix(claude): strip the reasoning-effort suffix from Claude model ids — Claude ids carrying an effort suffix (…-low … …-max) 404'd upstream and tripped the circuit breaker into a misleading "rate-limited" state; the suffix is now stripped before dispatch. (#3807 — thanks @zhiru)
• fix(sse): flush routed SSE chunks promptly (ping/zombie readiness filter) — combo stream-readiness now filters ping/zombie frames so routed SSE chunks stream out without waiting on the readiness window. (#3759 — thanks @rdself)
• fix(models): don't auto-hide transient (rate-limited / timeout) failures on Test All — a parallel Test All across many models could rate-limit an account and auto-hide every model that 429'd / timed out (dropping them from /v1/models); transient failures now surface an error state but stay visible. (#3849 — thanks @lukmanc405)
• fix(quota): surface OpenCode Go's missing quota-API as a latched diagnostic — OpenCode Go keys whose quota endpoints return 404/401 no longer hammer the dead endpoints; the gap is latched with a clear message and an OMNIROUTE_OPENCODE_GO_QUOTA_URL override hint. (#3838 — thanks @adivekar-utexas)
• fix(pricing): add the missing Kiro model pricing rows — Kiro models the registry serves (e.g. claude-sonnet-4.6) had no pricing row and reported $0.00; the rows were added. (#3835 — thanks @artickc)
• fix(ui): render country flags via flagcdn SVGs for Windows compatibility — Windows doesn't render regional-indicator flag emoji; flags now use flagcdn SVGs with an emoji fallback. (#3814 — thanks @rafacpti23)
• fix(ui): expand the request log table with a vertical resize handle — the request log table now shows ~10 rows and can be resized vertically. (#3820 — thanks @rafacpti23)
• fix(i18n): translate the missing embeddedServices keys across 37 locales — the embeddedServices strings showed __MISSING__ in 37 locales; they are now translated. (#3819 — thanks @rafacpti23)

🔒 Security & Hardening

• fix(security): CCR cross-tenant IDOR — per-principal scope store + bounded memory — the compression CCR scope store was shared across principals, allowing cross-tenant reads; it is now scoped per-principal with bounded memory. (#3859)
• feat(supply-chain): build provenance, SBOM, Trivy scan & OpenSSF Scorecard (advisory) — added npm build provenance, a CycloneDX SBOM, Trivy image scanning, and an OpenSSF Scorecard workflow (Quality Gates Fase 8 · Bloco A, advisory). (#3824)

🧹 Internal / Quality / Docs

• Consolidate the email-privacy control into Settings → Appearance — the per-page email-privacy toggles were replaced by a single global switch. (#3822 — thanks @rdself)
• docs(ui): clarify the routing-settings copy (strategy sync + sticky limit) — (#3843 — thanks @adivekar-utexas)
• Quality Gates — Fase 7 & 8 — promoted the dead-code / cognitive-complexity / type-coverage ratchets to blocking, installed advisory CI scanners (gitleaks / osv / actionlint / zizmor), and added property + golden + SSE-correctness tests and a runtime-resilience (chaos / heap-growth / k6 soak) suite. (#3809, #3858, #3808, #3854)
• fix(docs): add MDX frontmatter to SUPPLY_CHAIN.md — the new security doc lacked the title: frontmatter that MDX pages require, which broke the production Build + Docker Hub publish; the frontmatter was added. (#3864)
• chore(deps): bump aquasecurity/trivy-action 0.28.0 → 0.36.0 (#3862)
• chore(quality): reconcile the file-size ratchet baseline for Prettier-inflated v3.8.25 fixes + chat.ts growth — the per-file size baseline was re-frozen to absorb the formatting/line-count growth from this cycle's chat-core and combo fixes (manual edits, never an automatic upward ratchet). (#3823, #3833 — thanks @diegosouzapw)
• test(suite): green the unit suite at release time — align stale tests to this cycle's intended behavior + de-flake two new suites — release-gate housekeeping: updated tests that lagged behind intended behavior changes (OpenCode Go latched quota message #3838, the email-privacy control consolidated into Settings #3822, SOCKS5 default-on proxy-type message, the [id] provider-detail strangler-fig decomposition #3501, Vertex Express-mode keys, Antigravity discovery using a current user-callable model id) and the same-provider 503 fall-through resilience test; de-flaked the compression benchmark reproducibility test (sequential passes) and the ServiceSupervisor crash test (poll instead of fixed sleep). No production code changed. Also documented OMNIROUTE_MAX_PENDING_MIGRATIONS (#3416) in .env.example + ENVIRONMENT.md. (thanks @diegosouzapw)

---

[3.8.24] — TBD

See English CHANGELOG for v3.8.24 details.

---

[3.8.22] — TBD

See English CHANGELOG for v3.8.22 details.

---

[3.8.21] — 2026-06-11

See English CHANGELOG for v3.8.21 details.

---

[3.8.20] — Unreleased

Development cycle in progress.

---

[3.8.19] — Unreleased

Development cycle in progress.

---

[3.8.18] — Unreleased

Development cycle in progress.

---

[3.8.17] — Unreleased

Development cycle in progress.

---

[3.8.16] — Unreleased

Development cycle in progress.

---

[3.8.15] — Unreleased

Development cycle in progress.

---

[3.8.14] — Unreleased

Development cycle in progress.

---

[3.8.13] — Unreleased

Development cycle in progress.

---

[3.8.12] — Unreleased

Development cycle in progress.

---

[3.8.11] — Unreleased

Development cycle in progress.

---

[3.8.10] — Unreleased

---

[3.8.9] — Unreleased

---

[3.8.8] — 2026-06-01

Added

• Plugins framework (src/lib/plugins/, /api/plugins/*, /dashboard/plugins) — hooks + registry unification, plugin SDK (definePlugin), worker-thread sandbox, per-plugin hook rate limiting, SHA-256 integrity verification, semver-gated upgrade, and execution analytics. Plugin routes are loopback-only (isLocalOnlyPath) and child_process exec is opt-in via OMNIROUTE_PLUGINS_ALLOW_EXEC. (#2913 / #3041 — thanks @oyi77)
• Plugin system: response-hook wiring + startup load + example plugin — wires the plugin onResponse hook into the chat success path, loads active plugins on server startup so they survive restarts (pluginManager.loadAll() in server-init), and ships a welcome-banner example plugin (examples/plugins/) plus a comprehensive plugin test suite. (#3045 — thanks @oyi77)
• API key option: disable non-published models — a per-key flag restricting the key to discovered, public models (combos / auto/* / qtSd/* routing still allowed). (#3017 — thanks @androw)
• SessionPool — modular & provider-agnostic (open-sse/services/sessionPool/) — pooled cookie/session manager with round-robin fingerprint rotation (distinct fingerprint per pooled session), per-session cooldown/backoff, and a provider-agnostic webExecutorWrapper. Adds pool support for DuckDuckGo Web and LLM7 providers and an MCP poolTools toolset. (#2954 / #2978 — thanks @oyi77)
• AgentBridge (/dashboard/tools/agent-bridge) — MITM proxy consolidating 9 IDE agents (Antigravity, Kiro, GitHub Copilot, OpenAI Codex, Cursor IDE, Zed Industries, Claude Code, Open Code, Trae stub) with server card, per-agent setup wizard, model mapping table, bypass list, upstream CA cert support, and redirect from legacy /dashboard/system/mitm-proxy. See docs/frameworks/AGENTBRIDGE.md. (#2858 — thanks @diegosouzapw)
• Traffic Inspector (/dashboard/tools/traffic-inspector) — LLM-aware HTTPS debugger with 4 capture modes (AgentBridge hook, Custom Hosts DNS, HTTP_PROXY :8080, System-wide proxy), DevTools split UI, 7 detail tabs (Conversation, Headers, Request, Response, Timing, LLM Details, Stats), resizable panels, session recording (.har/.jsonl export), SSE stream merger, conversation normalizer (multi-provider), system-prompt fingerprint colorization, and annotations. See docs/frameworks/TRAFFIC_INSPECTOR.md.
• MITM handler base + 9 agent handlers (src/mitm/handlers/) — MitmHandlerBase abstract class with hookBufferStart/hookBufferUpdate for Traffic Inspector integration; concrete handlers for all 9 agents.
• MITM targets registry (src/mitm/targets/) — declarative MitmTarget shape per agent; emits DATA_DIR/mitm/targets.json for dynamic server.cjs resolution.
• Traffic Inspector core (src/mitm/inspector/) — TrafficBuffer in-memory ring, kindDetector, sseMerger (MIT port from chouzz/llm-interceptor), conversationNormalizer (MIT port), contextKey fingerprinting, httpProxyServer, systemProxyConfig.
• AgentBridge passthrough + bypass (src/mitm/passthrough.ts) — TCP tunnel for non-mapped hosts; bypass list with default sensitive-host patterns + user-defined patterns.
• Upstream CA cert (src/mitm/upstreamTrust.ts) — AGENTBRIDGE_UPSTREAM_CA_CERT for corporate TLS environments.
• Secret masking (src/mitm/maskSecrets.ts) — sk-/Bearer/generic token masking before any log or Traffic Inspector broadcast.
• DB migrations 073–075 — agent_bridge_state, agent_bridge_mappings, agent_bridge_bypass, inspector_custom_hosts, inspector_sessions, inspector_session_requests.
• ~28 API routes under /api/tools/agent-bridge/ (12 routes) and /api/tools/traffic-inspector/ (16+ routes). All LOCAL_ONLY + SPAWN_CAPABLE.
• i18n PT-BR + EN for all new keys in agentBridge.* and trafficInspector.* namespaces; all other locales fall back to EN automatically.
• E2E smoke tests — tests/e2e/agent-bridge.spec.ts, tests/e2e/traffic-inspector.spec.ts, tests/e2e/agent-bridge-traffic-cross.spec.ts (skip-gated on CI by RUN_AGENT_BRIDGE_E2E / RUN_TRAFFIC_INSPECTOR_E2E / RUN_CROSS_E2E).
• Documentation — docs/frameworks/AGENTBRIDGE.md and docs/frameworks/TRAFFIC_INSPECTOR.md; docs/architecture/REPOSITORY_MAP.md updated; docs/reference/openapi.yaml updated with ~28 new routes and 20+ new schemas.
• i18n: translate Ukrainian (uk-UA) menu and UI strings, plus complete uk-UA UI coverage (#2981 / #2988 — thanks @Lion-killer)
• providers: add SiliconFlow endpoint selector (#2975 — thanks @xz-dev)
• oauth: add Trae SOLO provider (work/code modes) (#2964 — thanks @S0yora)
• providers: add Qwen Web (chat.qwen.ai) web-cookie provider (#2947 — thanks @oyi77)
• Quota Share Engine — multi-provider quota pools — Monitoring/Costs reorg plus a Quota Share Engine: group selector, grouped pool cards, exclusive-quota API keys (allowedQuotas), quotaShared-* routing models via combos, a 3-step pool wizard (legacy Plans page retired), endpoint + key preview, and full pool editing. Adds quota-pool DB migrations. (#2859 / #3022 / #3032 — thanks @diegosouzapw)
• Dashboard page redesigns (Nav Restructure) — agent-skills + omni-skills with a dynamic 42-skill catalog and MCP/A2A discovery (#2827); CLI Code's + CLI Agents + ACP Agents pages (#2839); translator friendly redesign, 5 tabs → 2 (#2847); functional /batch + /batch/files redesign (#2849); Playground Studio + Search Tools Studio (#2869); memory engine redesign — sqlite-vec + hybrid RRF + Studio UI (#2873). (thanks @diegosouzapw)
• notion: add Notion as an MCP context source — 6 tools (notion_search, notion_list_databases, notion_get_database, notion_query_database, notion_read, notion_append_blocks) scoped under read:notion / write:notion, with dashboard "Context Sources" tab, settings API, and token persistence in key_value table (#2959)

Changed

• Sidebar Tools group: added agent-bridge and traffic-inspector items after cloud-agents.
• /api/tools/agent-bridge/ and /api/tools/traffic-inspector/ added to LOCAL_ONLY_API_PREFIXES and SPAWN_CAPABLE_PREFIXES in src/server/authz/routeGuard.ts.
• .env.example: documented 9 new env vars (AGENTBRIDGE_UPSTREAM_CA_CERT, INSPECTOR_BUFFER_SIZE, INSPECTOR_HTTP_PROXY_PORT, INSPECTOR_HTTP_PROXY_AUTOSTART, INSPECTOR_TLS_INTERCEPT, INSPECTOR_SYSTEM_PROXY_GUARD_MINUTES, INSPECTOR_MAX_BODY_KB, INSPECTOR_MASK_SECRETS, INSPECTOR_LLM_HOSTS_EXTRA, INSPECTOR_INTERNAL_INGEST_TOKEN).

Fixed

• codex/providers: POST /api/providers/[id]/refresh (the manual/auto "refresh token" endpoint) no longer rotates rotating-refresh providers (Codex/OpenAI share one Auth0 client_id). This was the last unguarded proactive-refresh entry point: when the dashboard auto-refreshed every expiring connection on a page load (or an old cached frontend bulk-called it), each Codex account's single-use refresh_token was rotated, and Auth0 revoked the whole token family (openai/codex#9648) — every account but the last died with [403] <!DOCTYPE. The endpoint now skips proactive rotation for rotating providers and defers to the reactive, serialized 401 path (same guard as refreshAndUpdateCredentials and the connection-test route).
• codex/quota: opening the Quota / Providers dashboard no longer disconnects Codex multi-account setups. The quota-sync path (refreshAndUpdateCredentials) proactively refreshed every connection — for rotating-refresh providers (Codex/OpenAI share one Auth0 client_id) it refreshed siblings concurrently, so Auth0 revoked the whole token family (openai/codex#9648) and every account but the last died with [403] <!DOCTYPE html>. The quota path now skips proactive refresh for rotating providers (rotationGroupFor) and reuses the current accesstoken, deferring genuine expiry to the reactive, serialized 401 path. Defense in depth: serializeRefresh now leaves a settle gap between two _queued sibling refreshes (default 2000 ms, tunable via CODEX_REFRESH_SPACING_MS, "0" to opt out) while releasing a lone refresh immediately, so the reactive path adds no latency.
• payload-rules: saved payload rules now survive a server restart. When no in-memory override is set (fresh process before the boot hook ran, or a separate module instance in the standalone build), getPayloadRulesConfig now reads the DB-persisted rules (the source of truth) before the file config, instead of silently returning the empty file default. (#2986)
• models/custom: custom models can now carry a per-model targetFormat override (e.g. an opencode-go custom model that must use the Anthropic Messages shape). Previously custom models always routed as OpenAI-compatible because targetFormat was neither persisted nor consulted at routing time. Threaded through addCustomModel/replaceCustomModels/updateCustomModel, the API schema/route, getModelInfo, and chatCore's targetFormat resolution. (#2905)
• providers/pollinations: route to gen.pollinations.ai/v1 instead of the retired text.pollinations.ai host, which now returns 404 "legacy API" for all models. The gen gateway is the current OpenAI-compatible endpoint. (#2987)
• executors/codex: drop the CLI-injected image_generation hosted tool for free-plan Codex accounts (workspacePlanType === "free"), which can't run it server-side and would otherwise get an upstream 400. Paid plans keep it. (mirrors CLIProxyAPI's free-plan guard; spun off from the #2980 analysis)
• dashboard: custom providers (openai-compatible-* / anthropic-compatible-*) now show their user-given node name instead of the raw UUID id across the active-requests panel, proxy logger, and home-page provider topology. The display-label resolver was extracted into a shared util reused by all surfaces (previously only the request-log viewer resolved it). (#2968)
• docker: the standalone launcher (Docker CMD) now honors OMNIROUTE_MEMORY_MB (default 512, clamped [64, 16384]) and overrides the image NODE_OPTIONS fallback, fixing random OOM crashes under load / with large SQLite DBs. Previously only omniroute serve honored the knob. (#2939)
• docker: add a web compose profile (omniroute-web, target runner-web, image omniroute:web) so web-cookie providers (gemini-web, claude-web, claude-turnstile) work out of the box — the default base image ships without Chromium/Playwright, which made those providers fail with "Executable doesn't exist at .../ms-playwright/chromium...". (#2832)
• routing/codex: fix two gpt-5.5 Codex defects (#2877). (A) For a Codex-only account, a bare gpt-5.5 Responses request was rerouted to codex with the model hardcoded to gpt-5.5-medium (chatHelpers.ts); the executor read that -medium suffix as an explicit modelEffort that (per #2331) overrode a client reasoning.effort=xhigh, silently demoting it — now it keeps the bare gpt-5.5 id so the client effort wins. (B) gpt-5.5-xhigh/-high/-low misrouted to openai (→ "No credentials" for codex-only users); the suffixed variants are now in CODEX_PREFERRED_UNPREFIXED_MODELS so they infer codex.
• sse/chatCore: remove a duplicate const settings declaration in handleChatCore (introduced alongside the per-key stream-default-mode feature). The same-scope redeclaration made esbuild/tsx fail with "The symbol 'settings' has already been declared", which turned every unit test that imports chatCore red and broke the production build. The earlier consolidated settings const is now reused.
• db/migrations: resolve a 077 migration version collision (077_api_key_stream_default_mode.sql vs 077_quota_pools.sql) that made getMigrationFiles() throw and blocked getDbInstance() at startup (app would not boot; every DB-touching test was red). Renumbered the dependency-free, idempotent quota_pools migration to 085, kept the non-idempotent api_key_stream_default_mode ALTER at 077, added a retroactive isSchemaAlreadyApplied guard (case 085), and a regression test enforcing unique migration prefixes.
• routing/reasoning-replay: OpenCode big-pickle (provider opencode/oc and opencode-zen) now declares the interleaved reasoning_content contract via a new RegistryModel.interleavedField field, so follow-up/tool-use turns replay reasoning_content. Previously big-pickle matched no replay pattern and failed with [400] The reasoning_content in the thinking mode must be passed back to the API (its DeepSeek-thinking upstream is not detectable from the model id, and requiresReasoningReplay does not consume supportsReasoning). getResolvedModelCapabilities now surfaces the registry interleavedField. (#2900)
• providers/github-copilot: built-in GitHub Copilot Claude Opus and Gemini models (claude-opus-4.7, claude-opus-4-5-20251101, gemini-3.1-pro-preview, gemini-3-flash-preview) no longer carry targetFormat: "openai-responses", so they route through chat/completions (the provider default, like the working claude-opus-4.6) instead of the Responses API, which Copilot does not serve for non-OpenAI models (returned [400]). Native OpenAI gpt-* models keep the Responses API. (#2911)
• translator/responses: Codex Desktop injects an image_generation hosted tool into every Responses API request (even text-only ones), which OmniRoute rejected with [400] image_generation tool type is not supported. It is now treated like tool_search: allowed past the tool-type validator and dropped silently from the tools array before forwarding to Chat Completions. (#2950)
• combo/builder: no-auth OpenCode Free combo entries now use the oc/ routing alias instead of the opencode/ prefix. parseModel("opencode/<model>") resolves to the opencode-zen api-key tier (via a manual ALIAS_TO_PROVIDER_ID override), so combos built with the bare provider id misrouted away from the no-auth opencode provider; oc/<model> resolves correctly. (#2901)
• resilience/providers: a route-restriction 403 (e.g. Fireworks Fire Pass fpk_* keys returning "…not authorized for this route." on /models, while chat still works) no longer marks the connection unavailable. Provider validation falls through to the chat probe for such 403s instead of returning "Invalid API key", and checkFallbackError short-circuits them to no cooldown. Genuine auth failures (401 / generic 403) still fail fast. (#2929)
• auth/opencode-zen: the OpenCode Zen free model now works in the Playground and combos without an API key. opencode-zen serves the public, signup-free endpoint (https://opencode.ai/zen/v1); when no api-key connection is configured, credential resolution now falls back to anonymous (no-auth) access instead of failing with "No credentials for provider: opencode-zen". A configured, active key is still used when present. (#2962)
• translator/responses: fixed an upstream [400] Messages with role 'tool' must be a response to a preceding message with 'tool_calls' when a Codex client sent a function_call with an empty/missing call_id. The orphaned function_call_output previously slipped past the orphan filter. Now empty-call_id function calls are skipped (no dangling assistant tool_call) and any tool result without a matching tool_call id is dropped. (#2893)
• deps: remove the proxifly npm dependency (#3000 — thanks @terence71-glitch)
• proxy: use connection proxy for OAuth refresh (#3012 — thanks @terence71-glitch)
• usage: export pure helper functions for unit testing (#3015 — thanks @oyi77)
• docs/docker: align memory default docs to 1024MB (#3006 — thanks @terence71-glitch)
• providers: fix DuckDuckGo missing API key & update OpenCode free model list (#3008 — thanks @NekoMonci12)
• claude: bump Claude Code identity to 2.1.158 and sync beta flags (#3010 — thanks @Tentoxa)
• test: increase DB and usage utils coverage to >60% (#3018 — thanks @oyi77)
• oom: resolve memory leak in Bottleneck limiter caches and provider registry (#2965 — thanks @soyelmismo)
• proxy: show registry provider proxies in dashboard after Custom proxy flow moved them into the proxy registry (#2963 — thanks @terence71-glitch)
• routing: add agy to executor map so it uses AntigravityExecutor (#2957 — thanks @ReqX)
• skills: avoid Claude assistant tool_result blocks (#2956 — thanks @terence71-glitch)
• perf: CPU leak from Bottleneck limiter accumulation + per-request optimizations (#2951 — thanks @soyelmismo)
• combo: combo credential resolution ignores target.providerId — prefer combo target's providerId over model-inferred provider (#2946 — thanks @oyi77)
• dashboard: v3.8.8 screen fixes — agent-bridge SSR + audit/logs/memory/playground (#2944)
• claude: sanitize tool schemas + cloak third-party tool names on native Claude OAuth (#2943 — thanks @NomenAK)
• auth: prevent Codex multi-account refresh_token family revocation (#2941)
• combo: fix combo vision passthrough and Codex tool history repair (#2940 — thanks @charithharshana)
• claude: map WebSearch to Responses web_search (#2938 — thanks @makcimbx)
• claude: strip empty Read pages tool input (#2937 — thanks @makcimbx)
• dashboard: improve self-service provider quota visibility (#2931 — thanks @guanbear)
• antigravity: avoid visible signatureless tool history (#2927 — thanks @dhaern)
• sse/web-search: bypass the web-search fallback on a Claude → Claude passthrough so native Claude requests aren't rewritten (#2960 — thanks @terence71-glitch)
• oom: prevent per-request memory accumulation (~256MB heap growth) (#2973 — thanks @soyelmismo)
• perf/proxy: parallelize provider proxy overlay lookups (#2984 — thanks @terence71-glitch)
• privacy/PII: resolve the PII feature flag correctly and fix PII response sanitization in streaming SSE requests (#3021 — thanks @dangeReis)
• electron: improve macOS window chrome (#3029 — thanks @bobbyunknown)
• i18n: fix missing API key scope translations (#3031 — thanks @guanbear)
• stream/responses: drop a leaked chat bootstrap chunk for Responses-API clients (#3035 — thanks @CitrusIce)
• docker: warn-only on the /app/data permission check instead of exit 1, so a non-writable bind mount no longer kills the container at boot (#3036 — thanks @wussh)
• mcp: resolve streamable-HTTP transport readiness reporting an offline status (#3037 — thanks @Chewji9875)
• dashboard: use a lightweight ping endpoint for the MaintenanceBanner (fixes #3040) (#3043 — thanks @herjarsa)
• test: resolve pre-existing test failures — env sync, PII, quota, sidebar (#3039 — thanks @oyi77)
• docs/mcp: regenerate the mcp-tools diagram for 43 tools and fix the tool count (#3028 — thanks @diegosouzapw)
• mcp: move enforceScopes guard before MCP_TOOL_MAP lookup, add inline scopes parameter to withScopeEnforcement(), and declare scopes on all 24 dynamic tool definitions (memory, skills, plugins, gamification, compression) to fix scope enforcement for dynamic MCP tool groups (#2958)

---

[3.8.7] — 2026-05-29

✨ New Features

• api (self-service): add GET /api/v1/me/status so a delegated API key can view its own usage (USD used, budget percent, token totals) and optional shared Codex account quota, backed by migration 075_api_key_self_service_usage_scopes (#2908 — thanks @guanbear).
• analytics: roll up usage logs to daily_usage_summary before raw log cleanup, and query a SQL UNION of raw and rolled-up data to prevent analytics history data loss (#2904 — thanks @unitythemaker).
• perf (RAM): reduce server memory footprint by capping 11 in-memory caches, limiting SQLite page cache, lazy-loading provider registries via Proxy, and optimizing Next.js startup database probes (#2903 — thanks @soyelmismo).

🔧 Bug Fixes

• token-accounting: prefer prompt_tokens over compatibility input_tokens for Anthropic Claude streams to avoid double-counting cached tokens (#2904 — thanks @unitythemaker).

• agy: add the Antigravity CLI (agy) as a standalone OAuth provider next to gemini-cli/antigravity. It reuses the antigravity inference backend (identical Google client, daily-cloudcode-pa.googleapis.com) but ships its own model catalog — notably the Claude models the backend exposes (claude-opus-4-6-thinking, claude-sonnet-4-6) — its own account pool, and connection methods: import the agy CLI token file (paste/upload), auto-detect a local CLI login (~/.gemini/antigravity-cli/antigravity-oauth-token), browser OAuth, and bulk/ZIP import. New routes: POST /api/providers/agy-auth/{import,import-bulk,zip-extract,apply-local}.

Breaking Changes

• proxy-logs: GET /api/usage/proxy-logs now returns clientIp instead of publicIp for each log entry. External consumers reading log.publicIp must update to log.clientIp. The underlying SQLite column (public_ip) is unchanged, so callers that query the database directly are unaffected (#2880 — thanks @rdself).

Known Inconsistency

• log-export: GET /api/logs/export?type=proxy-logs returns raw SQLite rows whose IP field is still named public_ip (the historical column name). This differs from the clientIp field exposed by GET /api/usage/proxy-logs. The two endpoints are intentionally inconsistent for now and will be aligned in a future migration (#2880).

✨ New Features

• usage: add per-API-key token limits scoped to model/provider/global with two-tier inline enforcement and in-memory cache accelerator (#2888 — thanks @mugnimaestra).
• providers: audit web cookie providers, fix 4 missing registry entries, and add DuckDuckGo AI Chat provider (#2862 — thanks @oyi77).
• compression: expand pt-BR pack with 34 new rules inspired by the troglodita project (#2818 — thanks @leninejunior).

🔧 Bug Fixes

• oauth: hotfix Windsurf login — drop dead PKCE flow, promote import-token, and resolve SQLite bind type errors (#2884 — thanks @yunaamelia).
• models: prune stale synced available models for inactive connections and dynamically map Antigravity MITM aliases loop-safely (#2886 — thanks @herjarsa).
• antigravity: harden signatureless tool history replay by making text representation inert (#2878 — thanks @dhaern).
• i18n: complete 144 missing Portuguese (pt-BR) locale keys and synchronize them with English (#2870 — thanks @alltomatos).
• opencode-go: add OpenCode Go provider limits quota fetcher to retrieve Z.AI quota windows (#2861 — thanks @RajvardhanPatil07).
• reasoning: gate reasoning trace replay injection on model interleaved capability metadata (#2843 — thanks @nickwizard).
• audio: construct multipart body manually for transcription form-data to prevent dropped boundary headers under Next.js fetch (#2842 — thanks @soyelmismo).
• gemini-cli: prefer real Google Cloud project IDs over default-project during model synchronization (#2841 — thanks @nickwizard).
• mcp: redirect console.log and console.warn startup messages to stderr in stdio MCP mode to prevent JSON-RPC parsing failures (#2840 — thanks @disonjer).
• antigravity: normalize unescaped tool calls and classify resource exhaustion 429 errors as lockout cooldowns (#2828 — thanks @Ardem2025).
• sse: repair RTK engine defaults to resolve consecutive-line deduplication and direct compression calls (#2825 — thanks @leninejunior).
• fix(usage): add opencode-go / opencode / opencode-zen quota fetcher so the provider limits page surfaces $12/5h, $30/wk, $60/mo windows alongside other quota-aware providers (#2852 — thanks @apoapostolov)

---

[3.8.6] — 2026-05-27

🧹 Chores

• gitignore: ignore .claude/settings.local.json so per-user Claude Code permissions never get committed by accident
• release: version bump and metadata sync (package.json, package-lock.json, electron, open-sse, openapi.yaml)

v3.8.6 is a maintenance/scaffolding patch. All feature and bug-fix work from the post-v3.8.5 cycle (44 commits — community PRs #2777, #2782–#2787, #2789, #2790, plus internal hotfixes) was already integrated into v3.8.5 and is documented under that section.

---

[3.8.5] — 2026-05-26

🔒 Security

• authz: redirect /home and /home/:path* to /login when unauthenticated (#2712)

🔧 Bug Fixes

• mcp: break callLogs ↔ compliance ESM cycle that deadlocks the bundled MCP server on Node.js 24 (#2650)
• deepseek: guard PoW solver Web Worker handler under Node strict mode (#2724)
• combos: include no-auth providers in the combo builder picker (#2737)
• translator: allow the web_search server-tool family in the Responses API translator (#2695)
• oauth: register the missing trae provider with import_token flow (#2658)
• model: merge settings-based aliases with the legacy DB alias namespace (#2618, #2208)
• kiro: clipboard fallback for HTTP / non-secure contexts (#2689)
• cli: raise omniroute serve ready timeout to 60s with TCP fallback for Windows cold start (#2460)

---

[Unreleased]

---

[3.8.23] — TBD

---

✨ New Features

🔧 Bug Fixes

---

[3.8.4] — 2026-05-25

Added

• Embedded services (work in progress — 9Router, CLIProxyAPI; see T-15 for full entry).

---

[3.8.3] — 2026-05-24

✨ New Features

• feat(combos): universal context handoff for cross-model conversation continuity — structured XML summary system (<context_handoff>) that preserves conversation continuity and handles state transfer when combo routing switches models. (#2653 — thanks @herjarsa)
• feat(docs): migrate /docs to Fumadocs MDX with nested routes — replaces the custom docs engine with Fumadocs, adding [...slug] catch-all routing, search API at /docs/api/search, source.config.ts content configuration, and meta.json navigation files across 8 doc sections (architecture/, compression/, frameworks/, guides/, ops/, reference/, routing/, security/). Includes 50+ URL redirects for backward compatibility via next.config.mjs. (#2614 — thanks @ovehbe)
• feat(dashboard): add search and filters to /dashboard/api-manager — filter bar with search by name/key, active-only toggle (persisted to localStorage), status filter (active/disabled/banned/expired), type filter (standard/manage/restricted), filter count badges, and empty state with "Clear Filters" button. (#2628 / #2641 — thanks @diegosouzapw)
• feat(dashboard): free-tier grouping with symbolic link in /dashboard/providers — groups and displays all free-tier providers from all categories dynamically using hasFree: true properties without removing them from their native lists. Displays category dot and amber dot with localizable tooltips, dedupes search results by provider ID, and corrects free tier count statistics. (#2632 — thanks @diegosouzapw)
• feat(dashboard): risk notice modal for sensitive providers — show a soft, informative warning modal when connecting to session-based or OAuth providers (like Claude, Cursor, Copilot) for the first time. Adds subscriptionRisk properties to 20 providers, localizable templates, and stores acknowledgment in localStorage. (#2633 / #2638 — thanks @diegosouzapw)
• feat(dashboard): refactor free-tier provider dashboard layout — cleans up visual clutter, reorganizes categories, hides redundant banners, and integrates free-tier categories nicely into the primary provider interface. (#2640 — thanks @diegosouzapw)
• feat(dashboard): mini-playground inline (Phase 4) — integrated interactive mini-playground capabilities to provider details pages, including support for specialized example cards (Embedding, Image, LLM Chat, Music, STT, TTS, Video, Web Fetch, Web Search), unified API key loading hooks, model listing hooks, and curl command builder. (#2648 — thanks @diegosouzapw)
• feat(webfetch): category support with dedicated media providers page and executors for Firecrawl, Jina Reader, and Tavily. (#2645 — thanks @diegosouzapw)
• feat(adapta): integrate Adapta Org (adapta-web) provider with automatic Clerk authentication refresh and custom onboarding tutorial modal. (#2643 — thanks @df4p)
• feat(i18n): complete translations for Simplified Chinese — translates 1220 missing keys bringing UI coverage to 98.8% with 0 placeholders. (#2655 — thanks @L-aros)

🔧 Bug Fixes

• fix(settings): Require Login modal Cancel button text and dismissal — modal now renders localized cancel label via the common namespace and closes correctly without modifying settings when cancelled. (#2649 — thanks @Chewji9875)
• fix(deepseek-web): re-apply SSE parser, prompt format, and error handling fixes — handles all 3 DeepSeek SSE stream formats (initial fragments, APPEND operations, bare string tokens), uses non-greedy regex for markdown image stripping, simplifies prompt to single-turn, checks json.code before token extraction, and uses accessToken fallback for session cache eviction on auth errors. (#2616 — thanks @ovehbe)
• fix(deepseek-web): SSE thinking/search routing and session lifecycle — properly routes thinking vs content fragments based on thinking_enabled flag, handles search results with citation indices, appends search result footnotes, refactors transformSSE() and collectSSEContent() with shared helpers. (#2624 — thanks @ovehbe)
• fix(codex): use allowlist to strip non-Responses-API fields in non-passthrough path — strips residual Chat Completions fields (stream_options, service_tier, store, metadata) from the request body when routing through the non-passthrough (translation) code path, preventing GPT-5.5 from receiving invalid parameters. (#2615 — thanks @diegosouzapw)
• fix(catalog): skip static PROVIDER_MODELS when synced models exist — prevents stale/duplicate model entries in /v1/models for auto-synced providers. (#2625 — thanks @herjarsa)
• fix(qoder): Cosy auth fallback for PAT tokens + vision support for qwen3-vl-plus — when a PAT token gets 401, falls back to Cosy auth against api1.qoder.sh; adds supportsVision: true to qwen3-vl-plus. (#2629 — thanks @herjarsa)
• fix(cli): register tsx loader and add opencode config subcommand — registers tsx/esm at CLI startup so dynamic .ts imports resolve; adds omniroute config opencode convenience alias. (#2631 — thanks @amogus22877769)
• fix(claude): improve Pi and OpenCode compatibility — adds Pi Coding Agent anchors to system transform removal, stores _toolNameMap as non-enumerable, strips context_management when thinking is disabled. (#2621 — thanks @unitythemaker)
• fix(passthrough): restore semantic passthrough system-role-only extraction — reverts full normalizeClaudeUpstreamMessages() to lighter extractSystemRoleMessages() in CC semantic passthrough paths, preventing document/tool chain corruption. (#2620 — thanks @Tentoxa)
• fix(kiro): stabilize conversationId across prompt compression — captures pre-compression body and uses the original first user message as seed for UUID v5, keeping Kiro's AWS conversation context stable. (#2630 — thanks @HALDRO)
• fix(t3-chat-web): close implementation gaps for t3.chat TanStack Start, tracking of stream_options, and retry configurations — parses TSS Turbo Stream Serialization from _serverFn/*, tracks request combo_strategy via database migration 062_usage_history_combo_strategy.sql, and makes batch retry backoffs custom-configurable via environment variables. (#2634 — thanks @oyi77)
• fix(reasoning): extend empty reasoning_content injection to prevent tool call loops in Kimi K2 and replay models — injects empty reasoning_content field to Kimi models during tool-calling sequences to bypass loop issues. (#2639 — thanks @herjarsa)
• fix(cli): Linux autostart via systemd user service on headless VPS — adds auto-generating systemd user service unit for headless setups on Linux, updating tray configs and system variables allowlist (LOGNAME and XDG_CURRENT_DESKTOP). (#2635 — thanks @janeza2)
• fix(combo): preserve <omniModel> tag in SSE stream output for combos when using context_cache_protection to ensure correct context pinning round-trips. (#2646 — thanks @herjarsa)
• fix(rtk): prevent false positives in RTK compression by skipping content-based filter matching for non-shell tool results (e.g. read_file, grep_search). (#2642 — thanks @HALDRO)
• fix(translator): enable Claude extended thinking for Copilot Responses-API requests — handles reasoning budget and translations for Copilot. (#2647 — thanks @ivan-mezentsev)
• fix(tests): remove duplicate assertion in schema coercion & fix(cli): ignore system vars in env check. (thanks @diegosouzapw)

📝 Maintenance

• chore(config): ignore additional agent workflow command files (.agents/commands/). (thanks @diegosouzapw)
• chore(config): ignore memory-bank and Cursor agent rules from tracking. (thanks @ovehbe)

---

[3.8.2] — 2026-05-22

✨ New Features

• feat(@omniroute/opencode-plugin): upstream-provider suffix in model display name — appends provider label to enriched names (e.g. Claude Opus 4.7 · Claude vs Claude Opus 4.7 · Kiro) so the OC TUI model picker can differentiate same-id models routed through different upstream connections. Default-on, opt-out via features.providerTag: false. (#2602 — thanks @mrmm)
• feat(@omniroute/opencode-plugin): provider-tag becomes a prefix + traffic-light compression emoji — provider label now prepends (Claude - Claude Opus 4.7) for better TUI column grouping, with smart abbreviation for long labels (GitHub Models → GHM). Compression pipelines render intensity as emoji (🟢🟡🟠🔴). (#2604 — thanks @mrmm)
• feat(providers): add 7 free-tier providers (Wave 1) — Arcee AI, InclusionAI, Krutrim, Liquid AI, MonsterAPI, Nomic, and Poolside now available as new API-key providers with provider icons, model specs, and full routing support. (#2479 — thanks @oyi77)
• feat(providers): add Astraflow provider support with global + China endpoints — new provider with dual-region base URLs for global and mainland China access. (#2486 — thanks @ucloudnb666)
• feat(providers): add claude-web provider — cookie-based Claude Web chat access without OAuth. (#2476 — thanks @oyi77)
• feat(providers): add 14 free-tier providers (Wave 1b) — 360AI, Baichuan, Baidu, ByteDance/Doubao, IDEO, Kuaishou/Kling, Kunlun/Skywork, SenseTime/SenseNova, Stepfun, Tencent HunYuan, Zhipu GLM, Replicate, RunPod, and Modal with provider icons, model specs, and routing support. (#2488 — thanks @oyi77)
• feat(hermes): add rich multi-role Hermes Agent CLI support — 7 configurable roles (default, delegation, vision, compression, web_extract, skills_hub, approval), per-role model selection with YAML config generation, dashboard card with preview, and home widget integration. (#2526 — thanks @apoapostolov)
• feat(cloud-agents): cloud agents UX overhaul — tabs (tasks/agents/settings), status filters, Material icons, duration formatting, cloud agent credentials and health API endpoints, memory stats endpoint. (#2516 — thanks @oyi77)
• feat(authz): manage-scope API keys may reach /api/mcp/* from non-loopback — Route Guard Tiers system (LOCAL_ONLY / ALWAYS_PROTECTED / MANAGEMENT), narrow carve-out for remote MCP access gated by manage scope; /api/cli-tools/runtime/* stays strict-loopback. Includes dashboard AuthzSection, inventory API, and comprehensive docs. (#2473 — thanks @mrmm)
• feat(home): home page customization for experienced users — pin Provider Quota to home, toggle Quick Start and Provider Topology visibility via Appearance settings. (#2531 — thanks @apoapostolov)
• feat(home): automatic refresh of Provider Quota — configurable interval (60s–600s) with toggle in Appearance settings; auto-refreshes pinned quota on the home page. (#2532 — thanks @apoapostolov)
• feat(@omniroute/opencode-plugin): OmniRoute OpenCode plugin — live models fetched from OmniRoute API, combo-aware model listing, Gemini request sanitization, multi-instance support, auth flow integration, and 10 test files. (#2529 — thanks @mrmm)
• feat(executors): forward OpenCode client headers to upstream providers — OpenCode-specific headers are now forwarded through the executor pipeline for improved compatibility. (#2538 — thanks @kang-heewon)
• feat(fireworks): add new models with modelIdPrefix support — generic registry field that stores short model IDs and prepends the full path prefix before upstream API calls. Adds 6 new Fireworks models, modelsUrl for dynamic sync, and Qwen3 reranker. (#2560 — thanks @HALDRO)
• feat(@omniroute/opencode-plugin): readable + filterable + offline-resilient model picker — usableOnly filter (only show providers with healthy connections), diskCache for offline hydration, Combo: prefix labeling, and compression metadata tags in combo display names. (#2572 — thanks @mrmm)
• feat(smart-pipeline): multi-stage pipeline for auto combo routing — rule-based + intent-classifier + domain-specific stages with configurable pipeline router, accuracy benchmarks, and comprehensive tests. (#2551 — thanks @oyi77)
• feat(ops): skip DB health check on startup via OMNIROUTE_SKIP_DB_HEALTHCHECK=1 — replaces slow integrity_check (7+ min on large WAL) with quick_check, and adds env var to skip entirely. (#2554 — thanks @soyelmismo)
• refactor(dashboard): Provider Quota grouped layout with vertical rail — restructures the page to a 2-column per-provider layout (left rail with icon/name/status, right content with dynamic per-provider columns), new providerColumns.ts / ProviderGroup.tsx / AccountRow.tsx components, env chip-filter row, bulk-refresh per group, and inline expanded panels. (#2528 — thanks @Gi99lin)
• feat(providers): add 26 free-tier providers missing from registry — Novita, Avian, Chutes, Kluster, Targon, Nineteen, Celery, Ditto, Atoma, and more. (#2590 — thanks @oyi77)
• feat(providers): add api-airforce free provider with 55 models. (#2587 — thanks @oyi77)
• feat(dashboard): configurable sidebar — presets, drag-and-drop ordering, smart-grouping, and new Settings → Sidebar page. (#2581 — thanks @Gi99lin)

🔧 Bug Fixes

• fix(validation): stop appending a second /models when the Gemini base URL already ends in /models — Google AI Studio connections using the default base URL were validating against .../v1beta/models/models and failing with 404 for every connection. (#2545)

• fix(cloudflare-ai): flatten OpenAI content-part arrays to plain strings for the Workers AI (cf/) executor — Workers AI's /ai/v1/chat/completions rejects content: [{type:"text",...}] with HTTP 400, so requests with array content now have their text parts joined into a string. (#2539)

• fix(i18n): replace leftover Portuguese strings in the English source with English on the Quota dashboards — the quota-share Beta notice (betaConfigSaved*) and the Provider Quota row's Edit cutoffs / Refresh now fallbacks were showing Portuguese. (#2540)

• fix(proxy): honor the legacy per-provider/global proxy config in resolveProxyForProvider — the Claude OAuth token exchange and token refresh only consulted the new proxy registry, so a proxy configured the legacy way (/api/settings/proxy?level=provider) was ignored and the exchange went out directly from the host, tripping Anthropic's IP rate_limit_error on VPS deployments. It now falls back to the legacy config, mirroring resolveProxyForConnection. (#2456)

• fix(antigravity): auto-discover a missing Cloud Code projectId via loadCodeAssist before failing — a freshly re-added Antigravity account whose stored projectId was empty (OAuth-time discovery returned nothing) now recovers the project on the first request instead of returning 422 Missing Google projectId, mirroring the gemini-cli bootstrap. (#2334, #2541)

• fix(stream): keep the /v1/responses SSE connection warm for strict clients — emit an early keepalive while the upstream produces its first token and lower the heartbeat cadence to 4s, so Codex CLI's reqwest client (≈5s idle-read timeout) no longer drops the stream "before completion" on slow/reasoning models. curl was unaffected because it has no idle timeout. (#2544)

• fix(electron): wait longer for the server on first launch and reload once it responds — long post-upgrade DB migrations could exceed the 30s readiness probe, leaving the desktop app stuck on the "Server starting" screen even though the backend was healthy. The probe now targets the auth-exempt health endpoint with a generous timeout and reloads the window once the server comes up. (#2460)

• fix(cli): mark bin/omniroute.mjs as executable (mode 755) so the globally-installed CLI runs directly without a manual chmod +x. (#2469 — thanks @disonjer)

• fix(settings): restore the Global System Prompt into the in-memory config on server startup and after JSON/SQLite import — it was only loaded by the PUT endpoint, so the toggle/prompt silently reverted to defaults after any restart or import. (#2470 — thanks @disonjer)

• fix(settings): append the Global System Prompt after existing system content instead of prepending it, so provider/agent instructions (Kiro, OpenCode, Hermes, …) injected into the system message no longer override the user's global prompt via recency bias. (#2468 — thanks @disonjer)

• fix(kiro): refresh imported social tokens (authMethod === "imported") via the Kiro social-auth endpoint instead of AWS SSO OIDC — imported tokens carry a registered clientId/clientSecret but a social-issued refresh token the OIDC client cannot refresh, so auto-refresh was failing with "provider returned no new token". (#2467 — thanks @disonjer)

• fix(antigravity): resolve the Cloud Code projectId from providerSpecificData as a fallback (and preserve it across token refresh) so the Gemini /v1beta streaming path stops returning a spurious 422 Missing Google projectId for connections that store the project there. (#2480)

• fix(api): GET /v1beta/models now lists only models whose provider has an active/validated connection, matching the OpenAI-format /v1/models behavior, instead of returning the entire catalog. (#2483)

• fix(cli): persist STORAGE_ENCRYPTION_KEY into DATA_DIR (not only ~/.omniroute) and refuse to auto-generate a fresh key when a storage.sqlite already exists — a new key cannot decrypt previously-encrypted credentials, so silently regenerating it locked users out of their database. The CLI now mirrors the server bootstrapEnv guard. (reported by Daniel Nach; original key persistence by @Chewji9875 — follow-up to #1622)

• fix(gemini): preserve and re-attach the thoughtSignature on Gemini thinking-model tool calls — thread the signature namespace through the FORMATS.GEMINI and FORMATS.GEMINI_CLI request translators so the cached signature (keyed by connection + tool-call id) is found on the follow-up turn. Fixes [400]: Function call is missing a thought_signature in functionCall parts on agentic Gemini tool use. (#2504)

• fix(translator): accept PDFs sent in the Responses-API input_file shape on the Gemini path, and the Gemini-style document shape on the Responses/Codex path — content parts are now normalized across input_file / file / document so a PDF reaches the model regardless of which field name the client used. (#2515)

• fix(stream): count thinking arrays and reasoning_details as useful stream output — a reasoning-only response (e.g. Mistral/StepFun with a low max_tokens) was misclassified as "Stream ended before producing useful content" and turned into a spurious 502; it is now recognized as valid output. (#2520)

• fix(claude): extract system/developer role messages in Claude Code semantic passthrough paths — moves role:"system" / role:"developer" messages from the messages[] array to the top-level system parameter before sending to Anthropic, which rejects them inside messages. Fixes memory injection context being silently dropped. (#2497 — thanks @unitythemaker)

• fix(vision-bridge): auto-route non-standard provider models through OmniRoute self-loop — vision-bridge now detects when a model doesn't natively support vision and automatically re-routes the image through OmniRoute's own endpoint for format translation. (#2487 — thanks @herjarsa)

• fix(mitm): add IPv6 DNS redirect, modular antigravity target, improved logging — MITM DNS handler now correctly redirects IPv6 (AAAA) queries alongside IPv4, adds a dedicated antigravity.ts target module, and enhances DNS/TLS logging for debugging. (#2514 — thanks @herjarsa)

• fix(usage): improve Claude and MiniMax plan label detection — better tier name resolution for Claude OAuth usage (tier/plan/subscription_type/org fields) and new MiniMax plan label inference from quota totals. (#2498 — thanks @Gi99lin)

• fix(codex): fan out image n requests in parallel — when Codex requests n > 1 images, the image-generation handler now dispatches them concurrently instead of sequentially, significantly reducing total latency. (#2499 — thanks @nmime)

• fix(embeddings): strip stale Content-Encoding headers from upstream response — prevents clients from receiving gzip-encoded responses with identity encoding declared, which caused silent data corruption. (#2477 — thanks @lordavadon2)

• fix(model): return clear error instead of silent OpenAI default for unrecognized models — previously, an unrecognized model silently fell back to OpenAI; now returns a 404 with a descriptive message listing known providers. (#2492 — thanks @herjarsa)

• fix(dark-mode): correct background token on Compression Override select — the combo compression override <select> was using a hard-coded white background that was invisible in dark mode. (#2513 — thanks @apoapostolov)

• fix(antigravity): align subscription tier detection with Antigravity Manager — extractCodeAssistSubscriptionTier now parses the correct nested field from the loadCodeAssist response, and a new extractCodeAssistOnboardTierId fallback handles the onboarding flow. Subscription info is cached per access-token with 5-min TTL. (#2496 — thanks @Gi99lin)

• fix(opencode-zen): add opencode provider alias and sync model list with live API — opencode-zen and opencode-go are now also reachable via the shorter opencode alias, and the default model list is kept in sync with the live /v1/models catalog. (#2508 — thanks @herjarsa)

• fix(combo): clarify log message when combo target is skipped due to unavailable credentials — previously logged a misleading "provider not found" message; now says "skipped: credentials unavailable". (#2494 — thanks @herjarsa)

• fix(security): replace Math.random with crypto.randomUUID in generateTaskId/ActivityId and fix URL hostname check in test — eliminates weak PRNG usage flagged by CodeQL. (#2489)

• fix(electron): downgrade to Electron 41.x for better-sqlite3 V8 compatibility — Electron 42.x shipped a V8 version that broke better-sqlite3 native bindings at runtime; pinning to 41.x restores stability.

• fix(@omniroute/opencode-provider): include limit.context in model entries for OpenCode context window detection — OpenCode reads limit.context to determine usable context length for compaction and overflow detection.

• fix(providers): make gitlawb/gitlawb-gmi model entry optional — prevents provider initialization failure when the model is not available in the catalog. (#2476 — thanks @oyi77)

• fix(translator): inject omniroute_web_search in the Responses-API flat tool shape ({ type, name }) when the target provider speaks the Responses API — previously it was always emitted in the Chat Completions nested shape, so Codex/relay upstreams rejected the request. (#2390)

• fix(kiro): serialize non-string role:"tool" message content before sending to CodeWhisperer — structured/array tool output was collapsing to content:[{ text: "" }], which Kiro rejects with 400 Improperly formed request. (#2446)

• fix(claude): gate the heavy-agent beta headers (context-1m, effort, advanced-tool-use) on Opus/Sonnet only — Haiku with OAuth was receiving context-1m and rejecting it with 400. Also sanitizes historical thinking block signatures in passthrough. (#2454 — thanks @havockdev)

• fix(perplexity-web): route requests through a Firefox-148 TLS-impersonating client so Perplexity's Cloudflare edge stops rejecting VPS/datacenter IPs with a 403 challenge. (#2459 — thanks @havockdev)

• fix(validation): guard apiKey/modelsUrl against non-string values before calling .startsWith() / .trim() in the provider connection-test path. (#2463)

• fix(cost): prevent double-billing of cache_creation_input_tokens — prompt_tokens from token extractors already includes both cache_read and cache_creation, so nonCachedInput now subtracts both cache types to avoid pricing cache at the full input rate. (#2522 — thanks @herjarsa)

• fix(handler): always normalize system role messages in Claude passthrough paths — normalizeClaudeUpstreamMessages() is now called unconditionally in both compatibleBridge and pure passthrough, ensuring role:"system" messages are always extracted to the top-level system parameter. (#2519 — thanks @herjarsa)

• fix(handler): capture Gemini thought_signature in non-streaming response path — the non-streaming translator now captures thoughtSignature from Gemini thinking model parts and persists them so follow-up turns can resolve them correctly. (#2518 — thanks @herjarsa)

• fix(kiro): replace broken social OAuth with device flow — rewrites Kiro's Google/GitHub social login from the broken PKCE kiro:// custom protocol to AWS Cognito device flow, which works correctly in web/proxy environments. (#2524 — thanks @disonjer)

• fix(providers): resolve opencode/ → opencode-zen slug mismatch + add 40+ new models — opencode is now a proper alias for opencode-zen in executor, model resolver, and provider registry; adds GPT 5.x, Claude 4.x, Gemini 3.x, Grok, Kimi, and other models with tests. (#2517 — thanks @herjarsa)

• fix(antigravity): fail over stalled Antigravity sessions — new ANTIGRAVITY_PRE_RESPONSE_TIMEOUT_CODE shared constant for pre-response timeout detection, automatic failover to next account when session stalls before headers arrive. Node.js engine range relaxed to >=20.20.2. (#2464 — thanks @dhaern)

• fix(deepseek-web): fix SSE parser, prompt format, and error handling — handles all 3 DeepSeek SSE stream formats (initial fragments, APPEND operations, bare string tokens), simplifies prompt to single-turn to prevent chat marker leakage, and checks json.code before token extraction. (#2502 — thanks @ovehbe)

• fix(codex): accept auth.json without auth_mode field on import — Codex CLI no longer writes auth_mode; import now accepts both formats as long as required tokens are present. Semantic cache read now requires explicit temperature: 0. (#2536 — thanks @janeza2)

• fix(freetheai): add /chat/completions to baseUrl to resolve 404 errors. (#2557 — thanks @lordavadon2)

• fix(qoder): route PAT tokens to Qoder native API instead of DashScope — detects pt- prefixed tokens and routes to api.qoder.com with proper User-Agent header. (#2559 — thanks @herjarsa)

• fix(perf): cache compiled RegExp in RTK compression hot path — eliminates thousands of redundant new RegExp() instantiations per second. (#2553 — thanks @soyelmismo)

• fix(reasoning-cache): auto-start periodic cleanup on module load — the server-init.ts job was never imported (dead code), causing the reasoning_cache table to grow indefinitely. Now runs 30-min cleanup cycles automatically. (#2552 — thanks @soyelmismo)

• fix(claude): omit context-1m beta for Sonnet — restrict to Opus-only to avoid long-context credit gate errors. Add afk-mode-2026-01-31, replace redact-thinking with thinking-token-count-2026-05-13. (#2568 — thanks @unitythemaker)

• fix(codex): relax auth_mode check in frontend import preview — accept undefined/null/"chatgpt" instead of requiring "chatgpt" strictly, matching the backend fix in #2536. (#2567 — thanks @janeza2)

• fix(kimi): declare vision capability for Kimi K2.6 in all 4 layers — providerRegistry, modelSpecs, catalog.ts keyword list, and Playground VISION_MODELS; previously the model silently rejected image uploads. (#2573 — thanks @herjarsa)

• fix(dashboard): paginate request-log viewer beyond 300 rows — getCallLogs now accepts offset with parameterized SQL (eliminates string-interpolated LIMIT); RequestLoggerV2 grows its window via "Load more" + IntersectionObserver infinite scroll, resetting on filter change. (#2576)

• fix(cli): use /api/monitoring/health for server readiness check — waitForServer() was polling the auth-protected /api/health (401), causing omniroute serve to hang indefinitely. (#2578 — thanks @amogus22877769)

• fix(combo): detect invalid model errors via structured error codes + regex fallback — when a combo target rejects a model (e.g. free account vs Pro), the router now recognizes model_not_found / deployment_not_found codes and 6 regex patterns, and falls through to the next target instead of stopping the loop. (#2534 — thanks @HALDRO)

• fix(security): post-review hardening batch — spawnSync arg-array replaces execSync string-template (command injection), CSP unsafe-eval gated on !app.isPackaged, requireManagementAuth guard on budget/bulk and resilience/reset endpoints, error messages sanitized in gemini-web/claude-web/copilot-web/oauth/agents catch blocks, circuit breaker persists lastFailureKind, and combo resets exhaustedProviders per set-retry iteration. (#2435)

• fix(@omniroute/opencode-plugin): honor geminiSanitization and fetchInterceptor feature flags — both were applied unconditionally; now each fetch layer is gated by its flag (default ON), and disabling both falls back to plain SDK fetch. (#2546)

• fix(#2575): check DB feature flag override in arePrivateProviderUrlsAllowed() — supports runtime toggle without restart. (#2595 — thanks @herjarsa)

• fix(mimo): add supportsVision flag to MiMo-V2.5, V2.5-Pro, and V2-Omni — previously image uploads were silently rejected. (#2592 — thanks @herjarsa)

• fix(ops): propagate OMNIROUTE_SKIP_DB_HEALTHCHECK env var to periodic DB health check scheduler — companion fix to #2554. (#2591 — thanks @soyelmismo)

• fix(github): remove incorrect openai-responses targetFormat from GitHub Copilot's Haiku/Sonnet models. (#2583 — thanks @oyi77)

• fix(copilot): stabilize responses configuration — removes 865 lines of unstable config, simplifies handler. (#2579 — thanks @ivan-mezentsev)

• fix(#2544): add SSE heartbeat keepalive to Responses API transform stream — prevents Codex CLI 0.130.0 from disconnecting during long thinking/reasoning phases. (#2599 — thanks @herjarsa)

• fix(memory): extract system role messages in semantic passthrough path to prevent 400 on memory injection — system messages were being passed as-is to providers that reject mixed roles. (#2474 — thanks @Tentoxa)

• fix(@omniroute/opencode-provider): include limit.context in model entries for OpenCode context window detection — previously OpenCode couldn't determine model context size. (#2482 — thanks @herjarsa)

• fix(mimo): add supportsVision flag to Kimi K2.6 in providerRegistry + comprehensive vision tests for MiMo V2.5/V2.5-Pro/V2-Omni. (#2600 — thanks @herjarsa)

• fix(proxy): prefer scoped proxies over registry global fallback — legacy provider-specific proxy was being shadowed by a registry-global fallback across both storage backends. Resolution now follows strict specificity: account → provider → combo → global. (#2606 — thanks @terence71-glitch)

• fix(@omniroute/opencode-plugin): canonical-twin dedup + alias-fallback enrichment — /v1/models returned the same model under both alias (cc/claude-opus-4-7) and canonical (claude/claude-opus-4-7) names; now drops ~75 canonical duplicates and rescues ~88 raw-id rows with proper provider prefix via alias-index fallback. Also emits cost, release_date, modalities fields in static catalog and raises provider label threshold to 12 chars (preserves AssemblyAI, Antigravity verbatim). (#2607 — thanks @mrmm)

• fix(registry): populate empty models arrays for HuggingFace (6 models) and HackClub (3 models) + fix Snowflake placeholder baseUrl to {account} template pattern. (#2611 — thanks @oyi77)

🌐 Internationalization

• i18n(zh-CN): translate 830 missing UI strings — replaces all __MISSING__: placeholders with proper Chinese translations. (#2523 — thanks @InkshadeWoods)
• i18n(dashboard): add missing dashboard keys and fix EN fallbacks — hundreds of hardcoded English strings across cache, caveman, costs, skills, memory, and evals pages replaced with t() calls. (#2500 — thanks @Gi99lin)
• i18n(pt-BR): complete and fix Brazilian Portuguese translation — comprehensive overhaul of pt-BR locale with ~3000 lines of quality translations, filling all missing keys and correcting existing entries. (#2543 — thanks @alltomatos)
• i18n(ru): comprehensive Russian translation update — ~2000 lines of corrected and filled translations. (#2550 — thanks @AgentAlexAI)
• i18n(all): comprehensive localization and UI refactoring — 42 locale files synchronized with missing keys, cloud-agents page i18n rewrite, and consistent t() usage across 21 dashboard components. (#2580 — thanks @alltomatos)
• i18n(all): translate freeTier provider strings across 41 locales — replaces __MISSING__:Free Tier Providers placeholders with proper translations in both common and providers namespaces. (#2609 — thanks @leninejunior)
• i18n(pt-BR): eliminate all 1270 remaining __MISSING__ markers — completes pt-BR translation across 41 namespaces to true 100% coverage. (#2610 — thanks @leninejunior)

📝 Maintenance

• chore: remove Akamai VPS deploy from release workflow and skills.
• chore(deps): bump actions/setup-node from v4 to v6 + randomBytes security fix for cloud agent task IDs. (#2589)
• chore(deps): bump actions/upload-artifact from v4 to v7. (#2588)
• chore: ignore .claude/worktrees from git tracking.
• chore(ci): auto-lock release branch on version publish — new CI workflow applies lock_branch protection when a GitHub Release is published. (#2542)
• docs: redesign README — marketing-first layout with accurate provider counts. (#2490)

---

[3.8.1] — 2026-05-21

✨ New Features

• feat(settings): Feature Flags Settings Page (Card Grid + DB overrides) — fully implements the feature flags UI dashboard using Variant A (Card Grid) with Glassmorphism, complete with global GET/PUT/DELETE API routes, Zod validation, debounced search, category filters, and full 30+ locale i18n support. Resolves priority hierarchy to DB > ENV > Defaults. (#2457)
• feat(db): multi-driver SQLite abstraction layer — new SqliteAdapter interface with 3 concrete adapters (betterSqliteAdapter, nodeSqliteAdapter, sqljsAdapter) and a driverFactory that cascades better-sqlite3 → node:sqlite → sql.js (WASM). Enables OmniRoute to run on any JavaScript runtime (Node.js, Bun, Deno, Cloudflare Workers) without native binary dependencies. better-sqlite3 moved to optionalDependencies. (#2447)
• feat(settings): Claude Fast Mode toggle in Settings › AI — opt-in toggle that forwards X-CPA-Force-Fast-Mode header so a paired CLIProxyAPI build can reach Anthropic Fast Mode (speed:"fast"). Model-gated to Opus models matching Anthropic's binary KT() check. (#2449 — thanks @NomenAK)
• feat(settings): Codex Fast Tier — tier dropdown (default/priority/flex) + per-model gate preventing 400 errors from OpenAI when the tier toggle was on for non-Fast-eligible models. (#2451 — thanks @NomenAK)
• feat: align Antigravity 2.0.1 support — updated client profile, upstream headers, and model aliases. (#2443 — thanks @dhaern)
• feat: enhance extractBearer to support x-api-key for Anthropic API style auth. (#2436 — thanks @thedtvn)
• feat(memory): wire createMemory to upsertSemanticMemoryPoint (Qdrant). (#2439 — thanks @NomenAK)

🔧 Bug Fixes & Refactors

• fix(deepseek-web): rewrite auth to userToken Bearer + WASM PoW solver. (#2452 — thanks @ovehbe)
• chore: update node dependencies and runtime support. (#2453 — thanks @backryun)
• fix(translator): fix 3 Kiro tool_result defects causing 400 on follow-up turns — missing tool_use_id mapping, orphan result blocks, and conversation ID collision on assistant-first turns. (#2447)
• fix(translator): treat developer role as system in OpenAI → Claude translation — openAIToClaude now extracts developer-role messages into systemParts (same as system) and filters them from the non-system message list, preventing identity context injected via the Responses API developer role from silently becoming an assistant turn when routing to a Claude-format provider. (#2407)
• fix(antigravity): deduplicate removeHeaderCaseInsensitive — export canonical implementation from antigravityClientProfile.ts and remove the local copy in antigravity.ts; export AntigravityCredentialsLike type for cross-module use. (#2433 — thanks @Gi99lin)
• refactor(docs): enhance frontmatter handling in DocPage — gray-matter Date object parsing bug fix. (#2448 — thanks @ovehbe)
• fix(jules): Jules API parity and cloud-agent provider registration. (#2438)
• fix(i18n): harden diff key extraction tag sanitization in extract-keys-from-diff.mjs.
• chore(i18n): refresh fr/es/de locales + add missing settings.update key. (#2437)
• fix(dashboard): allow bracketed combo names — align dashboard combo-name validator regex with the shared/server schema updated in PR #2354; names like Claude [1m] are now accepted in the create/edit form. (#2458 — thanks @congvc-dev)
• docs(agentrouter): recommend native provider as the simple path — guide now prefers the built-in AgentRouter provider instead of manual OpenAI-compatible configuration. (#2429 — thanks @leninejunior)
• feat(settings): surface Codex Fast Tier toggle in Settings › AI — companion UI toggle for the Codex Fast Tier feature. (#2440 — thanks @NomenAK)

🔒 Security Fixes

• fix(security): replace execSync string-template with spawnSync arg-array in plugin.mjs — eliminates shell command injection via malicious plugin names.
• fix(security): gate Electron CSP unsafe-eval on !app.isPackaged instead of URL substring match — was leaking unsafe-eval into production builds; merged duplicate connect-src directives.
• fix(api): add requireManagementAuth to /api/usage/budget/bulk and /api/resilience/reset — both endpoints exposed spend data and circuit-breaker controls without auth.
• fix(security): route catch-block error messages through sanitizeErrorMessage() in gemini-web, claude-web, copilot-web executors, oauth route, and cloud-agent task routes — prevents stack traces and internal paths leaking into HTTP responses.
• fix(codex): refreshCredentials returns null (not error-object) on token refresh failure — prevents base executor from spreading {error} onto active credentials.
• fix(tokenRefresh): safe unknown-error access in catch block (error instanceof Error ? error.message : String(error)).
• fix(combo): reset exhaustedProviders set at start of each set-retry iteration — providers excluded in a failing pass now get a second chance on retry.
• fix(circuitBreaker): persist and restore lastFailureKind via the options JSON column — kind-based cooldown overrides (cooldownByKind) now survive server restarts.

---

[3.8.0] — 2026-05-06

🚀 Post-release hotfixes e contribuições (2026-05-06 → 2026-05-20)

2026-05-20

• feat(batch): implement 10 feature requests harvested from issues — T3 Chat Web executor (cookie-based), per-request exhausted-provider tracking (#1731) to skip quota-drained providers mid-combo, Zed Docker detection, API key rotator health dashboard, Kiro multi-account isolation, context-window model filtering, cost blending in combos, combo config tests, provider validation branches, and postinstall support scripts. (#2414)
• feat(combos): add falloverBeforeRetry strategy — combo routing now falls over to the next target before retrying the same model, eliminating the tail-latency spike from exhausting all per-model retries on a failing endpoint. Also wraps the retry loop in a setTry outer loop for per-target retry coordination. (#2417 — thanks @hartmark)
• fix(gamification): resolve 6 implementation gaps — missing SELECT in checkActionCountBadges SQL (was silently skipping 8 badges), federation leaderboard auth enforcement, pagination offset parameter no longer silently discarded, admin anomaly view now computes real z-scores, addXp correctly calculates initial level from XP amount, and barrel index.ts for clean module exports. 72-test suite covering all fixes. (#2421 — thanks @oyi77)
• docs: add AgentRouter provider setup guide — step-by-step instructions for connecting OmniRoute to AgentRouter.org's Claude-compatible relay endpoint, covering API key configuration and wire-image headers. (#2422 — thanks @leninejunior)
• fix(claude): drop orphan tool_result blocks left behind when fixToolAdjacency strips a dangling tool_use — resolves HTTP 400 "unexpected tool_use_id in tool_result blocks" from the Anthropic API on truncated histories. fixToolPairs now re-runs after every fixToolAdjacency pass across all three call sites (contextManager.ts, base.ts, claudeCodeCompatible.ts). (discussion #2410)
• fix(playground): guard against null/non-string model IDs in Playground dropdowns — typeof m?.id !== "string" check prevents a silent crash in the provider discovery loop and filteredModels computation that was leaving all Playground dropdowns empty when /v1/models returned entries with id: null; adds deduplication via Set to eliminate duplicate React key warnings.
• fix(mitm): point MITM runtime manager re-export to the compiled .js entrypoint — fixes module resolution after build when the .ts source is no longer present.
• fix(storage): persist STORAGE_ENCRYPTION_KEY across upgrades (closes #1622) — ensures that SQLite encryption keys are preserved during version upgrades. (#2428 — thanks @Chewji9875)
• fix(auth): auto-reset credential apiKeyHealth status on successful connection test. (#2427 — thanks @clousky2020)
• fix(mitm): drop .js extension on manager.runtime re-export to fix webpack packaging issue. (#2425 — thanks @NomenAK)
• fix(image): support Antigravity image generation and add Gemini 3.5 Flash support. (#2423 — thanks @backryun)

2026-05-19

• chore(i18n): comprehensive dashboard i18n coverage — 6 rounds of parallel refactoring replacing hardcoded English/Portuguese text with t() calls across 57+ dashboard pages; 420+ new keys added to en.json covering settings, playground, analytics, apiManager, providers, skills, memory, agents, and 15 other namespaces (coverage: ~88%, up from ~20%).
• fix(offline): avoid SSR/CSR hydration mismatch on the offline status page — switches from a useState lazy initializer (which accessed navigator.onLine on the server) to useSyncExternalStore with a distinct false server snapshot, eliminating the React hydration warning.
• fix(cli-tools): guard modelId type before calling .indexOf() — prevents a TypeError when a model entry without a string id reaches the comparison logic in CLI Tools.
• fix(providers): add missing isLocalProvider import and update changelog.
• fix(resilience): add API Key health tracking with automatic rotation and UI toast alerts. (#2412 — thanks @clousky2020)
• feat(providers): support Gemini API keys for Gemini CLI executor. (#2408 — thanks @benzntech)
• feat(gamification): implement Gamification & Leaderboard System with non-blocking event-driven updates. (#2405 — thanks @oyi77)
• fix(providers): Kilo Code provider no longer blocks on a missing local kilocode CLI binary — the provider uses OAuth device flow + direct HTTPS to api.kilo.ai and never required the CLI at runtime; the connection test was hard-failing with "Local CLI runtime is not installed" even when the OAuth token was valid. CLI Tools integration (/api/cli-tools/kilo-settings) keeps its own runtime check. (#2404 — thanks @Flexible78)
• fix(db): bun add -g omniroute (and other runtimes that skip postinstall) no longer surfaces a generic 500 — isNativeSqliteLoadError now also detects "Could not locate the bindings file" / MODULE_NOT_FOUND, so the user gets the friendly rebuild guide instead. (#2358 — thanks @yamansin)
• fix(kiro): enable Google OAuth login option in the Kiro auth modal — surfaces the Google SSO button alongside the existing identity providers. (#2392 — thanks @congvc-dev)
• fix(security): drop hashing layer in sessionPoolKey after switching to a non-cryptographic key derivation strategy that clears CodeQL alert #247. (#2396)
• feat(providers): Gemini Web cookie-based provider — proxies google.com chat through a session cookie, allowing free Gemini access without API keys. (#2380 — thanks @oyi77)
• model: add Composer 2.5 to the Cursor provider catalog. (#2381 — thanks @backryun)
• fix: tool_use without adjacent tool_result causes Claude 400 — adjacency guard now also applies inside compressContext. (#2383 — thanks @oyi77)
• build(deps): bump electron from 42.0.1 to 42.1.0 in /electron. (#2397)
• build(deps): production group bumps — 4 updates. (#2398)
• build(deps): development group bumps — 4 updates. (#2399)
• chore: sync release/v3.8.0 with main (CodeQL hotfixes + Dependabot bumps) via merge commit.

2026-05-18

• fix(security): resolve CodeQL alerts #243/#244/#245 — incomplete URL substring sanitization and weak crypto signal hardening. (#2391)
• fix(security): switch sessionPoolKey derivation to HMAC-SHA256 to clear CodeQL alert #246 (insecure hash for sensitive data). (#2394)
• docs(readme): restore the 9router acknowledgment that was inadvertently dropped during the v3.8.0 README rework. (#2393)
• refactor(dashboard): comprehensive nav, providers, endpoint, runtime, quota, pricing, budget redesign + quota sharing preview (sidebar restructure → 12 collapsible sections, 22 new routes). (#2384)
• fix(dashboard): PR #2384 follow-up review fixes — Runtime quota i18n, Budget projection logic, QuotaShare i18n externalization, Provider Limits semantic markup, bulk endpoint usage. (#2389)
• feat(content): add Haiper, Leonardo, Ideogram, Suno, and Udio as content/media providers. (#2377 — thanks @oyi77)
• feat(@omniroute/opencode-provider): expand config helpers, add MCP entry, live model fetch, and combo builder. (#2375 — thanks @mrmm)
• fix(claude-oauth): enable system-transforms pipeline for the native Claude executor (closes 400 billing-gate). (#2370 — thanks @thepigdestroyer)
• feat(content): extend providers with video, audio, TTS, music capabilities — Pollinations, MiniMax, Together, Replicate across Audio TTS and Transcription registries. (#2369 — thanks @oyi77)
• feat(providers): add Veo AI Free as a web-wrapper provider for generating video, image, and TTS without an API key. (#2366 — thanks @oyi77)
• feat(providers): add Replicate as a free provider for OpenAI-compatible inference with community models. (#2364 — thanks @oyi77)
• fix(claude): avoided redundant deep cloning of Claude Code messages during semantic passthrough preparation, improving memory/CPU efficiency for large histories. (#2362 — thanks @terence71-glitch)
• fix(providers): register llm7 in the executor registry and route Cohere via OpenAI-compatible layer. (#2361, #2360)
• fix(rate-limiter): Redis is now opt-in — when REDIS_URL is unset, the rate limiter falls back to the in-memory store instead of spamming ECONNREFUSED. (#2357)
• fix(streaming): emit protocol-aware stream errors — createDisconnectAwareStream() now emits native Responses API or Claude API SSE error blocks based on the client protocol. (#2355 — thanks @dhaern)
• fix(combos): allow bracketed combo names (e.g. Claude [1m]) by updating validation schemas. (#2354 — thanks @congvc-dev)
• fix(claude-code): semantic passthrough — preserve Claude Code messages[] structure for native Claude OAuth and relay routes. (#2351 — thanks @terence71-glitch)
• fix(usage): extract flat cached_tokens and reasoning_tokens from OpenAI-compatible usage objects. (#2350 — thanks @TF0rd)
• fix(translator): DeepSeek tool-call response lookup reads cached reasoning before falling back to empty string. (#2349 — thanks @herjarsa)
• fix(ui/tooltip): render in portal + clamp to viewport so tooltips aren't clipped in modal dialogs. (#2352 — thanks @slider23)
• fix(auto-routing): replace bare getSettings() with getCachedSettings to stop 500 on auto/* requests. (#2346)
• fix(docker): ship Dashboard Docs markdown in the container image. (#2348)
• fix(combo/validator): treat upstream responses carrying a non-empty reasoning_content as valid output. (#2341)
• fix(account-fallback): classify Anthropic Usage Limit Reached as QUOTA_EXHAUSTED with a 1h cooldown. (#2321)
• feat(providers): add GitHub Models as a free provider — GPT-5, o-series, DeepSeek-R1, Llama 4, Grok 3. (#2344 — thanks @oyi77)
• feat(providers): add Hackclub AI as a free provider — 30+ models, no credit card required. (#2339 — thanks @oyi77)
• feat(providers): add Microsoft Copilot Web executor — WebSocket-based provider. (#2340 — thanks @oyi77)
• feat(routing): LKGP stores last known good account connectionId alongside provider. (#2338 — thanks @oyi77)
• feat(dashboard): add Claude Code auth import/export UI + i18n (three-PR series: libs, API routes, dashboard UI).
• feat(dashboard): add Gemini CLI auth import/export UI + i18n (three-PR series: libs, API routes, dashboard UI).
• fix(routing): implement embedding combos, local provider validation bypass, and resolve migration collisions.
• fix(build): import Monaco ESM API to fix webpack nls.messages-loader error.
• fix(ui): v3.8.0 polish — connections border, sticky tabs, EN translations, save toasts, auto-combo catalog. (#2305 — thanks @mrmm)
• fix(auth+build): Bearer manage scope on management routes + lazy-load deepseek PoW solver. (#2308 — thanks @mrmm)
• fix(claude): guard orphan tool_use/tool_result pairs before upstream send. (#2312 — thanks @mrmm)
• fix(ui): remove count from batch removal button. (#2309 — thanks @hartmark)
• fix: remove implicit API key request caps — removes default 1K/5K/20K rate caps. (#2289 — thanks @josephvoxone)
• fix(sse): strip stale Content-Encoding, Content-Length, and Transfer-Encoding headers on non-streaming forward. (#2264 — thanks @gleber)
• chore(providers): refresh provider model metadata and ordering. (#2318 — thanks @backryun)
• chore(providers): consolidate Alibaba provider entries. (#2319 — thanks @backryun)
• fix(streaming): harden stream readiness detection. (#2317 — thanks @dhaern)
• fix(v1/messages): default to non-streaming when stream field is absent for Anthropic format. (#2326 — thanks @thepigdestroyer)
• fix(claude): fitThinkingToMaxTokens caps thinking budget to model's output ceiling. (#2327 — thanks @thepigdestroyer)
• fix(codex): Codex reasoning priority resolves modelEffort before explicitReasoning. (#2335 — thanks @terence71-glitch)
• fix(providers): providers page no longer deadlocks when no providers are configured. (#2329 — thanks @slider23)
• chore(providers): update HuggingFace to use the new /v1/ router endpoint. (#2322 — thanks @backryun)
• fix(security): resolve CodeQL ReDoS + URL sanitization alerts.
• fix(auth): stop retrying unrecoverable token refresh failures and include connection id in token health check credentials.
• fix(auth): return synthetic credentials for noAuth free providers and show no-auth card in dashboard instead of OAuth modal.
• fix(endpoint): replace nested <button> with <div role=button> in tunnel toggle rows to fix hydration warnings.
• fix(migrations): resolve version collision at migration slot 056 and add batch deletion API. (#2294 — thanks @hartmark)
• feat(batch): global rate-limit header cache with 60s TTL + 24h retry window. (#2299 — thanks @hartmark)
• feat(cc-bridge): config-driven per-provider system-block transform DSL. (#2286, closes #2260 — thanks @mrmm)
• feat(deepseek-web): full DeepSeek web API executor with Keccak PoW solver. (#2295 — thanks @oyi77)
• feat(i18n): add Azerbaijani (az / 🇦🇿) language support — new locale in config/i18n.json, 42 total supported languages.
• build(deps): bump actions/checkout from 4 to 6 in CI workflows. (#2288)

2026-05-17

• fix(codex): bulk import Codex auth.json — multi-file upload, paste-from-clipboard, and ZIP archive support. (#2343)
• feat(codex): import single Codex auth.json as an OAuth connection (one-click migration from Codex Desktop). (#2336)
• feat(codex-auth): rename export action + gate "Apply Local" behind a confirmation modal to prevent accidental local config overwrite. (#2332)
• fix(providers): providers page empty-state — missing i18n keys and "Add Provider" CTA so first-time users can add a provider. (#2333, #2337)
• fix(providers): Fix Providers empty state blocking first provider setup. (thanks @slider23)
• feat(providers): bulk add API keys with Single/Bulk tabs.
• feat(ui): comprehensive dashboard UX rework including simple/advanced modes for RTK/Caveman, human-readable error badges, InfoTooltip/PresetSlider shared components, sidebar subtitles, and provider category filters. (#2315, #2316 — thanks @oyi77)
• feat(provider): add Gitlawb Opengateway provider (xiaomi-mimo + gmi-cloud) with hasFree flag support. (#2314 — thanks @oyi77)
• feat(i18n): add simple/advanced mode keys and missing provider filter keys (allProviders, audioProviders, showFreeOnly).

2026-05-16

• feat(deepseek-web): full DeepSeek web API executor with PoW solver — also landed via PR #2295. (thanks @oyi77)
• feat(batch): global rate-limit header cache with 60s TTL — also via #2299.
• feat(cc-bridge): config-driven per-provider system-block transform DSL — also via #2286.
• feat(dashboard): provider summary card, free test button, sidebar order, i18n fix.
• feat(dashboard): A2A audit page, stats bar on MCP audit, sidebar deduplication.
• feat(skills): add 5 CLI skill manifests + AgentSkills / OmniSkills dashboard pages. (#2284)
• fix(translator): map developer → system by default for non-OpenAI-family providers. (#2281)
• fix(api/combos): add API-key-safe GET /v1/combos endpoint. (#2300)
• fix(embeddings/registry): add DeepInfra to the embedding provider registry. (#2298)
• fix(opencode-zen): flag qwen3.6-plus and qwen3.6-plus-free with targetFormat: "claude". (#2292)
• fix(settings): default debugMode to true on fresh installations.
• fix(sse): remove dead-code flag leak in claudeCodeToolRemapper. (#2290 — thanks @thepigdestroyer)
• fix(sse): strip stale Content-Encoding, Content-Length, Transfer-Encoding from upstream responses. (#2291 — thanks @thepigdestroyer)
• fix(migrations): resolve version collisions and add schema repair for quota thresholds.

2026-05-15

• feat(cli): CLI v4 — Commander.js architecture, 50+ commands, interactive TUI, full i18n (42 locales), plugin system (Fases 0–9). (#2280)
• feat(skills): publish 3 operational SKILL.md manifests + AI Skills dashboard entry. (#2276)
• feat(termux): Android/Termux headless support — auto-detect Android platform for headless mode. (#2273 — thanks @t-way666)
• feat(limits): per-window quota cutoffs across all providers with usage data. (#2267 — thanks @payne0420)
• feat(api-keys): configurable default rate limits via DEFAULT_RATE_LIMIT_PER_DAY env var. (#2266 — thanks @gleber)
• feat(authz): managementPolicy accepts API keys with manage scope. (#2265 — thanks @gleber)
• feat(mcp): MCP accessibility-tree smart filter engine — collapses ≥30 repeated sibling lines, 60-80% token savings.
• feat(auth): CLI machine-ID HMAC-SHA256 token for zero-friction local auth without JWT/password.
• feat(security): route protection tiers — 5 tiers: public/read-only/protected/always/local-only.
• feat(compression): Caveman SHARED_BOUNDARIES — all 6 languages × 3 intensities embed boundary clause.
• feat(runtime): dynamic SQLite 5-step fallback chain — bundled → runtime-installed → lazy-install → node:sqlite → sql.js.
• feat(cli): standalone system tray with PowerShell fallback on Windows (omniroute --tray).
• fix(providers/command-code): send required skills and stream payload fields. (#2271 — thanks @ddarkr)
• chore: ignore .playwright-mcp/ generated artifacts. (#2269 — thanks @backryun)
• chore: tidy up deprecated models from Windsurf provider registry. (#2279 — thanks @backryun)
• chore(deps): node dependency updates. (#2259 — thanks @backryun)
• build(deps): bump mermaid from 11.14.0 to 11.15.0. (#2178)

2026-05-08 a 2026-05-14

• feat(guardrails/vision-bridge): add VISION_BRIDGE_BASE_URL + VISION_BRIDGE_API_KEY env overrides for non-Anthropic vision-bridge routing. (#2232)
• feat(claude-web): implement session-based Claude Web executor with auto-refresh authentication. (#2283 — thanks @oyi77)
• refactor(@omniroute/opencode-provider): complete rewrite of the npm helper — tsup build (CJS + ESM + .d.ts), schema-correct output, baseURL deduplication, input validation, 13 unit tests. Versioned as 0.1.0.
• BREAKING: dropped Node 20.x support. Minimum Node version is now 22.22.2 (or 24.0.0+).
• fix(auth): accept x-api-key header in extractApiKey so Anthropic-native clients hit the same per-key policy enforcement. (#2225)
• fix(translator/claude-to-openai): stop including cache_creation_input_tokens in prompt_tokens. (#2215)
• fix(kiro): harden OpenAI-to-Kiro translator for API compliance. (#2251 — thanks @8mbe)
• fix(models): sync managed model aliases with provider model visibility. (#2250 — thanks @InkshadeWoods)
• fix(models/cleanup): align managed model cleanup for imported models. (#2261 — thanks @InkshadeWoods)
• fix(executor/claude-code): store tool-name round-trip metadata in non-enumerable _toolNameMap. (#2254 — thanks @Rikonorus)
• fix(streaming): strip upstream Content-Encoding, Content-Length, Transfer-Encoding headers from SSE responses. (#2253 — thanks @Rikonorus)
• fix(security): remediate CodeQL vulnerabilities (ReDoS, cryptographic bias, stack trace exposure, weak password hashing). (#216, #215, #211, #208, #206, #210)
• fix(providers/blackbox-web): add BLACKBOX_WEB_VALIDATED_TOKEN env override and 403 token-error disambiguation. (#2252)
• fix(auth): REQUIRE_API_KEY=false invalid Bearer no longer 401s the whole request. (#2257)
• feat(resilience): add model cooldowns dashboard card with real-time list, individual/bulk re-enable, and auto-refresh.
• feat(resilience): useUpstream429BreakerHints toggle. (#2133 — thanks @eleata)
• feat(auto): zero-config auto-routing with auto/ prefix — dynamic virtual combo from connected providers with 6 variant profiles. (#2131 — thanks @oyi77)
• feat(kiro): headless auth via kiro-cli SQLite, image support, tool overflow handling, model list sync. (#2129 — thanks @christlau)
• feat(cursor): surface Cursor Pro plan usage on provider-limits dashboard. (#2128 — thanks @payne0420)
• feat(mitm): dynamic Linux certificate path detection for multi-distro MITM cert trust. (#2134 — thanks @flyingmongoose)
• feat(1proxy): add dedicated settings tab with proxy rotation support. (#2135 — thanks @oyi77)
• feat(responses): degrade background: true to synchronous execution with a warning. (#2164 — thanks @Yosee11)
• feat(api): aggregate combo model metadata in catalog endpoint. (#2166 — thanks @faisalill)
• feat(oauth): complete Windsurf and Devin CLI OAuth + API-token flows. (#2168 — thanks @Zhaba1337228)
• feat(antigravity): support custom Google Cloud project ID. (#2227 — thanks @nickwizard)
• feat(cli): CLI Integration Suite — 5 new management commands, 3 API endpoints, config generators for 6 tools. (#2240 — thanks @oyi77)
• fix(sanitizer): preserve reasoning_content on assistant messages with tool_calls. (#2140 — thanks @DavyMassoneto)
• fix(catalog): ensure individual models expose context_length via getTokenLimit() fallback chain. (#2136 — thanks @herjarsa)
• fix(docker): remove docs directory from .dockerignore. (#2137, #2120 — thanks @hartmark)
• fix(providers): restore cloud agent provider exports and logger import. (#2138 — thanks @backryun)
• fix(providers): remove duplicate CLOUD_AGENT_PROVIDERS declaration. (#2141 — thanks @backryun)
• fix(translator): preserve body.system in openai→claude when Claude Code sends native format. (#2130)
• fix(authz): classify /dashboard/onboarding as PUBLIC to unblock setup wizard. (#2127)
• fix(i18n): complete Simplified Chinese translations. (#2115 — thanks @boa-z)
• fix(sse): classify hour quota errors as QUOTA_EXHAUSTED. (#2119 — thanks @clousky2020)
• fix(sse): fix CC-compatible streaming bridge. (#2118 — thanks @rdself)
• fix(cliproxyapi): detect Anthropic-shaped request bodies and route to /v1/messages. (#2165 — thanks @Brkic-Nikola)
• fix(claudeHelper): preserve latest assistant thinking blocks verbatim. (#2224 — thanks @NomenAK)
• fix(deepseek): preserve reasoning_content through full pipeline for DeepSeek V4 models. (#2231 — thanks @kang-heewon)
• fix(chatcore): stop leaking provider credentials in response headers.
• fix(export): exclude telemetry/usage-history tables from JSON config backups by default. (#2125)
• build(deps): regenerate package-lock.json to match http-proxy-middleware 4.x bump. (#2228 — thanks @NomenAK)

2026-05-06 a 2026-05-07 (lançamento inicial v3.8.0)

• feat(zed): Zed IDE Docker support — when OmniRoute runs in Docker and Zed is on the host, the Import flow now returns a 422 with zedDockerEnvironment: true and the dashboard auto-expands a Manual Token Import panel (new POST /api/providers/zed/manual-import endpoint with Zod validation). Includes Docker detection utility (/.dockerenv + cgroup heuristics) and a setup guide at docs/providers/ZED-DOCKER.md. ([#2306])
• feat(workflow): /implement-features gains pre-flight triage script (scripts/features/feature-triage.mjs) classifying open feature requests into 8 buckets — fresh issues (<14d) stay dormant to give the community time to react, engagement override (≥5 👍 or ≥3 unique non-bot commenters) absorbs early, already-delivered detection via merged PRs + CHANGELOG + git log closes issues with version + PR reference, stale need_details/ (>30d) is closed politely, aged defer/ (>90d) is re-evaluated, and externally-closed issues clean up _ideia/ automatically. Idea files now carry a YAML frontmatter snapshot enabling incremental comment re-sync. 53 unit tests cover the new logic.
• feat(providers): add GitHub Models as a free provider — GPT-5, o-series, DeepSeek-R1, Llama 4, Grok 3 with GitHub PAT auth and dynamic model fetch from api.github.com. (#2344 — thanks @oyi77)
• feat(providers): add Hackclub AI as a free provider — 30+ models, no credit card required, optional API key auth with passthrough model support. (#2339 — thanks @oyi77)
• feat(providers): add Microsoft Copilot Web executor — WebSocket-based provider translating OpenAI chat completions to Copilot's proprietary event protocol with per-token session pool isolation. (#2340 — thanks @oyi77)
• feat(routing): LKGP stores last known good account connectionId alongside provider — combo routing now prioritizes the exact connection that last succeeded, with graceful provider-level fallback for old records. (#2338 — thanks @oyi77)
• feat(i18n): add Azerbaijani (az / 🇦🇿) language support — new locale in config/i18n.json (source of truth), src/i18n/messages/az.json (UI strings), docs/i18n/az/ (full documentation set), README language bar, docs i18n index, and both translation pipeline scripts (generate-multilang.mjs, i18n_autotranslate.py). Total supported languages: 42.
• feat(limits): per-window quota cutoffs across all providers with usage data — operators can set per-quota-window thresholds (e.g. session=95%, weekly=80%) with cascading resolver (connection → provider default → global 98%) and zero-latency gate when nothing is configured. New migration 056, new GET /api/providers/quota-windows endpoint, and Dashboard › Limits cutoff modal. (#2267 — thanks @payne0420)
• feat(api-keys): configurable default rate limits via DEFAULT_RATE_LIMIT_PER_DAY env var — replaces hardcoded 1000/day fallback with Zod-validated configuration while preserving secure defaults for existing deployments. (#2266 — thanks @gleber)
• feat(authz): managementPolicy accepts API keys with manage scope — enables headless/programmatic management (provisioning providers, setting rate limits) without a browser session. (#2265 — thanks @gleber)
• feat(termux): Android/Termux headless support — auto-detect Android platform for headless mode (no browser open), move wreq-js and tls-client-node to optionalDependencies for ARM compatibility, lazy-load WS proxy with graceful 503 when unavailable, set GYP_DEFINES for better-sqlite3 ARM build, extended build timeout to 600s. (#2273 — thanks @t-way666)
• feat(deepseek-web): full DeepSeek web API executor with Keccak PoW solver (DeepSeekHashV1), SSE streaming, and auto-refresh session management via ds_session_id. (#2295 — thanks @oyi77)
• feat(cc-bridge): config-driven per-provider system-block transform DSL — operators can now configure system prompt transformations per-provider via Dashboard settings UI. (#2286, closes #2260 — thanks @mrmm)
• feat(batch): global rate-limit header cache with 60s TTL + 24h time-based retry window — shares rate-limit throttle state across sequential batches and uses time-based retry limits for robust large-batch processing. (#2299 — thanks @hartmark)
• feat(providers): improve Cohere provider support, expanding models and accurately updating OpenAI context limits. (#2313 — thanks @backryun)
• feat(claude-web): implement session-based Claude Web executor with auto-refresh authentication — enables direct Claude Web API access without an API key. (#2283 — thanks @oyi77)
• feat(skills): add 5 CLI skill manifests + AgentSkills / OmniSkills dashboard pages — enables external AI agents to discover and invoke OmniRoute capabilities. (#2284)
• feat(providers): add llama.cpp as local provider — llama-cpp (alias llamacpp) added to LOCAL_PROVIDERS and SELF_HOSTED_CHAT_PROVIDER_IDS; default base URL http://127.0.0.1:8080/v1; no API key required; uses the default OpenAI-compatible executor (#1980)
• feat(providers): bulk add API keys with Single/Bulk tabs.
• feat(provider): add Gitlawb Opengateway provider (xiaomi-mimo + gmi-cloud) with hasFree flag support. (#2314 — thanks @oyi77)
• feat(ui): comprehensive dashboard UX rework including simple/advanced modes for RTK/Caveman, human-readable error badges, InfoTooltip/PresetSlider shared components, sidebar subtitles, and provider category filters. (#2315, #2316 — thanks @dhaern, @oyi77)
• feat(i18n): add simple/advanced mode keys and missing provider filter keys (allProviders, audioProviders, showFreeOnly).
• feat(cli): full i18n support — 42 locales, --lang flag, config lang get/set/list commands for CLI language selection. (#2285)
• feat(claude-code): semantic passthrough for Claude Code /v1/messages payloads — preserves client messages[] structure (document blocks, tool_use/tool_result chains, cache_control, unknown content types) for native Claude OAuth and anthropic-compatible-cc-* relay routes, skipping broad normalization that could rewrite valid Claude Code semantics. (#2351 — thanks @terence71-glitch)

Changed

• CLI: Refactored architecture to use Commander.js as framework. Monolith bin/cli-commands.mjs (2853 lines) removed — commands now live individually in bin/cli/commands/. No breaking changes in normal usage; all previously listed subcommands continue working.

Removed

• bin/cli-commands.mjs — replaced by modular structure in bin/cli/commands/.

• bin/cli/index.mjs — replaced by bin/cli/program.mjs + bin/cli/commands/registry.mjs.

• bin/cli/args.mjs — replaced by Commander.js native parsing support.

• refactor(@omniroute/opencode-provider): complete rewrite of the npm helper. The 1.0.0 artifact was non-functional — index.js re-exported from .ts (unrunnable at install time) and the emitted shape didn't match the OpenCode https://opencode.ai/config.json schema. The new release ships a real tsup build (CJS + ESM + .d.ts), schema-correct output (npm: "@ai-sdk/openai-compatible", with models catalog), baseURL deduplication (no more /v1/v1), input validation, 13 unit tests, and full documentation in docs/frameworks/OPENCODE.md. Versioned as 0.1.0 to signal the pre-1.0 reset.

• chore(npm): @omniroute/opencode-provider@0.1.0 published to npmjs.com under the new @omniroute org. Install with npm install --save-dev @omniroute/opencode-provider.

• BREAKING: dropped Node 20.x support. Minimum Node version is now 22.22.2 (or 24.0.0+). Required because http-proxy-middleware 4.x requires node >=22.15.0. Users on Node 20 must upgrade — see package.json engines field and the README Node badge.

Security

• fix(oauth/windsurf): Windsurf Firebase token refresh now reads WINDSURF_CONFIG.firebaseApiKey instead of process.env.WINDSURF_FIREBASE_API_KEY directly.
• fix(kiro/translator): assistant-first conversations no longer collide on a single conversationId.
• fix(utils/publicCreds): decodePublicCred() no longer silently mangles raw credential overrides that don't match RAW_VALUE_PATTERN.
• fix(auth/extractApiKey): x-api-key fallback now only triggers when the request also carries an anthropic-version header.
• fix(providers/qoder): the OAuth+PAT disambiguation message now actually surfaces.
• fix(authz/clientApi): when REQUIRE_API_KEY=false, an invalid Bearer no longer 401s the whole request — falls through to anonymous (matching the "no auth required" semantics of the flag) with a single warning log carrying the masked key id. Fixes the surprise 401s that hit CLI integrations (Codex Desktop auto-config, Hermes Agent) that ship a stale Bearer in their saved config. (#2257)

Fixed

• fix(providers/llm7): add llm7 to the executor registry (open-sse/config/providerRegistry.ts). The provider was advertised in the dashboard catalog but missing from the executor table, so every connection test failed with a credential error. Now routes through the standard OpenAI-compatible https://api.llm7.io/v1/chat/completions endpoint with optional bearer auth. (#2361)
• fix(providers/cohere): switch the Cohere upstream from https://api.cohere.com/v2/chat (native shape) to https://api.cohere.com/compatibility/v1/chat/completions (OpenAI-compatible). The native endpoint returned { message: { content: [...] } } which the combo test validator could not read, surfacing as Provider returned HTTP 200 but no text content. (#2360)
• fix(combo/dispatch): add defensive typeof target.modelStr === "string" guards around the LKGP fallback findIndex and the combo test target builder. Combo entries whose modelStr failed to resolve at routing time (regression after #2338 added per-account LKGP) used to crash the request with TypeError: e.startsWith is not a function; we now surface a clean error instead. (#2359)
• fix(rate-limiter): Redis is now opt-in. When REDIS_URL is unset, the rate limiter and API-key auth cache fall back silently to the in-memory store instead of spamming connect ECONNREFUSED 127.0.0.1:6379 for every request. Connection-error logging is also deduped so docker logs no longer flood under sustained outage. Single-instance deployments work out of the box; multi-instance deployments continue to use Redis when REDIS_URL is provided. (#2357)
• fix(auto-routing): stop the ReferenceError: getSettings is not defined 500 that every auto / auto/* request raised. src/sse/handlers/chat.ts called the bare getSettings symbol without importing it; replaced with the already-imported getCachedSettings (same shape, plus the auto-routing hot path benefits from the cache). (#2346)
• fix(combo/validator): treat upstream responses carrying a non-empty reasoning_content (or reasoning) field as valid output, even when content is null. Reasoning models like moonshotai/Kimi-K2.5-TEE, zai-org/GLM-5-TEE, and the QwQ family put their answer in reasoning_content only — the quality validator was rejecting them with 502: empty content and triggering unnecessary combo fallbacks. (#2341)
• fix(docker): the Dashboard Docs viewer now actually has documents to show. .dockerignore was hiding every file under docs/ except openapi.yaml, so the in-product /docs/* viewer threw ENOENT: no such file or directory, open '/app/docs/...' for every page. We now ship the ~5 MB English markdown tree and still exclude the ~45 MB of translations/screenshots/raster diagrams that were the original optimization target. (#2348)
• fix(account-fallback): Anthropic OAuth (Claude Code Pro/Team) 429 responses carrying phrases like Usage Limit Reached, Claude Pro usage limit reached, or you've reached your usage limit are now classified as QUOTA_EXHAUSTED with a 1h cooldown instead of RATE_LIMIT_EXCEEDED with a ~5s transient backoff. Previously every Claude Pro account cascaded into a tight retry loop until the 5h subscription window genuinely reset. Also honors absolute ISO timestamps embedded in the error body (Try again at 2026-05-17T10:00:00Z) so the cooldown matches the upstream's stated recovery time. (#2321)
• fix(ui/tooltip): the shared <Tooltip> component now renders into a React portal anchored to document.body by default, so tooltips in modal dialogs (combo editor, etc.) are no longer clipped by overflow:hidden ancestors. Adds an optional multiline prop that swaps the legacy whitespace-nowrap clamp for max-w-xs whitespace-normal break-words when the label is long. Coordinates are clamped to the viewport so triggers near the right edge don't bleed off-screen. (#2352)
• fix(claude): avoided redundant deep cloning of Claude Code messages during semantic passthrough preparation, improving memory/CPU efficiency for large histories. (#2362 — thanks @terence71-glitch)
• fix(streaming): emit protocol-aware stream errors — createDisconnectAwareStream() now emits native Responses API (response.failed) or Claude API (event: error) SSE error blocks based on the client protocol instead of falling back to raw Chat Completions chunks, resolving upstream client parse failures on mid-stream disconnects. (#2355 — thanks @dhaern)
• fix(combos): allow bracketed combo names (e.g. Claude [1m]) by updating validation schemas and pinning exact combo lookup behavior before model suffix parsing. (#2354 — thanks @congvc-dev)
• fix(v1/messages): POST /v1/messages now defaults to non-streaming when the stream field is absent and the Anthropic source format is detected — prevents STREAM_EARLY_EOF errors from Anthropic SDK clients that omit the field per spec. (#2326 — thanks @thepigdestroyer)
• fix(claude): fitThinkingToMaxTokens caps thinking budget to the model's output ceiling — eliminates HTTP 400 from Anthropic when max_tokens + budget exceeds model limits (e.g. Opus 4.7's 128K ceiling). (#2327 — thanks @thepigdestroyer)
• fix(codex): Codex reasoning priority now resolves modelEffort before explicitReasoning — aligns with expected precedence and fixes suffix alias mismatches. (#2335 — thanks @terence71-glitch)
• fix(translator): DeepSeek tool-call response lookup reads cached reasoning before falling back to empty string — preserves reasoning content in multi-turn tool-call flows. (#2349 — thanks @herjarsa)
• fix(providers): providers page no longer deadlocks when no providers are configured — setup hint is shown instead of an empty filtered list, allowing the first provider to be added. (#2329 — thanks @slider23)
• fix(usage): extract flat cached_tokens and reasoning_tokens from OpenAI-compatible usage objects — providers like Xiaomi MiMo that return these as top-level fields instead of nesting in prompt_tokens_details/completion_tokens_details now properly surface in call logs and dashboard. (#2350 — thanks @TF0rd)
• chore(providers): update HuggingFace to use the new /v1/ router endpoint with dynamic model list (router.huggingface.co/v1/), removing the stale static model list. (#2322 — thanks @backryun)
• fix(security): resolve CodeQL ReDoS + URL sanitization alerts.
• fix(auth): stop retrying unrecoverable token refresh failures and include connection id in token health check credentials.
• fix(auth): return synthetic credentials for noAuth free providers and show no-auth card in dashboard instead of OAuth modal.
• fix(endpoint): replace nested <button> with <div role=button> in tunnel toggle rows to fix hydration warnings.
• fix(claude): guard orphan tool_use/tool_result pairs before upstream send, resolving a critical Anthropic 400 error on truncated histories. (#2312 — thanks @mrmm)
• fix(ui): remove count from batch removal button for cleaner interface. (#2309 — thanks @hartmark)
• fix(sse): strip stale Content-Encoding, Content-Length, and Transfer-Encoding headers on non-streaming forward — fixes JSON truncation on gzipped Gemini responses where clients honoring Content-Length read only the compressed byte count of the decompressed payload, causing "Unterminated string in JSON" parse failures. RFC 7230 §6.1 compliant. (#2264 — thanks @gleber)
• fix(executor/claude-code): store tool-name round-trip metadata in non-enumerable _toolNameMap so it survives in-memory but is stripped by JSON.stringify() — prevents internal OmniRoute metadata from leaking to upstream providers. (#2254 — thanks @Rikonorus)
• fix(streaming): strip upstream Content-Encoding, Content-Length, and Transfer-Encoding headers from SSE responses — prevents client-side decompression corruption when the proxy serves plain-text event streams through nginx/caddy. (#2253 — thanks @Rikonorus)
• fix(kiro): harden OpenAI-to-Kiro translator for API compliance: recursively strip additionalProperties and empty required: [] from tool schemas; merge consecutive assistant messages; prepend synthetic user for assistant-first conversations; convert orphaned tool results to inline text; enforce origin: "AI_EDITOR" on all history user messages; deterministic uuidv5 session caching. Closes #2213. (#2251 — thanks @8mbe)
• fix(models): sync managed model aliases with provider model visibility — remove aliases when models are hidden/deleted, skip alias creation for hidden models during sync, restore aliases when unhidden, cross-connection safety guard prevents pruning aliases still valid from another connection. (#2250 — thanks @InkshadeWoods)
• fix(models/cleanup): align managed model cleanup for imported models — provider-level "Delete All" now also removes synced available model storage; delete-alias button only shown for alias-source rows; compatible models section uses proper 3-way source-aware delete logic. (#2261 — thanks @InkshadeWoods)
• fix(auth): accept x-api-key header in extractApiKey so Anthropic-native clients (Claude Code, @anthropic-ai/sdk) hit the same per-key policy enforcement as Bearer clients. Previously these requests were treated as anonymous, bypassing model/budget/rate-limit policies and showing up as NULL in usage_history.api_key_id (~50% of traffic invisible in Costs/Analytics). Authorization: Bearer still wins when both are present (back-compat). (#2225)
• fix(translator/claude-to-openai): stop including cache_creation_input_tokens in prompt_tokens. Anthropic pads short prompts up to a 1024-token minimum on cache creation, so a 2-token "hi" could be reported as ~2008 prompt_tokens and inflate downstream billing (Sub2API/NewAPI/OneAPI) ~250x. prompt_tokens now matches the dashboard "Total In" (input + cache_read); cache_creation_tokens is exposed separately in prompt_tokens_details.cache_creation_tokens for auditing. (#2215)
• fix(ui/claude-extra-usage): clarify the toggle-success notification text to spell out the toggle→effect relationship ("Claude extra-usage blocking enabled/disabled" instead of the ambiguous "blocked/allowed"). (#2157)
• fix(providers/qoder): disambiguate the "Local CLI runtime is not installed" error when a user pastes a Personal Access Token but the connection is in OAuth/CLI-flavored mode. The test route now surfaces a single actionable message ("switch this connection to API Key auth") instead of cascading CLI + 401 errors. (#2247)
• fix(dashboard/api-manager): route custom OpenAI-/Anthropic-compatible provider IDs through getProviderDisplayName so the model grouping label shows Compatible (openai) instead of leaking the raw synthetic openai-compatible-chat-<uuid> value. (#2021)
• fix(providers/blackbox-web): add BLACKBOX_WEB_VALIDATED_TOKEN env override and 403 token-error disambiguation. Blackbox /api/chat started rejecting requests whose validated field didn't match the frontend tk token, even with a valid cookie + active subscription. Operators with the real token can now set the env var; otherwise the previous random-UUID fallback still ships, and a 403 with a token-specific body now surfaces a one-line "set BLACKBOX_WEB_VALIDATED_TOKEN" hint instead of the generic "cookie expired" message. (#2252)
• fix(guardrails/vision-bridge): add VISION_BRIDGE_BASE_URL + VISION_BRIDGE_API_KEY env overrides so non-Anthropic vision-bridge calls can be routed through OmniRoute's own /v1 self-loop, Google's Gemini OpenAI-compat endpoint, OpenRouter, or any other OpenAI-compatible URL — instead of being hardcoded to https://api.openai.com/v1 (which failed with 401 for users without an OpenAI key, even when they configured visionBridgeModel: "google/gemini-2.0-flash"). Anthropic models keep their dedicated path. (#2232)
• docs(security): document the ToS-violation hot spot of ANTIGRAVITY_CREDITS=always in STEALTH_GUIDE.md, including why it draws Google abuse detection more aggressively than free-tier-only usage and the recommended posture (=retry, Auto-Combo spread, per-connection RPM caps). (#2246)
• fix(translator/developer-role): convert OpenAI developer role → system by default for non-OpenAI-family providers. Codex/Responses API clients hitting DeepSeek (and other OpenAI-compatible gateways: MiniMax, Mimo, GLM, Fireworks, Together, etc.) were getting 400: unknown variant 'developer' because the previous default preserved developer for any targetFormat=openai upstream. New default: preserve only for openai/azure-openai/azure/github (and any id containing "openai"); convert everywhere else. Operators can still force preservation per-model via the dashboard "Compatibility → preserveOpenAIDeveloperRole = true" toggle. (#2281)
• fix(api/combos): add API-key-safe GET /v1/combos endpoint that mirrors the /v1/models auth model. Previously /api/combos was management-gated, blocking read-only integrations (e.g. opencode-omniroute-auth plugin) that need to enrich combo capabilities from a normal Bearer API key. The new endpoint projects only public metadata (name, strategy, model ids, providerId, description) — internal routing details like connectionId, weights, and labels are stripped. /api/combos (management) is unchanged. (#2300)
• fix(embeddings/registry): add DeepInfra to the embedding provider registry. Custom embedding models on the DeepInfra provider (e.g. Qwen/Qwen3-Embedding-8B, BAAI/bge-large-en-v1.5) were failing with Unknown embedding provider: deepinfra because the registry only included Nebius/OpenAI/Together/Fireworks/NVIDIA/etc. Now ships 8 popular DeepInfra embedding models out of the box and routes through https://api.deepinfra.com/v1/openai/embeddings. (#2298)
• fix(opencode-zen): flag qwen3.6-plus and qwen3.6-plus-free with targetFormat: "claude". The opencode-zen upstream returns Claude-format SSE bodies (type: "message_start", no choices array) for these Qwen3.6 models even when the request hits the OpenAI-compatible /chat/completions endpoint, causing client-side Zod failures (expected "choices" (array), received undefined). Routing them through the Claude /messages endpoint + translator fixes the format mismatch. (#2292)
• fix(settings): default debugMode to true on fresh installations — the Debug sidebar section (Translator, Playground, Search Tools) was hidden on new installs because debugMode was not in the settings defaults object, making data?.debugMode === true evaluate to false. The toggle in System & Storage appeared active but had no effect until manually set. Now all sidebar sections are visible out of the box.
• fix(providers/command-code): send required skills and stream payload fields — Command Code upstream wrapper now includes skills: "" and forces params.stream: true to align with upstream API requirements. Validation probe defaults to deepseek/deepseek-v4-flash. (#2271 — thanks @ddarkr)
• fix(sse): strip stale Content-Encoding, Content-Length, and Transfer-Encoding from upstream responses — prevents JSON truncation and ZlibError on gzipped provider responses forwarded through the proxy. (#2291 — thanks @thepigdestroyer)
• fix(sse): remove dead-code flag leak in claudeCodeToolRemapper — eliminates a stale boolean flag that could cause incorrect tool remapping behavior on subsequent requests. (#2290 — thanks @thepigdestroyer)
• fix(ui): v3.8.0 polish — connections border, sticky tabs, EN translations, save toasts, auto-combo catalog. (#2305 — thanks @mrmm)
• fix: remove implicit API key request caps — removes the default daily/weekly/monthly rate caps (1K/5K/20K) that silently applied 429s to API keys without explicit limits configured, causing unexpected throttling for operators who hadn't set custom rate policies. (#2289 — thanks @josephvoxone)
• fix(auth+build): Bearer manage scope on management routes + lazy-load deepseek PoW solver — unblocks MCP remote usage and Docker Next.js standalone builds. (#2308 — thanks @mrmm)
• fix(migrations): resolve version collision at migration slot 056 by renaming the quota thresholds migration to 057, and add batch deletion API with bulk cleanup support and batch/file management UI. (#2294 — thanks @hartmark)
• chore: ignore .playwright-mcp/ generated artifacts (CSP error logs, accessibility tree snapshots) — removes tracked test artifacts and adds the directory to .gitignore. (#2269 — thanks @backryun)
• chore: tidy up deprecated models from Windsurf provider registry. (#2279 — thanks @backryun)
• build(deps): bump actions/checkout from 4 to 6 in CI workflows. (#2288)
• build(deps): regenerate package-lock.json to match http-proxy-middleware 4.x bump. (#2228 — thanks @NomenAK)
• fix(streaming): harden stream readiness detection — recognize OpenAI Responses API lifecycle events (response.created, response.in_progress, response.output_item.added) and Chat Completions start chunks as readiness signals; switch GLM from idle timeout to readiness timeout; compact Provider Limits cutoff UI with i18n fallback labels; fix DeepSeek PoW dynamic import warning; static locale for docs prerender. (#2317 — thanks @dhaern)
• chore(providers): refresh provider model metadata, sort dashboard entries by display name, fix docs generator relative links and frontmatter. (#2318 — thanks @backryun)
• chore(providers): consolidate Alibaba provider entries — merge alicode/alicode-intl into shared ALIBABA_DASHSCOPE_MODELS array, update 42 i18n llm.txt files. (#2319 — thanks @backryun)
• chore: narrow .claude/ gitignore to runtime files only and untrack scheduled_tasks.lock.
• Docs: 270 broken internal markdown links repaired.

🏆 v3.8.0 Hall of Fame — extended credits (post-release)

The following contributions landed after the initial v3.8.0 cut and supplement the 55+ community hall of fame below. Updated tallies:

Contributor	New PRs in this cycle	Full v3.8.0 PR list
@oyi77	#2338, #2339, #2340, #2344, #2364, #2366, #2369, #2377, #2380, #2383	(+ already listed: #2010, #2014, #2041, #2052, #2061, #2074, #2091, #2094, #2096, #2131, #2135, #2240, #2283, #2295)
@backryun	#2269, #2279, #2313, #2318, #2319, #2381	(+ already listed: #1992, #2033, #2088, #2123, #2138, #2141, #2150, #2177)
@thepigdestroyer	#2326, #2327, #2370	(+ already listed: #2290, #2291)
@mrmm	#2375, #2286 (closes #2260), #2305, #2308, #2312	(consolidates the row)
@dhaern	#2315, #2316, #2317, #2355	(+ already listed: #2028, #2039, #2087, #2090)
@hartmark	#2294, #2299, #2309	(+ already listed: #2045, #2137)
@gleber	#2264, #2265, #2266	(+ already listed: #2103)
@herjarsa	#2349	(+ already listed: #2030, #2136, #2152)
@congvc-dev	#2354, #2392	(+ already listed: #2004)
@terence71-glitch	#2335, #2351, #2362	(new contributor — 3 PRs)
@TF0rd	#2350	(new contributor — 1 PR)
@slider23	#2329, #2352	(new contributor — 2 PRs)
@t-way666	#2273	(new contributor — 1 PR)
@payne0420	#2267	(+ already listed: #2082, #2128)
@Rikonorus	#2253, #2254	(new contributor — 2 PRs)
@8mbe	#2251	(new contributor — 1 PR)
@InkshadeWoods	#2250, #2261	(+ already listed: #2202)
@clousky2020	#2412	(+ already listed: 15 PRs)
@benzntech	#2408	(+ already listed: 8 PRs)

Thanks also to @app/dependabot for keeping our dependency tree current via #2178, #2228, #2288, #2397, #2398, #2399.

---

Detalhes completos — features do release (2026-05-15)

• feat(providers): expanded capabilities for Pollinations, MiniMax, Together, and Replicate across Video, Audio TTS, and Transcription registries. (#2369 — thanks @oyi77)
• feat(providers): added Veo AI Free as a web-wrapper provider for generating video, image, and TTS without an API key. (#2366 — thanks @oyi77)
• feat(providers): added Replicate as a free provider for OpenAI-compatible inference with community models. (#2364 — thanks @oyi77)
• feat(mcp): MCP accessibility-tree smart filter engine — collapses ≥30 repeated sibling lines, preserves [ref=eXX] anchors, 60-80% savings on browser snapshot outputs (Task 1)
• docs(skills): publish 10 SKILL.md manifests for external AI agents — zero-friction onboarding for Claude Desktop, ChatGPT, Cursor, Cline (Task 2)
• feat(cli): standalone system tray with PowerShell fallback on Windows — no Electron required; omniroute --tray; autostart via LaunchAgent/.desktop/registry (Task 3)
• feat(auth): CLI machine-ID HMAC-SHA256 token — zero-friction local auth without JWT/password; loopback-only; constant-time compare (Task 4)
• feat(security): route protection tiers — 5 tiers: public/read-only/protected/always/local-only; spawn-capable routes enforce loopback even with valid JWT (Task 5)
• feat(compression): Caveman SHARED_BOUNDARIES — all 6 languages × 3 intensities embed boundary clause; alreadyApplied check order fixed (Task 6)
• feat(runtime): dynamic SQLite 5-step fallback chain — bundled → runtime-installed → lazy-install → node:sqlite → sql.js; magic-byte validation (ELF/Mach-O/PE) (Task 7)
• docs/ux: tier 1/2/3 marketing, onboarding tour, dashboard widget — README tier diagram, docs/marketing/TIERS.md, TierTour onboarding step, Tier Coverage widget (Task 8)
• docs(comparison): OMNIROUTE_VS_ALTERNATIVES.md — objective comparison vs LiteLLM, OpenRouter, Portkey

Changed

• getDbInstance() requires prior ensureDbInitialized() call — server startup awaits it automatically (see release notes for migration)
• Caveman prompts embed SHARED_BOUNDARIES verbatim (LITE/FULL/ULTRA × 6 languages)
• README "Why OmniRoute?" enhanced with 3-tier ASCII diagram and comparison table
• Onboarding wizard gains "How It Works" tier tour step (after Welcome, before Security)
• Home dashboard shows "Tier coverage" widget (configured + active counts per tier)

Security

• Hard Rule #15: spawn-capable routes must call assertRouteAllowed(req) (CLAUDE.md)
• CLI token rejected on non-loopback hosts even when the HMAC is correct
• always-protected routes (shutdown, db export) reject CLI tokens unconditionally

Documentation

• docs/security/CLI_TOKEN.md
• docs/security/ROUTE_GUARD_TIERS.md
• docs/ops/SQLITE_RUNTIME.md
• docs/marketing/TIERS.md
• docs/comparison/OMNIROUTE_VS_ALTERNATIVES.md
• docs/releases/v3.8.0.md

---

Dependencies

• chore(deps): node dependency updates — bump multiple runtime and dev dependencies to latest patch/minor versions. (#2259 — thanks @backryun)

Detalhes completos — features e fixes do lançamento (2026-05-06 a 2026-05-14)

✨ New Features

• feat(providers): add Command Code provider (#2199 — thanks @ddarkr)
• feat(providers): add ModelScope provider-specific 429 handling and retry logic (#2202 — thanks @InkshadeWoods)
• feat(providers): update Gemini CLI provider models catalog (#2196 — thanks @nickwizard)
• feat(antigravity): integrate Antigravity provider with dynamic maxOutputTokens calculation, identity fingerprinting overhaul, and Cloud Code envelope payload sanitization (#2055, #2063)
• feat(gemini-cli): add custom projectId support for Gemini CLI transport (UI, DB, executor) (#1991)
• feat(providers): add KIE media provider support with dynamic polling, text models, and expanded video models catalog (#2009 — thanks @wauputr4)
• feat(providers): add Z.AI provider support with GLM quota handling and new quota labels — thanks @JxnLexn
• feat(providers): add 9 new free AI providers — LLM7, Lepton, Kluster, UncloseAI, BazaarLink, Completions, Enally, FreeTheAi (#2096 — thanks @oyi77)
• feat(providers): batch delete provider connections via checkbox multi-select (#2094 — thanks @oyi77)
• feat(cursor): full OpenAI parity — tool calls, streaming, and session management (#2082 — thanks @payne0420)
• feat(cursor): surface Cursor Pro plan usage on provider-limits dashboard (#2128 — thanks @payne0420)
• feat(cli): comprehensive CLI enhancement suite with 20+ new commands including omniroute providers, omniroute combos, omniroute doctor (#2074 — thanks @oyi77)
• feat(cli): add modular CLI setup and provider management commands (#2046 — thanks @wauputr4)
• feat(mcp): add DeepSeek quota and limit monitoring feature (#2089 — thanks @HoaPham98)
• feat(circuit-breaker): classify 429 errors and apply per-kind cooldowns (#2116 — thanks @eleata)
• feat(multi): manifest-aware tier routing — W1-W4 complete (#2014 — thanks @oyi77)
• feat(combos): add reset-aware routing strategy for quota-based providers — thanks @JxnLexn
• feat(combo): add context_length input field to combo edit form (#2047 — thanks @ddarkr)
• feat(combo): add fallbackDelayMs to combo configuration and related settings — thanks @JxnLexn
• feat(chat): dynamic tool limit detection with proactive truncation (#2061 — thanks @oyi77)
• feat(chat): add STREAM_READINESS_TIMEOUT_MS and integrate into chat handling — thanks @JxnLexn
• feat(chat): enhance error handling for semaphore capacity with fallback logic — thanks @JxnLexn
• feat(sse): refresh Claude OAuth wire image to claude-cli/2.1.131 (#2011 — thanks @Tentoxa)
• feat(github): add targetFormat: openai-responses to all GitHub models (#2122 — thanks @abhinavjnu)
• feat(api): allow configuration via API calls — open management routes to Bearer keys with manage scope (#2103 — thanks @gleber)
• feat(api): update API bridge proxy timeout to 600,000ms (#2019 — thanks @JxnLexn)
• feat(api): aggregate combo model metadata in catalog endpoint — buildComboCatalogMetadata() inlines contextLength, strategy, and target count for combo entries (#2166 — thanks @faisalill)
• feat(usage): add service tier breakdown, codex fast service tier analytics, and account for fast tier — thanks @JxnLexn
• feat(qdrant): embedding model discovery (#2086 — thanks @rafacpti23)
• feat(auth): per-session sticky routing for Codex (#1887)
• feat(oauth): complete Windsurf and Devin CLI OAuth + API-token flows — WindsurfExecutor (gRPC-web/protobuf), DevinCliExecutor (ACP JSON-RPC 2.0 over stdio), model alias map, OAuth provider config (#2168 — thanks @Zhaba1337228)
• feat(inworld): enhance Inworld TTS support (#2123 — thanks @backryun)
• feat(kiro): headless auth via kiro-cli SQLite, image support, tool overflow handling, and model list sync (#2129 — thanks @christlau)
• feat(auto): zero-config auto-routing with auto/ prefix — dynamic virtual combo from connected providers with 6 variant profiles (coding, fast, cheap, offline, smart, lkgp), analytics tab, and settings UI (#2131 — thanks @oyi77)
• feat(resilience): add model cooldowns dashboard card with real-time list, individual/bulk re-enable, and auto-refresh (#2146 — thanks @rafacpti23)
• feat(resilience): useUpstream429BreakerHints toggle — per-provider default policy for upstream 429 hint trust at the circuit-breaker cooldown layer with tri-state PATCH semantics (#2133 — thanks @eleata)
• feat(search): add Ollama Search as a web search provider with registry integration and validation (#2176 — thanks @andrewmunsell)
• feat(search): add Z.AI Coding Plan Search via MCP protocol integration (#2238 — thanks @andrewmunsell)
• feat(debug): configurable chat log truncation limits via environment variables (CHAT_LOG_TEXT_LIMIT, CHAT_LOG_ARRAY_TAIL_ITEMS, CHAT_LOG_MAX_DEPTH, CHAT_LOG_MAX_OBJECT_KEYS) and CHAT_DEBUG_FILE mode for untruncated JSON payloads (#2156 — thanks @bypanghu)
• feat(responses): degrade background: true to synchronous execution with a warning instead of throwing unsupportedFeature (#2164 — thanks @Yosee11)
• feat(mitm): dynamic Linux certificate path detection for multi-distro MITM cert trust (Debian, Arch/CachyOS, Fedora/RHEL, openSUSE) with NSS browser database injection (#2134 — thanks @flyingmongoose)
• feat(1proxy): add dedicated settings tab with proxy rotation support (#2135 — thanks @oyi77)
• feat(antigravity): support custom Google Cloud project ID for Antigravity provider (#2227 — thanks @nickwizard)
• feat(cli): CLI Integration Suite — 5 new management commands (config, status, logs, update, provider), 3 API endpoints, config generators for 6 tools (Claude, Cline, Codex, Continue, KiloCode, OpenCode), zero-config auto/ routing, and @omniroute/opencode-provider npm package (#2240 — thanks @oyi77)

🐛 Bug Fixes

• fix(pricing): make getPricingForModel fully case-insensitive to ensure custom prices correctly reflect in new incoming requests cost calculations
• fix(gemini): prevent functionDeclarations from being dropped by the sanitizer when googleSearch tool is present (#2077)
• fix(pollinations): add jsonMode: true flag in the request transformation to enforce correct JSON structure from Pollinations API (#2109)
• fix(docker): update Dockerfile to copy /docs directory during build ensuring API catalog availability at runtime (#2083)
• fix(docker): include OpenAPI spec in runtime image (#2007 — thanks @tatsster)
• fix(providers): strip OpenAI-specific fields in Kiro translator to prevent 400 errors (#2037)
• fix(kiro): normalize tool-use payloads to prevent 400 errors from agents (#2104 — thanks @rilham97)
• fix(kiro): merge adjacent user history turns after role normalization (#2105 — thanks @Gioxaa)
• fix(ui): resolve text contrast issues for zero-config warning banner in light mode (#2050)
• fix(core): inject global system prompt correctly into downstream chat completions pipeline (#2080)
• fix(core): restore Claude Code adaptive thinking defaults and resolve audio transcription CORS regression
• fix(routing): add missing v1beta rewrites to next.config to resolve 404 on Gemini models endpoint (#2102)
• fix(routing): fix bare GPT-5.5 routing for Codex-only installations (#2054 — thanks @guanbear)
• fix(routing): add fuzzy auto-combo routing for auto/* model prefix (#2010 — thanks @oyi77)
• fix(cache): optimize cache_control preservation logic and explicitly align tool schema with upstream Claude Code expectations
• fix(db): preserve legacy SQLite database path on Windows to prevent data loss (#1973)
• fix(db): reduce hot-path persistence overhead (#2039 — thanks @dhaern)
• fix(db): resolve migration conflict by renumbering overlapping migration entries (#2041 — thanks @oyi77)
• fix(settings): resolve model alias persistence double stringification preventing UI updates (#2018)
• fix(routing): dynamically filter bare model auto-resolution by active provider connections to prevent dead-routing (#2029)
• fix(embeddings): add Google Gemini embeddings compatibility via OpenAI-compatible endpoint mapping (#2006)
• fix(sse): prevent Claude OAuth multi-account correlation via metadata.user_id (#2053 — thanks @Tentoxa)
• fix(sse): prevent Claude Code identity cloak overrides and fix fallback resilience (#2053 — thanks @Tentoxa)
• fix(sse): classify hour quota errors as QUOTA_EXHAUSTED (#2119 — thanks @clousky2020)
• fix(sse): fix CC-compatible streaming bridge (#2118 — thanks @rdself)
• fix(antigravity): sanitize Claude Cloud Code payloads (#2090 — thanks @dhaern)
• fix(antigravity): add duplex half for streaming bodies — thanks @Gi99lin
• fix(antigravity): align identity protocol and behavior with official AM — thanks @Gi99lin
• fix(chatgpt-web): plumb proxy through to native tls-client (#2022, #2023 — thanks @xssdem)
• fix(codex): expose native model IDs in catalog (#2012 — thanks @Tr0sT)
• fix(glm): add dedicated coding transport (#2087 — thanks @dhaern)
• fix(compression): support Responses input and expand Spanish compression rules (#2028 — thanks @dhaern)
• fix(catalog): auto-calculate combo context_length from target model limits (#2030 — thanks @herjarsa)
• fix(api): fix usage analytics and API key identity (#2008, #2092 — thanks @AveryanAlex, @yoviarpauzi)
• fix(api-key): allow Unicode letters in API key name validation (#1996 — thanks @rodrigogbbr-stack)
• fix(auth): allow bootstrap without password (#2048 — thanks @tces1)
• fix(proxy): clean up proxy page redundancy and fix 1proxy sync empty body error (#2052 — thanks @oyi77)
• fix(dashboard): resolve Unknown plan display in Provider Limits — thanks @congvc-dev
• fix(usage): add extensible CURRENCY_SYMBOLS mapping for deepseek currencies
• fix(runtime): harden timer handling and model pricing fallback
• fix(i18n): complete Simplified Chinese translations (#2115 — thanks @boa-z)
• fix(mitm): add Linux cert install and skip sudo password when root (#1999 — thanks @NekoMonci12)
• fix(mitm): prevent stub from loading at runtime via bypass module — thanks @NekoMonci12
• fix: remove Anthropic-Beta header from non-Anthropic providers to fix identity contamination (#1989)
• fix(cli): resolve .env loading failure for global npm installations
• fix(authz): classify /dashboard/onboarding as PUBLIC to unblock setup wizard (#2127)
• fix(chatcore): stop leaking provider credentials in response headers
• fix(analytics): precise SQL matching for auto/ prefix models
• fix(export): exclude telemetry/usage-history tables from JSON config backups by default to prevent unbounded file growth (#2125)
• fix(translator): preserve body.system in openai→claude translator when Claude Code sends native Anthropic system array through /chat/completions — fixes v3.7.9 regression where system prompt was silently dropped, triggering Anthropic 429 (#2130)
• fix(sanitizer): preserve reasoning_content on assistant messages with tool_calls or function_call — fixes Kimi and other thinking-enabled providers returning 400 errors when reasoning_content was incorrectly stripped (#2140 — thanks @DavyMassoneto)
• fix(catalog): ensure individual (non-combo) models expose context_length via getTokenLimit() fallback chain — prevents OpenCode and other clients from falling back to conservative ~4000 token limit (#2136 — thanks @herjarsa)
• fix(docker): remove docs directory from .dockerignore so API catalog documentation is available at runtime inside containers (#2137, #2120 — thanks @hartmark)
• fix(types): systematic any type elimination across 8 core files — antigravity.ts, accountFallback.ts, usage.ts, geminiHelper.ts, error.ts, apiKeys.ts, settings.ts, logger.ts (#2137 — thanks @hartmark)
• fix(providers): restore cloud agent provider exports and logger import (#2138 — thanks @backryun)
• fix(providers): remove duplicate CLOUD_AGENT_PROVIDERS declaration, move Kiro dash→dot Claude model aliases to PROVIDER_MODEL_ALIASES, and trim deprecated Kiro registry entries (#2141 — thanks @backryun)
• fix: Follow OpenAI specification, handle throttling in batch and fix UI (#2045)
• fix(cliproxyapi): probe /v1/models for health when CPA 6.x has no /health endpoint (#2189 — thanks @Brkic-Nikola)
• fix(cliproxyapi): detect Anthropic-shaped request bodies and route to /v1/messages, strip Capy extras, and round-trip mcp_* tool name rewrites to Mcp_* (#2165 — thanks @Brkic-Nikola)
• fix(cliproxyapi): detect Anthropic shape on minimal Capy bodies (#2192 — thanks @Brkic-Nikola)
• fix(stream): skip [DONE] terminator for Claude SSE clients (#2190 — thanks @Brkic-Nikola)
• fix(claudeHelper): emit data field on redacted_thinking, drop bogus signature (#2191 — thanks @Brkic-Nikola)
• fix(modelSpecs): cap thinking budget for Claude Opus 4.6 / 4.7 / Sonnet 4.6 (#2197 — thanks @Brkic-Nikola)
• fix(reasoning-cache): include xiaomi-mimo in replay provider/model detection (#2198 — thanks @Brkic-Nikola)
• fix(kiro): synthesize minimal tools schema when body.tools is omitted but message history contains tool_calls, preventing 400 errors from Claude Code and OpenCode (#2149 — thanks @Gioxaa)
• fix(kiro): avoid treating high-traffic 429s as quota exhaustion — use classify429FromError to prevent premature account deactivation (#2153 — thanks @Gioxaa)
• fix(responses): propagate include array (e.g. reasoning.encrypted_content) during Chat→Responses API translation, fixing broken thinking panel in Codex/OpenCode (#2154 — thanks @Gioxaa)
• fix(responses): emit reasoning summary as delta.reasoning_content (flat) instead of delta.reasoning.summary (nested) for Chat Completions client compatibility (#2159 — thanks @Gioxaa)
• fix(cloudflare): add state file write serialization lock to prevent race conditions in cloudflaredTunnel.ts (#2156 — thanks @bypanghu)
• fix(providers): allow optional-key providers to pass connection test (#2169 — thanks @andrewmunsell)
• fix(providers): correct pollinations requests and provider dashboard state
• fix(providers): fix Azure AI Foundry provider connection handling (#2236 — thanks @one-vs)
• fix(providers/command-code): fix validation request format for Command Code API (#2243 — thanks @ddarkr)
• fix(antigravity): strip generationConfig.thinkingConfig for Claude models routed through Antigravity to prevent upstream errors (#2217 — thanks @NomenAK)
• fix(antigravity): bootstrap project via loadCodeAssist + fetchAvailableModels fallback for robust startup (#2219 — thanks @NomenAK)
• fix(rateLimit): never .stop() during runtime reset, evict cache instead to prevent stale rate-limit state (#2218 — thanks @NomenAK)
• fix(ModelSync): shared loopback readiness gate + IPv4 force to prevent model sync failures on dual-stack hosts (#2221 — thanks @NomenAK)
• fix(proxyFetch): retry once on undici dispatcher failure before native fallback (#2222 — thanks @NomenAK)
• fix(model): local aliases override cross-proxy provider inference to prevent incorrect model resolution (#2223 — thanks @NomenAK)
• fix(claudeHelper): preserve latest assistant thinking blocks verbatim to prevent Anthropic HTTP 400 errors (#2224 — thanks @NomenAK)
• fix(deepseek): preserve reasoning_content through full pipeline for DeepSeek V4 models — prevents reasoning context loss on multi-turn conversations (#2231 — thanks @kang-heewon)
• fix(sse-heartbeat): shape-aware keepalives keep streams alive through stricter proxies (#2233 — thanks @NomenAK)
• fix(translator): coerce submit_pr_review functionalChanges/findings to arrays to prevent upstream schema errors (#2242 — thanks @NomenAK)
• fix(api): validate model cooldown delete payload
• fix(ci): run coverage gate serially, align resilience and thinking checks, align cloud code thinking and model catalog tests

🔒 Security

• fix(security): remediate CodeQL vulnerabilities (ReDoS, cryptographic bias, stack trace exposure, and weak password hashing) (#216, #215, #211, #208, #206, #210)
• fix(security): sanitize error messages in API routes to prevent stack trace exposure (CodeQL js/stack-trace-exposure) (#2209)
• fix(security): remediate regex validation backtracking path in core compression cleanup (#1990)
• fix(core): harden input handling and stabilization for prompt compression edge cases

📝 Documentation

• docs: add competitive marketing tables and SEO/AEO optimizations to README (#2091)
• docs: refresh providers, model catalogs, and docs for v3.8.0 (#2088)
• docs: update Claude MD and update GLM-CN max context to 200k (#2027)
• docs(env): add GITLAB_DUO_OAUTH_CLIENT_ID to .env.example (#2031)
• docs: add Brazilian WhatsApp group link to README (#2201 — thanks @rafacpti23)

🔧 Improvements

• refactor(executor): sanitizeReasoningEffortForProvider() hook in BaseExecutor.execute() — downgrades xhigh→high for unsupporting providers, strips effort for mistral/devstral and github claude models (#2162 — thanks @hachimed)
• refactor(translator): remove redundant provider guard from Claude thinking placeholder injection — applies to all targetFormat === FORMATS.CLAUDE bodies (#2161 — thanks @JohnDoe-oss)
• refactor(catalog): remove 11 .ts extension imports, eliminate all as any casts, add CustomModelEntry interface and ComboModelStep type predicate, normalize alias resolution with resolveCanonicalProviderId() (#2152 — thanks @herjarsa)
• feat(resilience): useUpstream429BreakerHints tri-state PATCH field — true/false persists, null resets to undefined (omitted from JSON) (#2146 tests — thanks @rafacpti23)

🧹 Chores & Maintenance

• chore(providers): prune redundant local provider icon assets in favor of @lobehub/icons web fonts (#1992)
• chore(providers): remove deprecated models (#2033)
• chore(providers): improve BazaarLink and Completions.me support (#2177 — thanks @backryun)
• chore(registry): refresh contextLength and maxOutputTokens for claude, kiro, github, kimi-coding, xiaomi-mimo, codex/gpt-5.5 models (#2163 — thanks @brucevoin)
• chore(models): tidy up Alibaba Coding Plan base URL, reorganize Cursor model list by family, fix gpt-4o model ID, update OpenCode Zen model (#2150 — thanks @backryun)
• chore(deps): resolve npm audit moderate vulnerability (hono)
• chore(deps): move gray-matter from devDependencies to dependencies (runtime requirement) (#2156 — thanks @bypanghu)
• deps: bump fast-uri from 3.1.0 to 3.1.2 (#2078)
• deps: bump hono from 4.12.14 to 4.12.18 (#2065, #2079)
• deps: bump the development group with 6 updates (#2184)
• deps: bump electron-builder from 26.9.1 to 26.10.0 (#2183)
• ci: update build-fork workflow to build from main branch (#2055)
• ci: skip SonarCloud scan on main pushes to optimize CI time
• test: stabilize cooldown abort coverage case in integration testing
• build(deps): regenerate package-lock.json to match http-proxy-middleware 4.x bump (#2228 — thanks @NomenAK)
• fix(requestLogger): exempt tools field from array truncation for full debug visibility (#2234 — thanks @NomenAK)

🏆 v3.8.0 Community Contributors

Thank you to all 55+ community contributors who made v3.8.0 possible! 🎉

Contributor	PRs	Contributions
@NomenAK	12	#2217, #2218, #2219, #2221, #2222, #2223, #2224, #2228, #2233, #2234, #2242, #2192
@oyi77	14	#2010, #2014, #2041, #2052, #2061, #2074, #2091, #2094, #2096, #2131, #2135, #2240, #2283, #2295
@backryun	9	#1992, #2033, #2088, #2123, #2138, #2141, #2150, #2177, #2279
@Brkic-Nikola	6	#2165, #2189, #2190, #2191, #2192, #2197
@Gioxaa	5	#2105, #2149, #2153, #2154, #2159
@dhaern	4	#2028, #2039, #2087, #2090
@andrewmunsell	3	#2169, #2176, #2238
@ddarkr	4	#2047, #2199, #2243, #2271
@nickwizard	3	#1991, #2196, #2227
@herjarsa	3	#2030, #2136, #2152
@rafacpti23	3	#2086, #2146, #2201
@Tentoxa	2	#2011, #2053
@wauputr4	2	#2009, #2046
@hartmark	4	#2045, #2137, #2294, #2299
@payne0420	2	#2082, #2128
@bypanghu	2	#2027, #2156
@eleata	2	#2116, #2133
@Tr0sT	1	#2012
@AveryanAlex	1	#2008
@rodrigogbbr-stack	1	#1996
@NekoMonci12	1	#1999
@congvc-dev	1	#2004
@tatsster	1	#2007
@xssdem	1	#2023
@wucm667	1	#2031
@tces1	1	#2048
@guanbear	1	#2054
@Gi99lin	1	#2055
@ivan-mezentsev	1	#2063
@JxnLexn	1	#2019
@yoviarpauzi	1	#2092
@gleber	1	#2103
@rilham97	1	#2104
@boa-z	1	#2115
@rdself	1	#2118
@clousky2020	1	#2119
@abhinavjnu	1	#2122
@HoaPham98	1	#2089
@christlau	1	#2129
@flyingmongoose	1	#2134
@05dunski	1	#1978 (cherry-picked)
@DavyMassoneto	1	#2140
@Zhaba1337228	1	#2168
@faisalill	1	#2166
@Yosee11	1	#2164
@hachimed	1	#2162
@JohnDoe-oss	1	#2161
@brucevoin	1	#2163
@InkshadeWoods	1	#2202
@kang-heewon	1	#2231
@one-vs	1	#2236
@thepigdestroyer	2	#2290, #2291
@josephvoxone	1	#2289
@mrmm	3	#2286, #2305, #2308

[3.7.9] — 2026-05-03

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• feat(compression): major upgrade to Caveman and RTK compression pipelines (#1876, #1889):

  • Add RTK tool-output compression, stacked Caveman + RTK pipelines, compression combo assignments, dashboard context pages, MCP management tools, and language-aware Caveman rule packs.
  • Expand RTK parity with a 39-filter catalog, RTK-style JSON DSL stages, inline verify/benchmark coverage, trust-gated custom filters, expanded command detection, and redacted raw-output recovery.
  • Expose rule intensities, track USD savings, unify config validation, and persist MCP savings.
  • Expand Caveman parity and MCP metadata compression.

• feat(provider): update Jina AI model catalog to support Embeddings and Rerank natively (#1874 — thanks @backryun)

• feat(provider): add NanoGPT image generation provider (#1899 — thanks @Aculeasis)

• feat(ui): move proxy configuration to dedicated System → Proxy page (#1907 — thanks @oyi77)

• feat(ui): add K/M/B/T cost shortener utility (#1902 — thanks @oyi77)

• feat(providers): implement bulk paste for extra API keys (#1916 — thanks @0xtbug)

• feat(analytics): usage history API key backfill + dark mode pricing (#1896 — thanks @Gi99lin)

• feat(logs): show RTK and Caveman compression token savings accurately in request log UI (#1923 — thanks @emdash)

• feat(routing): auto-skip exhausted quota accounts (Issue #1952)

• feat(docs): docs site overhaul (#1976 — thanks @oyi77)

• feat(db): consolidate all database settings into SystemStorageTab (closes #1935) (#1947 — thanks @oyi77)

• feat(sse): codex 429 mid-task failover with account rotation (#1888 — thanks @smartenok-ops)

• feat(auto-assessment): add auto-assessment engine for combo self-healing (#1918 — thanks @oyi77)

• feat(usage): DeepSeek V4 native cache token extraction (#1930 — thanks @smartenok-ops)

• feat(cost): enhance cost formatting and add Codex GPT-5.5 pricing support (#1944 — thanks @JxnLexn)

🐛 Bug Fixes

• fix(auth): implement session affinity sticky routing logic
• fix(dashboard): derive display base URL from origin instead of hardcoding localhost (#1960 — thanks @jeanfbrito)
• fix(proxy): use credentials.connectionId instead of non-existent credentials.id for image proxy resolution (#1929 — thanks @Aculeasis)
• fix(routing): codex bare-name disambiguation + family-native fallback (#1933 — thanks @smartenok-ops)
• fix(infrastructure): move wreq-js to optionalDependencies and add Node 25/26 to secure runtime policy (#1924)
• fix(providers): resolve ChatGPT Web authentication failure by aligning TLS fingerprint User-Agent strings (#1925)
• fix(mitm): support root user for MITM sudo handling (#1948 — thanks @NekoMonci12)
• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941, #1945)
• fix(auth): fix Codex assistant final_answer response sanitization (#1965)
• fix(mcp): reclassify MCP endpoints to ensure API key authentication works even when dashboard auth is enabled (#1970)
• fix(providers): allow local OpenAI-compatible endpoints (like Ollama) to be added without an API key (fixes #1893)
• fix(providers): bypass AgentRouter unauthorized_client_error by spoofing Claude CLI headers via Anthropic endpoints (fixes #1921)
• fix(copilot): emit compatible reasoning text deltas (#1919 — thanks @ivan-mezentsev)
• fix(api-manager): show validation errors inline in modals, not behind (#1920 — thanks @andrewmunsell)
• fix(compression): align seeded standard savings combo with stacked default, preserve stacked defaults, and secure metadata routes.
• fix(gemini-cli): separate Cloud Code transport from Antigravity (#1869 — thanks @dhaern)
• fix(codex): map prompt field to input array for Cursor compatibility (fixes #1872)
• fix(core): align stream parameter default to false per strict OpenAI spec (fixes #1873)
• fix(ui): restore Next.js CSP unsafe-eval in production script-src to fix unresponsive Onboarding button (fixes #1883)
• fix(proxy): globally strip prompt_cache_retention in BaseExecutor to prevent upstream 400 errors from strict endpoints like droid/gemini-2-pro (fixes #1884)
• fix(ui): include isOpen dependency in EditConnectionModal state sync to ensure maxConcurrent is properly hydrated when reopening the modal (fixes #1859)
• fix(security): remediate 4 polynomial-redos CodeQL alerts in compression regexes by bounding repetitions and removing overlapping quantifiers
• fix(codex): flatten Chat Completions tool format to Codex Responses format in normalizeCodexTools — prevents Missing required parameter: tools[0].name upstream errors (#1914 — thanks @tranduykhanh030)
• fix(proxy): add proxy-aware execution context to image generation route — proxy settings are now correctly applied for image providers behind restricted networks (#1904 — thanks @Aculeasis)
• fix(translator): inject properties: {} into zero-argument MCP tool schemas during Anthropic→OpenAI translation — prevents 400 errors from OpenAI strict schema validation (#1898 — thanks @bryceIT)
• fix(codex): sanitize raw responses input (#1895 — thanks @dhaern)
• fix(combos): align strategy contracts (#1892 — thanks @dhaern)
• fix(combos): fix combo provider breaker profile handling (#1891 — thanks @rdself)
• fix(migrations): duplicate-column no-op fix (#1886 — thanks @smartenok-ops)
• fix(auth): per-connection OAuth refresh mutex (#1885 — thanks @smartenok-ops)
• fix(auth): require dashboard management auth for compression preview

🔄 Updates

• chore(provider): Add reka models list (#1956 — thanks @backryun)
• chore(model): Update new models, Delete Deprecated models (#1949 — thanks @backryun)

📝 Documentation

• docs(compression): document RTK+Caveman stacked savings ranges

🏆 Release Attribution & Retroactive Credits

• @payne0420 (PR #1828 / #1839) — Implementation of the Rate Limit Watchdog and environment overrides. (This feature was manually backported to v3.7.8, causing the automatic GitHub Release notes to omit the author's credit).

---

[3.7.8] — 2026-05-01

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): add Grok 4.3 and Xiaomi Mimo TTS provider (#1837)

• feat(core): implement Rate Limit Watchdog with environment override capability to detect and reset stalled queues (#1839)

• feat(providers): add muse-spark-web provider with multiple models and reasoning support (#1843)

• feat(1proxy): integrate 1proxy free proxy marketplace with dashboard management and new MCP tools (closes #1788) (#1847)

🐛 Bug Fixes

• fix(codex): sanitize Responses replay state to prevent internal assistant commentary from leaking (#1868 — thanks @dhaern)
• fix(cli): add capture-backed Gemini CLI fingerprint (#1866)
• fix(ui): hide combo compression controls when the global setting is disabled (#1840)
• fix(db): tolerate missing request_detail_logs table for legacy deployments (#1848)
• fix(core): remove unneeded `store` payload parameter for providers lacking support (closes #1841)
• fix(core): ensure safeOutboundFetch and A2A routers return 503 Service Unavailable when security guardrails are triggered
• fix(usage): correct Unix seconds vs milliseconds parsing logic for Kiro AI quota reset (closes #1849)
• fix(ui): apply robust NaN handling, ensure 24h consistency, and fix missing hour slots in Compression Analytics (closes #1844)
• fix(ui): implement short number formatting for token consumption metrics on cache pages to prevent overflow (closes #1842)
• fix(combo): stabilize provider routing at 500+ connections by bounding semaphore queues and adjusting circuit breaker tracking (closes #1846) (#1854)
• fix(maritalk): update Maritalk model list, use Authorization Key header, and align with latest API endpoints (#1856)
• fix(grok-web): stabilize tool calling (bash, readFile, webSearch) and response parsing by mapping native Grok intents to standard OpenAI payloads (#1857)
• fix(providers): correctly map and expose the Upstage embedding and chat model catalogs (#1855)
• fix(executor): apply proper urlSuffix and custom authHeaders for unknown registry-based providers in DefaultExecutor (closes #1846) (#1861)

🛠️ Maintenance

• fix(workflow): build docker images on version tags (#1838)

---

[3.7.7] — 2026-04-30

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• Prompt Compression Pipeline: Implemented a multi-phase prompt compression engine including lite (whitespace/duplication collapse), aggressive (summarization, tool compression), and ultra modes (heuristic pruning and SLM stub) (#1633, #1738, #1739, #1741)

• Compression Dashboard & Analytics: Added a compression settings UI, real-time log viewer, pipeline statistics tracking, and interactive playground preview (#1756)

• Compression Caching & MCP: Added caching-aware strategy adjustments to the compression pipeline, alongside new MCP tools for status and configuration (#1758)

• Analytics Custom Filters: Added custom date range selection, API key filtering, and NULL key analytics backfilling to the Costs Dashboard (#1830)

🐛 Bug Fixes

• Combo Routing: Fixed an issue where Gemini -preview models were incorrectly normalized to their canonical names, causing 404 errors during combo routing (#1834)
• Codex Native Passthrough: Added support for Cursor 5.5 sending messages arrays to the responses/compact endpoint, preventing upstream rejections with empty requests (#1832)
• Rate-limit Watchdog: Implemented a new rate-limit watchdog with environment override capabilities and Stage Tracing to prevent and diagnose silent wedges (#1828)
• Encryption Resiliency: Prevent sending encrypted tokens to providers by returning null on decryption failure (#763d353)
• i18n & Locales: Fixed OpenCode baseUrl locale placeholders and added compression keys across 32 languages
• Startup Stability: Hardened resilience integration server startup logic (#9aa89b17)

🛠️ Maintenance

• Tests & Docs: Expanded the test suite with 61 unit/integration tests for the compression pipeline and updated AGENTS.md
• Workflow: Fixed the changelog extraction logic to accurately capture GitHub release descriptions

---

[3.7.6] — 2026-04-30

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(api-keys): add rename support in the permissions modal — editable key name field with validation (#1796)

• feat(chatgpt-web): support thinking_effort parameter (Standard/Extended) for thinking-capable models (#1821)

• feat(dashboard): implement remaining v3.7.6 dashboard features — Costs overview, Translator pipeline, and Endpoint tabs improvements

• feat(tools): inject fallback tool names to prevent upstream 400 errors on providers that require tool names (#1775)

• feat(db): auto-restore probe-failed database on startup to prevent data loss after failed upgrades (#1810)

• feat(analytics): add cost-based usage insights and activity streaks in the analytics dashboard

🔒 Security

• fix(security): resolve ReDoS vulnerability in Codex executor regex patterns (#1797, #1789)

🐛 Bug Fixes

• fix(stability): resolve codex input validation, enable combo circuit breaker, and fix broken unit tests (#1804, #1805)
• fix(stability): safely cast inputs to strings before calling .trim() to avoid crashes on numeric fields in proxy modal (#1825)
• fix(stability): clear active requests and recover providers after connection failures (#1824)
• fix(xiaomi-mimo): update models to V2.5, fix Token Plan validation and default region (#1823)
• fix(codex): omit compact client metadata to prevent upstream rejections (#1822)
• fix(dashboard): fix endpoint visibility, A2A status display, and API catalog consistency (#1806)
• fix(analytics): use pure SQL aggregations — no history rows loaded into memory (#1802)
• fix(dashboard): correct loadPresets ReferenceError in CostOverviewTab
• fix(mitm): enforce transparent interception on port 443 only

🧹 Chores

• chore(workflow): mandate implementation plan generation in /resolve-issues workflow before coding
• chore(release): expand contributor credits to 155 PRs across full project history

🏆 Community Contributors Acknowledgment

We identified that 155 community PRs across the entire project history (from inception through v3.7.5) were manually integrated into release branches but closed instead of properly merged through GitHub, preventing contributors from receiving merge credit on their profiles. We sincerely apologize for this oversight and have since updated our workflows to ensure this never happens again.

The following contributors had their code and ideas integrated across multiple releases without proper merge credit. Thank you for your invaluable contributions to OmniRoute:

Contributor	PRs (Total)	All Contributions
@rdself	28	#542, #705, #717, #737, #738, #841, #851, #853, #875, #880, #888, #891, #903, #904, #974, #1069, #1089, #1196, #1267, #1272, #1299, #1300, #1356, #1357, #1441, #1443, #1549, #1742
@oyi77	27	#644, #672, #700, #850, #859, #862, #868, #874, #881, #883, #908, #926, #931, #983, #990, #1019, #1020, #1021, #1103, #1281, #1286, #1363, #1368, #1377, #1411, #1689, #1717
@clousky2020	15	#1244, #1323, #1365, #1366, #1408, #1442, #1484, #1595, #1598, #1599, #1611, #1618, #1620, #1621, #1644
@benzntech	8	#158, #1264, #1435, #1436, #1437, #1440, #1444, #1677
@kang-heewon	5	#530, #854, #884, #1235, #1574
@herjarsa	4	#1472, #1474, #1477, #1480
@backryun	4	#1358, #1609, #1627, #1722
@tombii	4	#708, #856, #900, #1013
@christopher-s	3	#868, #885, #992
@zen0bit	3	#561, #650, #912
@k0valik	3	#554, #587, #596
@zhangqiang8vip	2	#470, #575
@wlfonseca	2	#997, #1016
@RaviTharuma	2	#1188, #1277
@prakersh	2	#419, #480
@payne0420	2	#1593, #1670
@only4copilot	2	#855, #1039
@jay77721	2	#581, #582
@hijak	2	#295, #578
@hartmark	2	#1494, #1500
@defhouse	2	#906, #946
@xiaoge1688	1	#1304
@xandr0s	1	#1376
@willbnu	1	#882
@slewis3600	1	#1624
@sergey-v9	1	#594
@razllivan	1	#987
@nmime	1	#1271
@Moutia-Ben-Yahia	1	#1663
@Mind-Dragon	1	#467
@mercs2910	1	#1001
@MAINER4IK	1	#196
@luandiasrj	1	#996
@knopki	1	#1434
@kfiramar	1	#389
@ken2190	1	#166
@keith8496	1	#569
@jonesfernandess	1	#1118
@JasonLandbridge	1	#1626
@i1hwan	1	#1386
@Gorchakov-Pressure	1	#754
@foxy1402	1	#934
@dt418	1	#896
@dhaern	1	#1647
@DavyMassoneto	1	#211
@dail45	1	#1413
@congvc-dev	1	#1569
@be0hhh	1	#1581
@andruwa13	1	#1457
@AndrewDragonIV	1	#898
@AndersonFirmino	1	#362
@alexsvdk	1	#1280
@abhinavjnu	1	#550

---

[3.7.5] — 2026-04-29

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(tunnels): integrate native ngrok tunnel support with dashboard UI parity (#1753)

🐛 Bug Fixes

• fix(dashboard): add manual 'Clear All' button to terminate stalled long-running requests in Active Requests panel (#1799)
• fix(schema): remove empty string values from optional tool parameters to prevent upstream validation errors (#1674)
• fix(providers): ensure proper streaming cleanup and semaphore release to prevent stalls with nanoGPT (#1781)
• fix(db): wrap quota_snapshots access in try/catch to gracefully handle pending database migrations (#1784)
• feat(providers): add support for glm-cn (BigModel) provider (#1770)
• fix(grok-web): fix Grok validator and cookie parsing (#1793)
• fix(antigravity): scrub internal OmniRoute headers (#1794)
• fix(chatgpt-web): restore validator + expand model catalog to ChatGPT Plus tier (#1792)
• fix(codex): stabilize Copilot responses replay state (#1791)
• fix(antigravity): cap Claude bridge output tokens (#1785)
• fix(schema): strip default properties from tool-call JSON schemas during egress to prevent injection errors (#1782)
• fix(db): add quota_snapshots table to core DB schema initialization to prevent startup failures on fresh installs
• fix(models): apply blocked providers filter to non-chat catalog models (image, embedding, audio, etc.) (#1752)
• fix(antigravity): stabilize streaming payload parsing and deduplicate usage/model metadata refreshes (#1748)
• fix(antigravity): normalize Gemini bridge payloads — sanitize tool names, cap output tokens, and fix thinking budget (#1769)
• fix(sse): propagate AbortSignal to pre-fetch semaphore and rate-limit awaits to prevent memory leaks (#1771)
• fix(models): fix model sync import handling — separate synced models from custom models to prevent data loss (#1755)
• fix(codex): improve VS Code Copilot /responses reasoning and tool follow-ups (#1750)
• fix(memory): resolve build issues and implement memory UPSERT logic to prevent duplicate entries (#1763)
• fix(kiro): support organization IDC OAuth with regional endpoints and refresh (#1754)
• fix(combo): include 429 in provider circuit breaker to stop infinite retry loops on exhausted quotas (#1767)
• fix(claude): respect client-set thinking/effort params — only inject adaptive thinking and high effort when the client hasn't explicitly set them, preventing forced quota drain on Claude Max accounts (#1761)
• fix(blackbox-web): correct cookie name and populate session/subscription fields (#1776)
• fix(codex): align client identity metadata (#1778)
• fix(claude): fix support for claude-cli using Gemini provider (#1779)
• test(reasoning-cache): isolate DB state using mkdtempSync to prevent 401 middleware errors

🛠️ Maintenance

• chore(docs): add MseeP.ai security assessment badge to README (#1727)
• chore(xiaomi): update Xiaomi provider model list (#1759)
• chore(db): move DB health endpoint to management API (#1757)
• chore(ui): speed up endpoint initial render with background task loading (#1760)
• chore(workflows): add strict PR contributor credit policy to prevent future merge credit loss

---

[3.7.4] — 2026-04-28

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(ui): add endpoint tunnel visibility settings (#1743)

• feat(cli): refresh CLI fingerprint provider profiles (#1746)

• feat(proxy): implement bulk proxy import via pipe-delimited parser with update-or-create (upsert) logic and real-time preview table

• feat(pwa): add fullscreen installable PWA with manifest, service worker, and cross-platform app icons (#1728)

🔒 Security

• security: replace insecure Math.random with crypto.getRandomValues for fallback UUID generation to resolve CodeQL CWE-338 finding (#182)

🐛 Bug Fixes

• fix(cc-compatible): fix CC-compatible relay format and UI copy (#1742)

• fix(codex): normalize max reasoning effort for Codex routing (#1744)

• fix(claude-code): fix Claude Code gateway config helper (#1745)

• fix(db): reconcile legacy create_reasoning_cache migration tracking to prevent version shadowing on 032 and resolve startup warnings (#1734)

• fix(db): intercept 007 migration to use idempotent IF NOT EXISTS logic via PRAGMA table_info, preventing syntax crashes on fresh installs (#1733)

• fix(cc-compatible): preserve Claude Code system skeleton to prevent request rejection by strict compatible upstream providers (#1740)

• fix(providers): add API key validation for image-only providers and fix Stability AI requests to use multipart/form-data instead of JSON (#1726)

• fix(codex): preserve previous_response_id and conversation_id fields when input array is empty to prevent schema validation errors (#1729)

• fix(searxng): bypass UI validation block when apiKeyOptional is true and fix typing errors in provider dashboard to allow saving search providers without credentials (#1721)

• fix(proxy): disable HTTP keep-alive and pipelining in Undici proxy dispatcher to prevent "Socket hang up" rotation failures

• stream: correctly identify thought and error blocks in Antigravity/Gemini SSE streams to prevent premature 502 timeouts (#1725, #1705)

🛠️ Maintenance

• workflow: add phase 4 release monitoring instructions to /generate-release workflow
• test: fix typescript compilation errors in unit tests to keep CI typecheck pipeline fully green
• test: update responses store expectations for empty input arrays

---

[3.7.3] — 2026-04-28

🐛 Bug Fixes

• fix(claude): strip existing billing headers from system array before injecting to prevent Anthropic prompt cache misses — stacked x-anthropic-billing-header blocks invalidated prefix matching, causing ~100% cache_create instead of cache_read (#1712)
• fix(claude): strip output_config.format for non-Anthropic Claude-compatible providers during passthrough — third-party Claude endpoints (MiniMax, DeepSeek via aggregators) reject structured output fields with 400 errors (#1719)
• fix(combo): set terminal error state on response quality validation failure — prevents misleading ALL_ACCOUNTS_INACTIVE 503 when the real issue is response quality validation (#1707, #1710)
• fix(combo): treat combo fallback as target-level orchestration — all non-ok responses (including generic 400s) now fall through to the next target instead of being terminal; removes complex bad-request allowlist regex (#1713)
• fix(codex): restore namespace MCP tools and hosted-tool whitelist — regression from #1581 that silently dropped all MCP tool groups and Responses-API hosted tools (#1715)
• fix(codex): add neutral instructions for bare chat requests — Codex Responses backend rejects requests without instructions, making Codex unusable for normal chat (#1709)
• fix(proxy): wrap proxy assignment queries in try-catch for missing proxy_assignments table — Electron installs where migration 004 hasn't run no longer crash with no such table error (#1706)
• fix(migration): improve Windows file URL path resolution in migration runner — adds direct URL path extraction and process.cwd() fallback for CI-built bundles with leaked build-time paths (#1704)
• fix(ui): fix light mode active request payload modal — add missing --color-card theme token, use opaque bg-surface instead of translucent bg-card/70, add backdrop blur (#1714)

🔄 Updates

• chore(image-models): refresh image generation model registry — replace stale FLUX aliases with FLUX Kontext / FLUX.2 mappings, remove deprecated FLUX Redux/Depth/Canny variants (#1722)

---

[3.7.2] — 2026-04-28

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(authz): introduce centralized proxy-based authz pipeline and lifecycle policy (#1632)

• feat(logs): configure call log pipeline artifacts (#1650)

• feat(network): add guarded remote image fetch utility

• feat(codex): enable native Codex websocket responses on beta-gated models (#1658)

• feat(muse-spark-web): continue the same meta.ai conversation across turns (#1673)

🐛 Bug Fixes

• fix(responses): sanitize empty string placeholders from tool-call optional arguments in stream delta accumulation to avoid breaking strict clients (#1674)
• fix(codex): prevent unexpected protocol leakage and fabricated instructions on bare chat completion requests without tools (#1686)
• fix(executors): truncate tools array to 128 items max in GitHub Copilot and OpenCode executors to mitigate 400 Bad Request errors from upstream (#1687)
• fix: add body-read timeout to prevent stuck pending requests (#1680)
• fix(rate-limit): replace unsupported Bottleneck maxWait option with job-level expiration to prevent indefinite queue stalls (#1694)
• fix(sse): sanitize OpenAI tool schemas for strict upstream validators — strips null from enum arrays, normalizes tuple items, filters invalid required keys (#1692)
• fix(stream): fail zombie SSE streams before accepting response — returns 504 instead of hanging indefinitely, enables combo fallback (#1693)
• fix(combo): complete context truncation hotfix — cache getCombos() with 10s TTL, pass allCombosData to resolveComboTargets() for nested combo resolution, consolidate duplicated context overflow regex patterns (#1685)
• fix(codex): raise default quota threshold from 90% to 99% to avoid premature account blocking when usable quota remains (#1697)
• fix(memory): use user role for GLM/ZAI/Qianfan providers — providers with strict role constraints (no system role) now correctly receive memory context as a user message instead of a system message, preventing 422 validation errors (#1701)
• fix(oauth): target specific connection by ID on re-auth token exchange — prevents duplicate account creation when re-authenticating an existing OAuth connection (#1702 — thanks @namhhitvn)
• feat(email-privacy): integrate email visibility toggle in RequestLoggerV2 — log detail modal now respects global email privacy state, hiding email addresses by default (#1700 — thanks @namhhitvn)
• fix(combo): trigger fallback on Anthropic Invalid signature in thinking block errors instead of returning 400 directly (#1696)
• fix: combo retry loop stops immediately on client disconnect (499) (#1681)
• fix(search): support optional bearer auth for SearXNG (#1683)
• fix(vision): respect native GPT vision support — prevents VisionBridge from intercepting models that already handle images natively (#1678)
• fix(qwen): use security.auth format instead of modelProviders for Qwen Code config generation (#1677)
• fix(codex): remove stale websocket transport lookup that caused fallback errors (#1676)
• fix(chatgpt-web): bound tls-client native deadlocks so requests never hang forever (#1664)
• fix(codex): default gpt-5.5 to HTTP transport instead of WebSocket (#1660)
• fix(codex): [urgent] fix gpt-5.5 websocket transport and model labels (#1656)
• fix(grokweb): update Request and Response Specifications (#1655)
• fix(blackbox-web): set isPremium flag to true to enable premium model access (#1661)
• fix(core): avoid OpenAI stream options for Anthropic-compatible providers (#1654)
• fix(electron): resolve MCP server start failure on Windows (#1662)
• fix(electron): make Windows smoke test non-blocking (continue-on-error), pre-create userData dir for Windows + stream logs in CI, and add --no-sandbox and sandbox env for CI smoke tests
• fix(codex): fix getWreqWebsocket ReferenceError causing 502 on all Codex requests (#1652, #1653)
• fix(codex): default store to false — Codex OAuth backend rejects store=true (#1635)
• fix(db): add post-migration guards for missing batches table and combos.sort_order column on DB upgrades (#1648, #1657)
• fix(db): renumber duplicate migration 032 to prevent collision
• fix(perplexity-web): update API version and user-agent to match upstream requirements (#1666)
• fix(docker): copy SQLite migration files and explicitly trace in standalone build (#1665)
• fix(muse-spark-web): update to Meta's Ecto-era persisted query — fixes 502 Unknown type "RewriteOptionsInput" after Meta retired the Abra mutation (#1668)
• fix(dev): enable Turbopack by default and repair Codex CORS headers (#1669)
• fix(authz): restore REQUIRE_API_KEY support in clientApi policy
• fix(auth): align fallback API key format with test setup

🛠️ Maintenance

• build(prepublish): make Next.js build bundler configurable (webpack/turbopack)
• ci: align sonar analysis scope
• ci: stabilize release branch checks
• ci: remove expired advanced security scans job

🧪 Tests

• test: fix TypeScript configuration errors in plan3-p0.test.ts
• test: fix implicit any types across test suites
• test: disable type checking in flaky unit tests
• test: fix failing tests due to recent refactors
• fix(tests): align integration tests with authz pipeline refactor
• fix(tests): align test assertions with v3.7.2 source code changes
• fix(tests): CORS test now checks object body instead of entire file
• fix(e2e): fix E2E flakiness and implicit any type errors

---

[3.7.1] — 2026-04-26

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Add GPT-5.5 support to the Codex provider — includes 1.05M context window, tool calling, vision, and reasoning capabilities with proper pricing entries across cx and openai providers. Refactors splitCodexReasoningSuffix() into a shared helper for cleaner effort-level parsing (#1617 — thanks @Zhaba1337228).

• feat(cli): Add omniroute reset-encrypted-columns recovery command — nulls encrypted credential columns (api_key, access_token, refresh_token, id_token) in provider_connections while preserving provider metadata, giving users affected by #1622 a clean recovery path without losing configurations.

• feat(i18n): Expand locale coverage with nine new language packs (Bengali, Farsi, Gujarati, Indonesian, Marathi, Swahili, Tamil, Telugu, Urdu), bringing total language support from 32 to 41 locales.

🐛 Bug Fixes

• fix(rate-limit): Add per-model rate limiting for GitHub Copilot provider — a 429 on one model (e.g. gpt-5.1-codex-max) no longer locks the entire connection, matching the existing Gemini per-model quota pattern (#1624 — thanks @slewis3600).
• fix(cli-tools): Preserve existing OpenCode configuration (MCP servers, custom providers, comments) when saving OmniRoute settings — uses jsonc-parser for tree-preserving edits instead of destructive JSON roundtrip. Fix API key clipboard copy to use raw keys instead of masked placeholders. Add theme-aware OpenCode light/dark SVG logos (#1626 — thanks @JasonLandbridge).
• fix(cli-tools): Fix OpenCode guide step 3 {{baseUrl}} double-brace placeholder to use ICU-style {baseUrl} across all 41 locales, restoring next-intl interpolation (#1626).
• fix(codex): Make wreq-js native module import lazy and optional to prevent server crash on startup when the platform-specific binary is missing — affects pnpm installs, Docker Alpine, macOS ARM, and Windows (#1612, #1613, #1616).
• fix(i18n): Add 14 missing translation keys (logs.runningRequests, logs.model, logs.provider, logs.account, logs.elapsed, logs.count, logs.payloads, etc.) for the Active Requests panel across all locales. Replace 83 placeholder values in usage/evals namespace. Add 5 missing health namespace keys for rate limit status.
• fix(encryption): Prevent STORAGE_ENCRYPTION_KEY from being silently regenerated during npm install -g upgrades, which made all previously-encrypted provider credentials permanently unrecoverable due to AES-GCM auth-tag mismatch (#1622).
• fix(startup): Add decrypt-probe diagnostic at server bootstrap — if STORAGE_ENCRYPTION_KEY doesn't match encrypted credentials in the database, a prominent warning is logged directing users to restore the key or use the new recovery command.
• fix(cli-tools): Allow null API key values in cliModelConfigSchema to prevent 400 Bad Request errors when saving cloud-based CLI tool configurations. Fix error handling across all 10 ToolCard components to safely extract messages from structured error objects, preventing React Error #31 crashes.
• fix(docker): Set NPM_CONFIG_LEGACY_PEER_DEPS=true in the Docker builder layer before npm ci and remove duplicate postinstallSupport.mjs COPY instruction — fixes container image build failures introduced in v3.7.0 (#1630 — thanks @rdself).
• fix(antigravity): Hide deprecated Gemini-routed Claude 4.5 models from public catalogs and model lists. Legacy gemini-claude-* aliases now silently resolve to current Claude 4.6 equivalents. Replace dynamic reverse-alias generation with an explicit allowlist for predictable model visibility (#1631 — thanks @backryun).
• fix(types): Add explicit type annotations to sync-env test helpers and dynamic import casts to satisfy typecheck:noimplicit:core CI gate.
• fix(reasoning): Implement Reasoning Replay Cache — hybrid memory/SQLite persistence for reasoning_content in multi-turn tool-calling flows. Automatically captures reasoning from DeepSeek V4, Kimi K2, Qwen-Thinking, and GLM models and re-injects it on follow-up turns to prevent HTTP 400 errors from strict reasoning-content validation. Includes dashboard telemetry tab, REST API, and 21 unit tests (#1628 — thanks @JasonLandbridge).
• fix(postinstall): Extend postinstall native module repair to cover wreq-js — detects missing platform-specific .node binaries inside app/node_modules/wreq-js/rust/ and copies them from the root install. Fixes global pnpm installs on macOS arm64 where the standalone app directory only contained Linux binaries (#1634 — thanks @MarcosT96).
• fix(migration): Prevent compat-renamed migration slots from shadowing new migrations at the same version number. After rewriting 028_provider_connection_max_concurrent → 029, the runner now verifies the old version slot is clear, ensuring 028_create_files_and_batches runs on v3.6.x → v3.7.x upgrades. Adds batches table as a physical schema sentinel for upgrade recovery (#1637 — thanks @V8-Software).
• fix(registry): Route GitHub Copilot GPT 5.4/5.5 models through the Responses API (targetFormat: "openai-responses"). Fixes gpt-5.4-mini and gpt-5.4 being rejected on /chat/completions by GitHub (#1641 — thanks @dhaern).
• fix(usage): Correct MiniMax token plan quota display — the newer /v1/token_plan/remains endpoint reports used counts, not remaining counts. Rounds floating-point percentage artifacts in Provider Limits UI (#1642 — thanks @CruxExperts).
• fix(codex): Lazy-load wreq-js WebSocket transport via createRequire instead of top-level import. Server boots cleanly when native module is unavailable and returns 503 only when Codex WebSocket is actually requested. Fixes #1612 (#1640 — thanks @dendyadinirwana).
• fix(electron): Package Electron runtime dependencies into resources/app/node_modules/ via separate extraResources FileSet. Adds cross-platform packaged app smoke test script and CI integration to prevent future regressions. Closes #1636 (#1639 — thanks @prateek).
• feat(account-fallback): Add model-level daily quota lockout. When a provider returns 429 with quota_exhausted, cooldown is set to tomorrow 00:00 instead of exponential backoff. Detects daily quota patterns via isDailyQuotaExhausted() in chat handler (#1644 — thanks @clousky2020).
• fix(codex): Use per-conversation session_id/conversation_id from client body as prompt_cache_key instead of account-wide workspaceId. The official Codex CLI uses conversation_id (a unique UUID per session); using the shared workspaceId capped cache hit-rate at ~49%. Includes 10 unit tests (#1643).
• fix(claude): Stabilize billing header fingerprint to prevent Anthropic prompt-cache prefix invalidation. The fingerprint was derived from the first user message text, which changes every turn, mutating system[] and forcing ~100% cache_create. Now uses a stable per-day hash, preserving ~96% cache_read hit rate (#1638).
• fix(transport): Harden GitHub and Kiro streaming — thread clientHeaders through BaseExecutor.buildHeaders() to eliminate mutable singleton state race condition on concurrent requests. Remove redundant [DONE] stripping TransformStream from GitHub executor. Add defensive parseToolInput() for malformed Kiro tool call arguments. Hoist TextEncoder/TextDecoder to module singletons and use zero-copy subarray() (#1645 — thanks @dhaern).
• fix(transport): Prevent memory bloat and database exhaustion from large, fragmented streaming responses. Implemented ByteQueue in kiro.ts for zero-copy binary accumulation, refactored antigravity.ts for incremental SSE parsing, and enforced a strict 512KB tiered truncation limit (MAX_CALL_LOG_ARTIFACT_BYTES) on stream request logs and call artifacts (#1647).
• chore(ci): Update build environment dependencies — bump Node to 24.15.0, actions/checkout@v6, docker/build-push-action@v7, pin actions/setup-python to major tag (#1646 — thanks @backryun).

📝 Documentation

• docs(env): Add OMNIROUTE_ALLOW_PRIVATE_PROVIDER_URLS to .env.example with documentation for LM Studio and other local provider use cases (#1623).

---

[3.7.0] — 2026-04-26

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Add CrofAI as a built-in API-key provider with quota/usage monitoring wired into the dashboard Limits page (#1604, #1606).

• feat(skills): Add workspace-scoped built-in skills (file_read, file_write, http_request, eval_code, execute_command) with real sandbox execution via Docker, replacing stub responses. Browser skills now fail explicitly when runtime is not configured.

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(provider): add ChatGPT Web (Plus/Pro) session provider (#1593)

• feat(provider): add Baidu Qianfan chat provider (#1582)

• feat(codex): support GPT-5.5 responses websocket (#1573)

• feat(sse): Codex CLI image_generation + DALL-E-style image route (#1544)

• feat(dashboard): Complete the reconciled v3.7.0 dashboard task set: MCP cache tools and count, video endpoint visibility, provider taxonomy, upstream proxy visibility, provider count badges, costs overview, eval suite management, Custom CLI builder, ACP-focused Agents copy, Translator stream transformer, logs convergence, learned rate-limit health cards, docs expansion, and active request payload inspection.

• feat(mcp): Register omniroute_cache_stats and omniroute_cache_flush across MCP schemas, server registration, handlers, docs, and tests.

• feat(providers): Complete the v3.7.0 provider onboarding wave with self-hosted/local providers (lm-studio, vllm, lemonade, llamafile, triton, docker-model-runner, xinference, oobabooga), OpenAI-compatible gateways (glhf, cablyai, thebai, fenayai, empower, poe), enterprise providers (datarobot, azure-openai, azure-ai, bedrock, watsonx, oci, sap), specialty providers (clarifai, modal, reka, nous-research, nlpcloud, petals, vertex-partner), amazon-q, GitLab/GitLab Duo, and Chutes.ai.

• feat(providers): Add Cloudflare Workers AI integration and UI support for robust backend execution.

• feat(telemetry): Implement proactive public IP capture from client headers (x-forwarded-for, x-real-ip, etc.) within safeLogEvents for accurate database observability.

• feat(audio): Add AWS Polly as an audio speech provider with SigV4 request signing, static engine catalog, provider validation, managed-provider UI coverage, and sanitization for AWS secret/session fields.

• feat(search): Add You.com search provider support with dashboard discovery, validation, livecrawl option handling, and search handler normalization.

• feat(video): Add RunwayML task-based video generation support, task polling, provider catalog metadata, validation, and dashboard/model-list coverage.

• feat(providers): Add search functionality to the providers dashboard with i18n support. (#1511 — thanks @th-ch)

• feat(providers): Register 6 new models in the opencode-go provider catalog. (#1510 — thanks @kang-heewon)

• feat(providers): Add ModelScope provider (Chinese AI marketplace) with Kimi K2.5, GLM-5, and Step-3.5-Flash integration. (#1430 — thanks @clousky2020)

• feat(providers): Add LM Studio as an OpenAI-compatible local provider for self-hosted model inference.

• feat(providers): Add Grok 4.3 thinking model support for xAI web executor requests.

• feat(core): Implement provider-level Circuit Breaker to prevent cascading failures across connections, enforcing a 10-minute cooldown after 5 consecutive transient failures. (#1430)

• feat(core): Add daily quota exhaustion lock to detect "quota exceeded" signals and lock the specific model until midnight. (#1430)

• feat(core): Auto-inject stream_options.include_usage = true for OpenAI format streams to guarantee token usage is reported correctly during streaming. (#1423)

• feat(core): Add OpenAI Batch Processing API support — submit, monitor, and manage batch jobs through the proxy with full lifecycle tracking.

• feat(vision-bridge): Add automatic image description fallback for non-vision models via VisionBridgeGuardrail (priority 5). Intercepts image-bearing requests to non-vision models, extracts descriptions via a configurable vision model (default: gpt-4o-mini), and replaces images with text before forwarding. Fails open on any error. (#1476)

• feat(dashboard): Introduce real-time model status badges with countdown timers in the provider detail and combo panel interfaces. (#1430)

• feat(dashboard): Add Batch/File management data grid with full i18n translations for batch processing workflows. (#1479)

• feat(usage): MiniMax + MiniMax-CN quota tracking in provider limits dashboard. (#1516)

• feat(providers): Fix OpenRouter remote discovery and unify managed model sync. (#1521)

• feat(providers): Implement provider and account-level concurrency cap enforcement (maxConcurrent) using robust semaphore mechanisms. (#1524)

• feat(core): Implement Hermes CLI config generation and message content stripping. (#1475)

• feat(combos): Add expert combo configuration mode for advanced routing controls. (#1547)

• feat(providers): Register Codex auto review and expand icon coverage.

• feat(tunnels): Add Tailscale tunnel management routes and runtime helpers for install, login, daemon start, enable/disable, and health checks.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(chatgpt-web): Fix empty-file race in tlsFetchStreaming where waitForFile accepted zero-byte files, silently degrading streaming requests to buffered mode. Replaced with waitForContent requiring file.size > 0 with early exit on request settlement. (#1597 — thanks @trader-payne)

• fix(chatgpt-web): Fix stale NextAuth session-token cookies surviving rotation shape changes (unchunked↔chunked). mergeRefreshedCookie now drops all session-token family members via SESSION_TOKEN_FAMILY_RE before appending the refreshed set, preventing auth failures from dual cookie submission. (#1597 — thanks @trader-payne)

• fix(codex): WebSocket memory retention and weekly limit handling (#1581)

• fix(providers): Default models list logic (#1577)

• fix(ui): Dashboard endpoint URL hydration respects NEXT_PUBLIC_BASE_URL when behind a reverse proxy (#1579)

• fix(providers): Restore strict PascalCase header masquerading for Claude Code to resolve HTTP 429 upstream errors (#1556)

• fix(sse): make Responses passthrough robust for size-sensitive clients (#1580)

• fix(codex): update client version for gpt-5.5 (#1578)

• fix(vision-bridge): force GPT-family image fallback (#1571)

• fix(claude): skip adaptive thinking defaults for unsupported models (#1563)

• fix(claude): preserve tool_result adjacency in native and CC-compatible paths (#1555)

• fix(reasoning): Preserve OpenAI Chat Completions reasoning_effort through assistant-prefill requests and label OpenAI request protocols explicitly as OpenAI-Chat or OpenAI-Responses. (#1550)

• fix(codex): Fix Codex auto-review model routing so review traffic resolves to the intended configured model. (#1551)

• fix(resilience): Route HTTP 429 cooldowns through runtime settings so cooldown behavior follows the configured resilience profile. (#1548)

• fix(providers): Normalize Anthropic header keys to lowercase in the provider registry to avoid duplicate or case-variant upstream headers. (#1527)

• fix(providers): Preserve audio, embedding, rerank, image, video, and OpenAI-compatible alias metadata when /v1/models merges static and discovered catalogs.

• fix(providers): Discover Azure OpenAI deployments from resource endpoints using api-key auth and configurable API versions.

• fix(providers): Keep local OpenAI-style providers authless when no API key is configured, including the Lemonade Server default endpoint.

• fix(translator): Preserve Antigravity default system instructions and caller-provided system prompts as separate Gemini systemInstruction parts instead of concatenating them.

• fix(security): Sanitize provider-specific AWS secrets and session tokens from provider management API responses.

• fix(release): Resolve combo prefixing, Electron packaging, CLI auth, and release-branch integration regressions. (#1471, #1492, #1496, #1497, #1486)

• fix(providers): Resolve 400 errors for GLM and Antigravity Claude adapter during request translation by scoping prompt caching to compatible Anthropic endpoints and flattening system instructions. (#1514, #1520, #1522)

• fix(core): Strip reasoning_content from OpenAI format messages for non-reasoning models to prevent upstream HTTP 400 validation errors. (#1505)

• fix(sse): Map Claude output_config/thinking to OpenAI reasoning_effort for proper Antigravity tool translation. (#1528)

• fix(combo): Fallback to next model on all-accounts-rate-limited (HTTP 503/429) to maintain high availability. (#1523)

• fix(api): Harden batch and file endpoints for auth and recovery to prevent schema state collisions.

• fix(ui): Add missing UI wiring for "Add Memory" and "Import" buttons on the /dashboard/memory page. (#1506)

• fix(ui): Prevent Dark Mode FOUC (Flash of Unstyled Content) by injecting a synchronous theme initialization script into the root layout.tsx.

• fix(ui): Fix mobile layout text overflow in provider and combo cards, and enable touch-friendly reordering arrows across all combo strategies.

• fix(core): Add periodic runtime log rotation checks to prevent disk exhaustion in long-running instances. (#1504 — thanks @ether-btc)

• fix(build): Resolve missing process module in webpack client build for pino-abstract-transport. (#1509 — thanks @hartmark)

• fix(ui): Add dark mode support for native dropdown <option> elements on Linux/Windows, resolving invisible text in settings and combo builders (#1488)

• fix(batch): Add batch item dispatching to specific handlers based on URL to support embeddings and other modalities (#1495 — thanks @hartmark)

• fix(dashboard): Correct TOML round-trip corruption in Codex config serializer by dequoting keys and preserving array/boolean structures properly. (#1438 — thanks @benzntech)

• fix(security): Resolve CodeQL alert 164 (ReDoS in extraction) and 163 (incomplete URL sanitization). (#163, #164)

• fix(providers): Add optional chaining to connection object before accessing providerSpecificData, preventing runtime errors when the connection is null/undefined.

• fix(codex): Preserve namespace MCP tools forwarded to Codex Responses API, preventing tool name stripping during translation. (#1483)

• fix(codex): Deduplicate case-variant anthropic-version header in Claude Code patch to prevent duplicate header injection. (#1481)

• fix(fallback): Use shared CircuitBreaker instead of undefined constants, fixing runtime errors in provider failure handling. (#1485)

• fix(fallback): Merge new provider failure threshold fields (providerFailureThreshold, providerFailureWindowMs, providerCooldownMs) into resilience profiles.

• fix(fallback): Remove 429 from PROVIDER_FAILURE_ERROR_CODES — rate limits are already handled by model-level and account-level locks; including them in the provider-wide circuit breaker caused premature cooldown.

• fix(sse): Enable tool calling for GPT OSS and DeepSeek Reasoner models. (#1455)

• fix(encryption): Return null on decryption failure to prevent sending encrypted tokens to providers. (#1462)

• fix(combo): Resolve cross-provider thinking 400 errors and HTTP clipboard issues during combo routing. (#1444)

• fix(core): Resolve skills, memory, and encryption system issues affecting startup and runtime stability. (#1456)

• fix(core): Fix model ID parsing for providers with slashes in model names — use indexOf/substring instead of split to handle models like modelscope/moonshotai/Kimi-K2.5.

• fix(core): Fix reference counting in ModelStatusContext — changed registeredModels from Set to Map<string, number> to prevent polling stop when one component unmounts while others still track the same model.

• fix(security): Prompt injection guard failures now return an explicit 500 response instead of silently passing through (fail-closed policy).

• fix(security): Encryption now derives new keys from a secret-based salt while falling back to the legacy static-salt key during decryption, preserving existing stored credentials.

• fix(combo): Resolve context truncation bug in combo routing to prevent incomplete execution states. (#1517)

• fix(compression): Implement bidirectional tool_pair cleaning for anthropic inputs (fixes #1592).

• fix: Resolve v3.7.0 stabilization issues including dashboard navigation routing, ProxyRegistryManager component layout, and models API response merging (#1566, #1560, #1559).

• fix(cli): Preserve TOML integer/boolean types in Codex config round-trip to prevent tui.model_availability_nux validation errors.

• fix(tailscale): Support sudo auth prompts and live daemon socket detection for non-root tunnel management.

• fix(dashboard): Stabilize usage tab loading and refresh behavior to prevent empty state flashes.

• fix(i18n): Translate 519 untranslated pt-BR keys and add missing Windsurf/Cline/Kimi docs keys.

• fix(i18n): Add missing dashboard message keys across all 30 locales.

• fix(cli): Align OpenCode config preview and add multi-model selection (#1602).

• fix(security): Harden management API auth and OpenAPI try-proxy endpoint.

• fix(security): Resolve vulnerability scan findings for auth-guarded routes.

♻️ Refactoring

• refactor(fallback): Make provider failure thresholds configurable via PROVIDER_PROFILES instead of hardcoded constants, supporting different failure tolerance per provider type. (#1449)
• refactor(resilience): Unify resilience controls across the codebase for consistent circuit breaker and fallback behavior. (#1449)
• refactor(core): Implement shared path utilities, add custom date formatting, improve type safety, and unify database imports across modules.
• refactor(security): Harden backup archive creation by switching to execFileSync, validate ACP agent IDs, expand shared CORS handling.
• refactor(release): Remove obsolete agent workflow playbooks and the stale compiled src/lib/dataPaths.js artifact. (#1541)

🧪 Tests

• test(providers): Add targeted coverage for AWS Polly SigV4 speech/validation, Azure OpenAI deployment discovery, Lemonade local discovery, provider dashboard taxonomy, managed provider catalog behavior, and merged /v1/models alias metadata.
• test(catalog): Add v3.7.0 catalog coverage for Pollinations text models, Perplexity Sonar via Puter, and NVIDIA free-model alias resolution.
• test(vision-bridge): Add 51 unit tests covering all VisionBridge spec scenarios (VB-S01 through VB-S10), including helper functions for callVisionModel, extractImageParts, replaceImageParts, and resolveImageAsDataUri.
• test(batch-api): Isolate batch API unit tests with temp DATA_DIR to prevent schema state collisions.
• test(settings-api): Add test harness with createSettingsApiHarness function for proper temp directory setup and storage reset between tests.
• test(security): Update prompt injection test for fail-closed policy alignment.
• test(core): Restore local test fixes for encryption and resilience modules.
• test(next): Align transpile package expectations for the Next.js standalone build.
• test(ci): Fix CI-only test failures from environment differences — clear INITIAL_PASSWORD and JWT_SECRET in integration tests, handle XDG_CONFIG_HOME for guide-settings tests.

📚 Documentation

• docs: Update the root changelog with all release-branch changes through 2026-04-24, including PRs #1544, #1555, #1551, #1550, #1548, #1547, #1541, #1538, #1536, and #1527.
• docs: Fix broken README and localized documentation links. (#1536)
• docs: Add dashboard docs coverage for current API endpoints, management APIs, ACP, MCP tools, provider onboarding, and v3.7.0 task reconciliation.
• docs: Add Arch Linux AUR install notes for community package support. (#1478)
• docs(i18n): Improve Ukrainian (uk-UA) translation quality — full Ukrainian translation for README, SECURITY, A2A-SERVER, API_REFERENCE, AUTO-COMBO, and USER_GUIDE documents. Fix mixed Latin/Cyrillic typos, translate model table entries, and standardize section headers.

🛠️ Maintenance

• chore: Add .tmp/ to .gitignore to keep local build/test artifacts out of release diffs. (#1538)
• chore(release): Clarify release version parity and changelog segregation rules for generated release workflows.

📦 Dependencies

• deps: Bump the development group with 4 updates. (#1464)
• deps: Bump the production group with 4 updates. (#1463)
• deps: Update @lobehub/icons to 5.5.4, add explicit react-is@19.2.5 for Recharts, pin npm installs to skip unused peer auto-installs, and override Electron's transitive @xmldom/xmldom to 0.9.10 so audit findings stay closed.

---

[3.6.9] — 2026-04-19

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Mark Qwen OAuth provider as deprecated following the upstream free tier shutdown on 2026-04-15. Adds deprecation warning to CLI tool UI and rewrites saveQwenConfig to inject OmniRoute as a multi-provider (openai, anthropic, gemini) via .qwen/settings.json and .qwen/.env (#1437)

• feat(cc-compatible): Align Claude Code-compatible request shape with the official Claude CLI protocol, including proper system skeleton and request normalization (#1411)

• feat(skills): Provider-aware marketplace UX with scored AUTO injection and memory pipeline hardening. Skills now show relevance scores and can automatically inject context into requests (#1411)

• feat(claude-code): Update Claude Code obfuscation to version 2.1.114, centralize hardcoded version strings, and use standard logger (#1403)

• feat(cli-tools): Add direct configuration file generation and override support for Qwen Code local settings (#1394)

• feat(providers): Derive Claude CLI model defaults dynamically from provider registry to stay current with upstream API changes (#1393)

• feat(core): Implement persistent API key, backup pruning, and GPU optimization (#1350, #1367, #1369)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(cli-tools): Prevent masked API keys (sk-31c4****8600) from being written to CLI tool config files. The dashboard UI now passes key.id to the backend, which resolves the unmasked key from the database via a new resolveApiKey() helper. Fixes auth failures across all CLI tools (Claude, Codex, Cline, Kilo, Droid, OpenClaw, Antigravity) (#1435)

• fix(cc-compatible): Trim the default Claude Code-compatible system prompt skeleton from a multi-paragraph instruction set down to a single identifier line, reducing redundant token usage since Claude Code already injects its own extensive system context (#1433)

• fix(security): Resolve SSRF environment static evaluation bug where the outbound URL guard could be bypassed via computed expressions (#1427)

• fix(auth): Reload fresh token state and unify expiry persistence to prevent stale credentials from causing cascading auth failures

• fix(core): Stabilization fixes for token refresh, usage translation, and testing infrastructure

• fix(api): Stop sending unsupported parameters to Gemini and Codex upstream APIs, preventing 400 Bad Request errors

• fix(skills): Optimize AUTO scoring algorithm and include Responses API input context for more accurate skill relevance matching (#1418)

• fix(responses): Preserve reasoning content when translating Chat Completions format to Responses API format, preventing loss of chain-of-thought data (#1414)

• fix(cc-compatible): Add Claude CLI system skeleton for OpenAI-format inputs to ensure consistent behavior when CC-compatible providers receive OpenAI-style payloads

• fix(providers): Add ref to GEMINI_UNSUPPORTED_SCHEMA_KEYS to fix 400 errors from Gemini CLI when tool schemas contain JSON Schema $ref fields

• fix(codex): Prevent proactive token refresh from consuming valid tokens and strip the unsupported background parameter from upstream requests

• fix(providers): Fix usage.prompt_tokens under-reporting when translating Claude caching responses to OpenAI format (#1426)

• fix(core): Fix token refresh resilience for Codex providers. Unrecoverable OAuth refresh errors (token_expired and invalid_token) now correctly mark the connection as invalid to prompt user re-authentication, rather than silently failing (#1415)

• fix(providers): Fix Gemini tool calling by removing the unsupported additionalProperties schema field, resolving 400 errors during complex tool invocations (#1421)

• fix(providers): Remove arbitrary user thought signature injection in Gemini responses to comply with updated API constraints (#1410)

• fix(providers): Fix Gemini API part count mismatch for streaming responses (#1412)

• fix(codex): Respect openaiStoreEnabled setting during native passthrough for Responses API to prevent unsupported upstream arguments (#1432)

• fix(ui): Makes dropdown text visible in dark mode within the Combo Builder modal (#1409)

• fix(chatcore): Apply proactive compression before provider translation to prevent token limit errors in combo routes (#1406)

• fix(claude-code): Scope thinking stripping to executor boundaries to prevent issues with normal API requests (#1401)

• fix(claude-code): Scope obfuscation logic to CLI clients only and fix associated test assertions

• fix(mitm): Resolve MITM not working when connecting Antigravity (#1399)

• fix(security): Resolve CodeQL password hash alert and fix TruffleHog CI failure (#161)

• fix(combo): Fallback to the next model when all provider accounts return a 503 rate-limited signal instead of aborting the routing sequence (#1398)

• fix(codex): Strip server-generated IDs from response items in input to prevent 404 lookup errors in multi-turn Codex Conversations (#1397)

• fix(codex): Optimize Chat Completions paths by converting system to developer roles instead of hoisting them into instructions, enabling prompt caching for system messages on GPT-5 models (#1400)

• fix(providers): Resolve Claude passthrough corruption (#1359), Kimi-k2 reasoning header rejections (#1360), thinking parameter leaks (#1361), and Ollama proxy redirect drops (#1381)

• fix(core): Proxy lookup in key validation respects the new ProxyRegistry environments, and proxy contexts correctly inherit downwards during token refresh preventing expiration loops (#1384, #1390)

• fix(providers): Treat upstream legacy validation HTTP 5xx responses as a valid bypass for Qoder PAT tokens to prevent false negative invalidation (#1391)

• fix(electron): Resolve type error in Header electronAPI properties

• fix(security): Resolve CodeQL security alerts including safe prototype bindings (#151, #152, #154, #155-159)

• fix(tsc): Silence baseUrl deprecation warnings for TypeScript 5.5+ configurations

🧪 Tests

• test(core): Resolve typescript strictness complaints and fix combo-routing-engine test regression
• test(core): Resolve remaining strict type errors across all unit test files
• test(providers): Fix provider service assertion for anthropic-compatible header format
• test(codex): Align codex passthrough assertions with explicit store retention policy
• test(codex): Fix store assertion for codex responses
• test(cli): Resolve strict null checks in Qoder unit tests

🛠️ Maintenance

• chore: Sync infrastructure with docker postinstall components and secondary CodeQL analysis rules
• chore: Enforce contributor credit rule in review-prs workflow
• chore: Fix TS errors and update review-prs workflow for improved automation
• ci: Allow manual CI dispatch for release branches
• ci: Shard long-running test suites and relax timeouts for stability
• ci: Restore release v3.6.9 build pipeline and fix flaky tests
• docs: Update generate-release workflow to use full changelog for PR body
• docs: Enforce PR merge instead of manual close in workflows

---

[3.6.8] — 2026-04-17

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(providers): Support xhigh reasoning tier exclusively on Claude models that expose it (#1356)

• feat(providers): Add CC Compatible connection-level 1M context toggle (#1357)

• feat(core): Add full support for Node.js 24 LTS (Krypton) environments with continuous integration coverage (#1340)

• feat(dashboard): Display Antigravity credit balance in dashboard Limits & Quotas (#1338)

• feat(i18n): Add internationalization support for combo features and dashboard components; sync translations across 31 keys (#1318)

• feat(providers): Add Claude Opus 4.7 to Claude Code OAuth models natively with extended context and caching (#1347)

• feat(core): Add stopSequences support and expand tool definitions to include Google Search capabilities

• feat(auth): Enforce dashboard session authentication on all management API routes, preventing unauthenticated access to configuration endpoints

• feat(runtime): Add hot-reloadable guardrails and model diagnostics for real-time rule evaluation without restarts

• feat(core): Add payload rules, tag-based routing, and scheduled budget systems for fine-grained request governance

• feat(providers): Expose Antigravity preview model aliases and Gemini CLI onboarding flow for first-time setup

• feat(antigravity): Add client model aliases and thoughtSignature bypass modes for Antigravity OAuth connections

• feat(providers): Expand image provider registry with extended model support including SD3.5, FLUX, and DALL-E 3 HD configurations

• feat(combos): Add new routing strategies and full i18n support for agent features section across 31 languages

🔒 Security

• security: Resolve 18 GitHub CodeQL scan alerts including ReDoS, incomplete sanitization, and bad HTML filtering regexp patterns
• fix(auth): Seal privilege escalation vector by enforcing JWT session checking exclusively on /api/keys management endpoints (#1353)
• fix(providers): Resolve Codex token refresh race condition via mutex getAccessToken preventing refresh_token_reused Auth0 revocations

🔧 Maintenance & Architecture

• refactor(core): Split CLI runner and decouple migration engine for extensibility (#1358)
• refactor(audit): Rewire audit dashboard from dead in-memory configAudit store to live SQLite audit_log table — 331+ hidden compliance entries now visible in /dashboard/audit
• build(deps): Bump softprops/action-gh-release from v2 to v3
• ci: Bump GitHub Actions CI node-version to Node.js 24 natively
• fix(types): Resolve TypeScript compilation errors in claudeCodeCompatible.ts (type predicates, cache_control index access) and proxyFetch.ts (signal nullability)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(context): Scale reserved context tokens dynamically using a 15% sliding window for smaller models

• test(core): Replace unit test with integration test for proactive context compression to align with isolated runner rules (#1378)

• fix(services): Pass origin provider to refreshWithRetry to avoid tripping the generic "unknown" circuit breaker (fixes Codex accounts erroneously disabling)

• fix(db): Prevent native module ABI load crashes from assuming database corruption and skipping databases

• fix(db): Increase mass-migration threshold from 5 to 50 pending migrations to protect legacy users upgrading node

• fix(db): Prevent migration runner safety aborts from triggering on fresh DATA_DIR installations by detecting new databases (#1328)

• fix(mcp): Checkpoint and close MCP audit SQLite database safely on process signals and shutdown (#1348)

• fix(mcp): Fully decouple MCP audit SQLite connection caching via globalThis to fix unhandled teardown in standalone Next.js chunks (#1349)

• fix(cli): Avoid creating app router directory during postinstall initialization on non-built source trees (#1351)

• fix(codex): Correctly translate system role to developer in input array to unlock GPT-5 automatic prompt caching (#1346)

• fix(core): Pass client headers to executor in chatCore (#1335)

• fix(providers): Separate test batch calls and ignore unknown connections

• fix(providers): Add grok-web SSO cookie validation handler (#1334)

• fix(db): Preserve key_value settings (dashboard passwords, saved aliases) across DB heuristic recreation cycles (#1333)

• fix(routing): Allow combo fallback to cascade context overflow 400 errors instead of immediate aborts (#1331)

• fix(core): Resolve thinking leaks, consecutive roles, and missing thoughtSignatures for Antigravity translator (#1316)

• fix(translator): Only apply thoughtSignature to the first functionCall part in Gemini parallel tool calls, preventing duplicate signatures

• fix(providers): Default to batch testing execution blocks for web, search, and audio modalities to prevent connection timeouts

• fix(cli): Resolve Node 22 TS entrypoint incompatibility by using esbuild compilation (#1315)

• fix(chat): Preserve max_output_tokens for Responses API targets in chatCore sanitization (#1313)

• fix(api): API Manager usage stats showing 0 for all registered keys (#1310)

• fix(api): Support image-only models in catalog and allow authless search providers to bypass validation requirements

• fix(routes): Require prompts for media generation requests (/images, /videos, /music), returning 400 on missing payloads

• fix(dashboard): Auto-scroll ActivityHeatmap to show current date (#1309)

• fix(dashboard): Restore horizontal layout with w-max wrapper in heatmap components

• fix(i18n): Update nodeIncompatibleHint to recommend Node 24 LTS across all 31 languages

• fix(i18n): Add Chinese i18n support to remaining dashboard components (Loading.tsx, DataTable, etc.)

• fix(requestLogger): Add missing cacheSource and tps columns to i18n log detail views

[3.6.6] — 2026-04-15

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(storage): Add database backup cleanup controls, UI management, and customizable retention period env vars (#1304)

• feat(providers): Add Freepik Pikaso image generation provider with support for cookie/subscription-based auth modes (#1277)

• feat(providers): Add Perplexity Web (Session) Provider — Routes through Perplexity's internal SSE API using a session cookie, giving native proxy access without separate API costs to GPT-5.4, Claude Opus, Gemini 3.1 Pro, and Nemotron via preferences mapping (#1289)

• feat(api): Sync Tokens & V1 WebSocket Bridge — Dedicated sync token storage, issuance, revocation, and bundle download routes backed by stable config bundle versioning with ETag support. Exposes /v1/ws WebSocket upgrade route and a custom Next.js server bridge (scripts/v1-ws-bridge.mjs) so OpenAI-compatible WebSocket traffic can be proxied through the gateway. Compliance auditing expanded with structured metadata, pagination, request context, auth/provider credential events, and SSRF-blocked validation logging. New migrations: 024_create_sync_tokens.sql. New modules: syncTokens.ts, src/lib/sync/bundle.ts, src/lib/sync/tokens.ts, src/lib/ws/handshake.ts, src/lib/apiBridgeServer.ts, src/lib/compliance/providerAudit.ts.

• feat(models): GLM Thinking Preset & Hybrid Token Counting — GLM Thinking (glmt) registered as a first-class provider preset with shared GLM model metadata, pricing, per-connection usage sync, dashboard support, and maxTokens: 65536 / thinkingBudgetTokens: 24576 request defaults with 900s extended timeout. Provider-side /messages/count_tokens endpoint used when a Claude-compatible upstream supports it; gracefully falls back to estimation on missing models, missing credentials, or upstream failures. Startup seeding of default model aliases (src/lib/modelAliasSeed.ts) normalizes common cross-proxy model dialects so canonical slash-based model IDs are not misrouted. New file open-sse/config/glmProvider.ts.

• feat(core): Hardened Outbound Provider Calls & Cooldown Retries — Guarded outbound fetch helpers (src/shared/network/safeOutboundFetch.ts, src/shared/network/outboundUrlGuard.ts) blocking private/local URLs with configurable retry, timeout normalisation, and route-level status propagation for provider validation and model discovery. Cooldown-aware chat retries (src/sse/services/cooldownAwareRetry.ts) with configurable requestRetry and maxRetryIntervalSec settings and model-scoped cooldown responses. Improved rate-limit learning from headers and error bodies so short upstream lockouts can recover automatically. Runtime environment validation (src/lib/env/runtimeEnv.ts) checks env at startup. Pollinations now requires an API key. Antigravity and Codex header handling aligned via open-sse/config/antigravityUpstream.ts and open-sse/config/codexClient.ts. Gemini tool names restored in translated responses; synthetic Claude text block injected when upstream SSE completes empty.

• feat(logs): Add TPS (Tokens Per Second) metric to log details modal metadata grid (#1182)

• feat(memory+skills): Full-featured Memory & Skills systems with FTS5 SQLite search, dynamic UI pagination, backend observability, and extensive test coverage (#1228)

• feat(bailian-quota): Add Alibaba Coding Plan quota monitoring, multi-window quota extraction, and UI credential validation (#1235)

• feat(storage): Call Log Storage Refactor — Extracted heavy request/response JSON payloads from the core SQLite database (storage.sqlite) into filesystem artifacts stored within DATA_DIR/call_logs. This massively reduces WAL bloat and eliminates SQLITE_FULL crashes on high-traffic nodes (#1307).

• feat(providers): Add Grok Web (Subscription) Provider — Routes through the xAI web interface for subscription users via cookie session mapping (#1295).

• feat(api): Advanced Media Support — Extends OpenAI generic proxy layer to natively support image, embeddings, audio-transcriptions, and audio-speech workflows (#1297).

• feat(cli-tools): Qwen Code CLI Integration — Full integration for Qwen Code local execution mapping, model resolution, and dynamic API key fetching (#1266, #1263).

• feat(oauth): Supports cursor-agent CLI as a native Cursor credential source alongside the standard configuration (#1258).

• feat(models): Custom and imported models now merge correctly into filter lists for all available global providers (#1191).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(providers): match correct endpoint api.xiaomimimo.com for Xiaomi MiMo (#1303)

• fix(core): strip provider alias routing prefix from payload for custom endpoints to fix Azure OpenAI 400 errors (#1261)

• fix(core): ProxyFetch Undici dispatcher automatically bypasses LAN/local addresses, preventing fetch failures on internal OpenRouter requests (#1254)

• fix(core): Gemini thought stream signature detection upgraded to use native part.thought boolean, preventing reasoning text leaks (#1298)

• deps: bump hono from 4.12.12 to 4.12.14 to resolve CVE SSR HTML injection vulnerability (#1306, #59)

• deps: update dompurify to 3.4.0 in frontend overrides mitigating XSS HTML Injection (CVE-XYZ / Dependabot #60)

• test: Disable SQLite automatic backups during continuous integration (CI) tests to resolve E2E timeout issues limiting runner scaling (#24481475058)

• feat(core): Proactive Context Compression — chatCore now proactively compresses oversized message contexts before hitting upstream providers to dramatically reduce context_length_exceeded errors. Employs binary-search message pruning with structural integrity guarantees tracking explicit tool_use boundaries ensuring truncated tool inputs drop paired outputs appropriately (#1292, #1293)

• fix(cli): Resolve codex routing config parsing by strictly quoting section keys array, enforcing responses wire_api with fallback, and standardizing select-model button positioning mirroring Claude UI

• fix(providers): Correct Lobehub provider icons rendering by removing unsupported local references ensuring local SVG/PNG fallback mechanism invokes natively

• fix(db): Implement Database migration tracking safety abort safeguards (pre-migration backups via VACUUM INTO and mass renumbering warnings) to protect existing database structures on startup upgrades (#1281)

• fix(dashboard): Cleaned up target codex config.toml structure preventing recursive section rendering by enforcing quotes on section dot paths and mapping correct UI OMNIROUTE_API_KEY names.

• fix(mcp): Add dedicated explicit timeout constraint overrides for search handlers (#1280)

• fix(crypto): Add validation guard to encryption layer to surface clear UI errors when cryptographic environment variables are missing, replacing raw Node.js TypeErrors. Legacy env vars OMNIROUTE_CRYPT_KEY and OMNIROUTE_API_KEY_BASE64 now also accepted as fallbacks (#1165)

• fix(providers): Update Pollinations provider definition to require API keys and specify their new limited pollen/hour free tier (#1177)

• Streaming \n\n Artifact Fix (#1211): Changed <omniModel> tag-stripping regex from ? to * quantifier across combo.ts, comboAgentMiddleware.ts, and contextHandoff.ts to greedily strip all accumulated JSON-escaped newline sequences surrounding the tag. This prevents literal \n\n prefix artifacts from appearing in consumer streaming responses

• E2E Combo Test Locator: Fixed Playwright strict-mode violation in combo-unification.spec.ts by replacing ambiguous getByRole locator with a compound filter locator for the "All" strategy tab

• fix(cc-compatible): Trim beta flags and preserve cache passthrough for third-party HTTP proxy compatibility (#1230)

• fix(providers): Update Xiaomi MiMo endpoints to the live token-plan, migrating away from dead API URLs (#1238)

• fix: Forward client x-initiator header to GitHub Copilot upstream to accurately distinguish agent vs user turns (#1227)

• fix: Resolve backlog bugs including streaming edge cases, unhandled rejections, and quota parse failures (#1206, #1220, #1231, #1175, #1187, #1218, #1202)

• fix(tests): Resolve memory migration and skills route pagination bugs arising from PR overlaps

• fix(i18n): Add missing Chinese i18n support to dashboard components (DataTable, EmptyState, etc), update en.json/zh-CN.json routing keys, and natively resolve JSX defaults via next-intl (#1274)

🔧 Internal Improvements

• Compliance Audit Expansion: src/lib/compliance/index.ts expanded with structured metadata, pagination support, request context enrichment, and new providerAudit.ts module logging auth and provider credential events, SSRF-blocked validation attempts, and provider CRUD operations
• Config Sync Bundle: src/lib/sync/bundle.ts exports buildConfigBundle() generating a versioned JSON snapshot of settings, provider connections, nodes, model aliases, combos, and API keys (passwords redacted) with ETag support for bandwidth-efficient polling
• Codex Client Constants: Centralized CODEX_CLIENT_VERSION, CODEX_USER_AGENT_PLATFORM, and pattern-validated env overrides (CODEX_CLIENT_VERSION, CODEX_USER_AGENT) in open-sse/config/codexClient.ts
• Antigravity Upstream Constants: open-sse/config/antigravityUpstream.ts consolidates all Antigravity base URLs and model/fetchAvailableModels discovery path builders
• Model Alias Seed: src/lib/modelAliasSeed.ts seeds 30+ cross-proxy model dialect aliases (e.g. openai/gpt-5 → gpt-5, anthropic/claude-opus-4-6 → cc/claude-opus-4-6) at startup via idempotent upsert
• Test Coverage: 15+ new unit test suites covering sync routes, WebSocket bridge, compliance index, GLM provider config, cooldown-aware retry, safe outbound fetch, stream utilities, Codex executor, provider validation branches, model cross-proxy compatibility, and model alias seeding
• TypeScript Migration: Finalized migration of remaining JS tests (proxy-load and testFromFile) to TypeScript ES modules, ensuring a fully synchronized TS stack.
• Reliability & Resilience: Added exponential backoff to models.dev auto-sync to combat transient network failures, raised interval floor to 1 hour, and added LKGP debug logging for enhanced observability during routing. (#1286)

---

[3.6.5] — 2026-04-13

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Antigravity AI Credits Fallback: Automatically retries with GOOGLE_ONE_AI credit injection when free-tier quota is exhausted. Per-account credit balance (5-hour TTL) is cached from SSE remainingCredits and exposed as a numeric badge in the Provider Usage dashboard (#1190 — thanks @sFaxsy)

• Claude Code Native Parity: Full header/body signing parity with the Claude Code 2.1.87 OAuth client — CCH xxHash64 body signing with singleton WASM initialization promise (fixing race conditions), dynamic per-request fingerprint, bidirectional TitleCase ↔ lowercase tool name remapping (14 tools), API constraint enforcement (temperature=1 for thinking, max 4 cache_control blocks, auto-inject ephemeral on last user message), and optional ZWJ obfuscation. Wired into BaseExecutor for automatic CCH signing on all anthropic-compatible-cc-* providers and into chatCore for synchronous parity pipeline steps (#1188 — thanks @RaviTharuma)

• Per-Connection Codex Defaults: Codex Fast Service Tier and Reasoning Effort settings are now per-connection instead of a single global toggle. Existing connections are migrated automatically on startup via an idempotent backfill migration (#1176 — thanks @rdself)

• Cursor Usage Dashboard: New getCursorUsage() fetches quotas from Cursor's /api/usage, /api/auth/me, and /api/subscription endpoints. Displays standard requests, on-demand usage, and per-plan limits (Free/Pro/Business/Team). Client version bumped to 3.1.0 and x-cursor-user-agent header added for parity

• Database Health Check System: Automated periodic SQLite integrity monitoring via runDbHealthCheck() — detects orphan quota/domain rows, broken combo references, stale snapshots, and invalid JSON state. Runs every 6 hours (configurable via OMNIROUTE_DB_HEALTHCHECK_INTERVAL_MS), with auto-repair and pre-repair backup. Exposed as MCP tool #18 (omniroute_db_health_check) with Zod schemas and autoRepair option. Dashboard panel in Health page with status card, issue count, repaired count, and one-click repair button

• OpenAI Responses API Store Opt-In: Per-connection openaiStoreEnabled flag controls whether the store field is preserved or forced to false on Codex Responses API requests. When enabled, previous_response_id, prompt_cache_key, session_id, and conversation_id fields are round-tripped through the Chat Completions → Responses translation, enabling multi-turn context caching on supported providers

• Email Privacy Toggle (Combos Page): Global email visibility toggle (EmailPrivacyToggle) added to the Combos page header with responsive layout, tooltip guidance, and per-connection label masking via pickDisplayValue(). All combo builder options, provider connection lists, and quota screens now respect the global privacy state from emailPrivacyStore

• skills.sh Integration: Added skills.sh as an external skill provider. Users can now search, browse, and install agent skills directly from a new "skills.sh" tab in the Skills dashboard. Includes backend API resolvers, frontend implementation with search/install states, and a dedicated unit test suite (#1223 — thanks @RaviTharuma)

• Stabilization Settings: Added persistence support for lkgpEnabled and backgroundDegradation settings, integrated into instrumentation-node.ts for improved lifecycle awareness (#1212)

• xxhash-wasm dependency: Added xxhash-wasm@^1.1.0 for CCH signing (xxHash64 with seed 0x6E52736AC806831E)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Codex stream: false via Combo (ALL_ACCOUNTS_INACTIVE): Fixed a critical bug where Codex combos returned ALL_ACCOUNTS_INACTIVE or empty content when the client sent stream: false. Root cause was triple: (1) CodexExecutor.transformRequest() mutated body.stream in-place to true, contaminating the combo's quality check which skipped validation thinking it was streaming; (2) the non-stream SSE parser used the wrong format (Chat Completions instead of Responses API) for Codex SSE output; (3) combo quality validation read the mutated body.stream instead of the client's original intent. Fixed by: cloning the body via structuredClone() in CodexExecutor, detecting Codex/Responses SSE format in the non-stream fallback path (with auto-translation back to Chat Completions), and capturing clientRequestedStream before the combo loop

• Gemini CLI Tool Schema Rejection: Fixed 400 Bad Request errors from the Google API by strictly filtering non-standard vendor extensions (starting with x-) and deprecated fields from tool parameter schemas (#1206)

• SOCKS5 Proxy Interop (Node.js 22): Resolved invalid onRequestStart method crashes caused by undici version mismatches between dispatchers and the built-in fetch. Hardened proxyFetch.ts to strictly use the library's fetch implementation for custom dispatchers (#1219)

• Search Cache Coalescing with TTL=0: Fixed a bug where providers configured with cacheTTLMs: 0 (caching explicitly disabled) still had concurrent requests coalesced and returned { cached: true }. Now each call gets its own independent upstream fetch (#1178 — thanks @sjhddh)

• Antigravity Credit Cache Alignment (PR #1190): Reconciled accountId derivation between AntigravityExecutor.collectStreamToResponse and getAntigravityUsage to use consistent cache keys (email || sub || "unknown"). Previously, SSE-parsed credit balances could be written under a different key than the one read by the usage dashboard, causing stale/missing credit badges

• Non-streaming reasoning_content Duplication: Fixed clients rendering duplicated reasoning panels when both reasoning_content and visible content were present in non-streaming responses. responseSanitizer now strips reasoning_content from messages that already have visible text content, preserving it only for reasoning-only messages

• Streaming Regression Fix: Hardened the sanitize TransformStream in the combo engine to strip both literal and JSON-escaped newline sequences, eliminating leading \n\n prefixes in assistant responses (#1211)

• Gemini Empty Choice Fix: Ensured initial assistant deltas always include an empty content: "" string to satisfy strict OpenAI client requirements and prevent empty choice responses in tools (#1209)

• Gemini Tools Sanitizer Deduplication: Extracted shared tool conversion logic into buildGeminiTools() helper (geminiToolsSanitizer.ts), eliminating duplicate implementations between openai-to-gemini.ts and claude-to-gemini.ts. The new helper correctly handles web_search / web_search_preview tool types by emitting googleSearch tools with priority over function declarations

• Qwen/Qoder Thinking+Tool_Choice Conflict: Added sanitizeQwenThinkingToolChoice() to both DefaultExecutor (for Qwen provider) and QoderExecutor to prevent provider-side 400 errors when clients send tool_choice alongside thinking/reasoning parameters that are mutually exclusive upstream

• API Key Deletion Orphan Cleanup: Deleting an API key now also removes associated domain_budgets and domain_cost_history rows, preventing orphan data accumulation

• CC-compatible test assertion: Fixed pre-existing test that expected no cache_control on system blocks — the billing header system block now carries cache_control: { type: "ephemeral" } per PR #1188 design

• Codex Combo Smoke Test False Positives: Fixed combo tests incorrectly reporting ERROR for valid Codex streaming responses when response.output is empty but text deltas were emitted. The summary now falls back to accumulated delta text (#1176 — thanks @rdself)

• Electron Builder Version Mismatch: Fixed Electron desktop startup failures on Windows packaged builds caused by native modules (better-sqlite3) being under app.asar.unpacked while helpers were in app/node_modules. resolveServerNodePath() now merges both locations with deduplication and existence checks (#1172 — thanks @backryun)

🔧 Internal Improvements

• SSE Parser: Responses API Non-Stream Conversion: Added full parseSSEToResponsesOutput() implementation in sseParser.ts (255+ lines) — reconstructs complete Responses API objects from SSE event streams, handling response.output_text.delta/done, response.reasoning_summary_text.delta/done, response.function_call_arguments.delta/done, and terminal events. Used by the new chatCore non-stream fallback path for Codex
• Cursor Executor Version Sync: Updated Cursor client User-Agent to 3.1.0 and centralized version constants (CURSOR_CLIENT_VERSION, CURSOR_USER_AGENT) for consistent fingerprinting across executor, usage fetcher, and OAuth flows
• Responses API Translator Parity: convertResponsesApiFormat() now accepts credentials and passes them through to the translator, enabling store-aware field propagation. Round-trip preservation of previous_response_id, prompt_cache_key, session_id, and conversation_id fields
• Provider Schema Validation: Added openaiStoreEnabled boolean validation to providerSpecificData Zod schema
• Combo Error Response Normalization: Empty combo targets now return 404 (comboModelNotFoundResponse) instead of generic 503, improving client-side error differentiation
• Dependency Updates: Bumps typescript-eslint to 8.58.2 (dev), axios to 1.15.0 (prod), and next to 16.2.2 (prod) (#1224, #1225)

⚠️ Breaking Changes

• DELETE /api/settings/codex-service-tier removed: This endpoint no longer exists. Codex Service Tier configuration has moved to per-connection providerSpecificData.requestDefaults. Existing connections are migrated automatically on first startup after upgrade. Any external scripts or integrations that call this endpoint should be updated — use PUT /api/providers/:id with providerSpecificData.requestDefaults.serviceTier instead (#1176).
• CCH signing on CC-compatible providers: All requests to anthropic-compatible-cc-* providers now include an xxHash64 integrity token (cch=...) in the billing header. Providers that do not validate CCH will ignore it (no behavioral change), but any custom middleware inspecting the billing header should expect a 5-character hex token instead of the 00000 placeholder

---

[3.6.4] — 2026-04-12

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Combo Builder v2 (Wizard UI): Completely redesigned the combo creation/editing interface as a multi-stage wizard with stages: Basics → Steps → Strategy → Review. The builder fetches provider, model, and connection metadata via a new GET /api/combos/builder/options endpoint, enabling precise provider/model/account selection with duplicate detection and automatic next-connection suggestion. Heavy UI components (ModelSelectModal, ProxyConfigModal, ModelRoutingSection) are now lazily loaded via next/dynamic for faster initial page render

• Combo Step Architecture (Schema v2): Introduced a structured step model (ComboModelStep, ComboRefStep) replacing the legacy flat string/object combo entries. Steps carry explicit id, kind, providerId, connectionId, weight, and label fields, enabling pinned-account routing, cross-combo references, and per-step metrics. All combo CRUD operations normalize entries through the new src/lib/combos/steps.ts module. Zod schemas updated with comboModelStepInputSchema and comboRefStepInputSchema unions

• Composite Tiers System: Added tiered model routing via config.compositeTiers — each tier maps a named stage to a specific combo step with optional fallback chains. Includes comprehensive validation (src/lib/combos/compositeTiers.ts) ensuring step existence, preventing circular fallback, and validating default tier references. Zod schema enforcement blocks composite tiers on global defaults (concrete combos only)

• Model Capabilities Registry: Created src/lib/modelCapabilities.ts providing getResolvedModelCapabilities() — a unified resolver that merges static specs, provider registry data, and live-synced capabilities into a single ResolvedModelCapabilities object covering tool calling, reasoning, vision, context window, thinking budget, modalities, and model lifecycle metadata

• Observability Module: Extracted health and telemetry payload construction into src/lib/monitoring/observability.ts with buildHealthPayload(), buildTelemetryPayload(), and buildSessionsSummary() builders. The health endpoint now returns session activity, quota monitor status, and per-provider breakdowns alongside existing system metrics

• Session & Quota Monitor Dashboard: Added live Session Activity and Quota Monitors panels to the Health dashboard, showing active session counts, sticky-bound sessions, per-API-key breakdowns, and top session details alongside quota monitor alerting/exhausted/error status with per-provider drill-down

• Combo Health Per-Target Analytics: The combo-health API now resolves per-target metrics using the new resolveNestedComboTargets() function, providing step-level success rates, latency, and historical usage breakdowns per execution key — enabling per-account, per-connection health visibility

• Auto-Combo → Combos Unification: Merged the separate /dashboard/auto-combo page into the main /dashboard/combos page. Auto/LKGP combos are now managed alongside all other combos with a new strategy filter tabs system (All / Intelligent / Deterministic). The old auto-combo route redirects to /dashboard/combos?filter=intelligent. Removed the auto-combo sidebar entry, consolidating navigation into the single Combos item

• Intelligent Routing Panel (IntelligentComboPanel): New inline panel (371 lines) within the combos page that shows real-time provider scores, 6-factor scoring breakdown (quota, health, cost, latency, task fitness, stability), mode pack selector, incident mode status, and excluded providers for auto/lkgp combos — replacing the former standalone auto-combo dashboard

• Builder Intelligent Step (BuilderIntelligentStep): New conditional wizard step (280 lines) that appears in the Builder v2 flow only when strategy=auto or strategy=lkgp is selected. Exposes candidate pool selection, mode pack presets, router sub-strategy selector, exploration rate slider, budget cap, and collapsible advanced scoring weights configuration

• Intelligent Routing Module (intelligentRouting.ts): Extracted strategy categorization and filtering logic into a dedicated shared module (210 lines) with getStrategyCategory(), isIntelligentStrategy(), filterCombosByStrategyCategory(), normalizeIntelligentRoutingFilter(), and normalizeIntelligentRoutingConfig() utility functions

• LKGP Standalone Strategy: Implemented lkgp (Last Known Good Provider) as a fully functional standalone combo strategy. Previously, lkgp as a combo strategy silently fell through to priority ordering — the LKGP lookup only ran inside the auto engine. Now strategy: "lkgp" correctly queries the LKGP state, moves the last successful provider to the top of the target list, and saves the LKGP state after each successful request. Falls back to priority ordering when no LKGP state exists

• Unified Routing Rules & Model Aliases: Consolidated the routing rules and model alias management controls into the Settings page, reducing fragmentation across the dashboard

⚡ Performance

• Middleware Lazy Loading: Refactored src/proxy.ts to lazy-import apiAuth, db/settings, and modelSyncScheduler modules, reducing middleware cold-start overhead. Added inline isPublicApiRoute() to avoid loading the full auth module for public routes
• E2E Auth Bypass: Added NEXT_PUBLIC_OMNIROUTE_E2E_MODE environment flag to bypass authentication gates for dashboard and management API routes during Playwright E2E test runs

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• P2C Credential Selection: Implemented Power-of-Two-Choices (P2C) connection scoring in src/sse/services/auth.ts with quota headroom awareness, error/recency penalties, and forced/excluded connection support. The new getProviderCredentialsWithQuotaPreflight() function integrates quota preflight checks directly into credential selection, eliminating the separate Codex-only preflight path

• Fixed-Account Combo Steps: Combo steps with explicit connectionId now correctly bypass provider-level model cooldowns and circuit breakers, preventing a single account failure from blocking pinned-connection routing for the same model

• Combo Metrics Per-Target Tracking: Extended comboMetrics.ts to track byTarget metrics keyed by execution path, recording per-step provider, providerId, connectionId, and label alongside existing per-model aggregates

• Call Logs Schema Expansion: Added requested_model, request_type, tokens_cache_read, tokens_cache_creation, tokens_reasoning, combo_step_id, and combo_execution_key columns to call_logs with auto-migration. Added composite index idx_cl_combo_target for efficient per-target historical queries

• Quota Monitor Enrichment: Expanded quotaMonitor.ts with full lifecycle state tracking (status, startedAt, lastPolledAt, consecutiveFailures, totalPolls, totalAlerts), ISO-formatted snapshots via getQuotaMonitorSnapshots(), and sorted summary via getQuotaMonitorSummary()

• Codex Quota Fetcher Hardening: Improved codexQuotaFetcher.ts with safer connection registration and quota fetch error handling

• LKGP Save Refactored to Async/Await: Replaced fire-and-forget .then() chain for LKGP persistence after successful combo routing with proper async/await + try/catch, preventing unhandled promise rejections and ensuring LKGP state is reliably saved before the response is returned

• Duplicate auto in Combo Strategy Schema: Removed duplicate "auto" entry from comboStrategySchema (was listed on both line 104 and 108). Harmless to Zod runtime but cleaned up to avoid confusion. Schema now has exactly 13 unique strategy values

• Legacy Combo Refs Normalization: Fixed combo step normalization to preserve legacy string combo references during CRUD operations, preventing data loss when editing combos created before the v2 step architecture

🔒 Security

• Auth Bypass on Backup Routes (Critical): Added isAuthenticated guards to /api/db-backups/exportAll (full database export) and /api/db-backups (list, create, and restore backups) — both were previously accessible without authentication
• Auth Guard on Translator Save: Added isAuthenticated guard to /api/translator/save for defense-in-depth consistency
• API Key Secret Hardening: Removed the hardcoded "omniroute-default-insecure-api-key-secret" fallback from apiKey.ts — the function now fails fast if API_KEY_SECRET is unset, relying on the startup validator to auto-generate it
• NPM Tarball Leak Fix: Added app/.env* to .npmignore to prevent the working .env file from being shipped inside the npm tarball distribution
• Electron Builder CVE Fix: Bumped electron-builder to 26.8.1 to resolve tar CVEs in the desktop build pipeline

🔧 Maintenance & Infrastructure

• DB Migration 021: Added combo_call_log_targets migration for combo_step_id and combo_execution_key columns in call_logs
• Combo CRUD Normalization: db/combos.ts now normalizes all stored combo entries through the step normalization pipeline on read, ensuring consistent step IDs and kind annotations regardless of when the combo was created
• Playwright Config: Updated Playwright configuration and run-next-playwright.mjs script for improved E2E test orchestration
• Build Script: Updated build-next-isolated.mjs with additional reliability improvements
• Auto-Combo UI Cleanup: Deleted AutoComboModal.tsx (161 lines), replaced auto-combo/page.tsx (478→5 lines) with a server-side redirect to /dashboard/combos?filter=intelligent
• Sidebar Consolidation: Removed "auto-combo" from HIDEABLE_SIDEBAR_ITEM_IDS and PRIMARY_SIDEBAR_ITEMS — normalizeHiddenSidebarItems() silently discards any stale "auto-combo" entries in user settings
• Schema Cleanup: Removed obsolete createAutoComboSchema from schemas.ts. Exported comboStrategySchema for direct use in test and filter modules
• A2A Agent Card Update: Renamed skill ID from auto-combo to intelligent-routing with updated description referencing the unified combos dashboard
• Builder Draft Refactor: Extended builderDraft.ts with dynamic stage list generation via getComboBuilderStages() and isIntelligentBuilderStrategy(). Stage navigation (getNextComboBuilderStage, getPreviousComboBuilderStage, canAccessComboBuilderStage) now accepts options to conditionally include/skip the intelligent wizard step
• i18n Consolidation: Removed the standalone "autoCombo" i18n block (22 keys) from all 30 language files. Migrated keys into the "combos" block with new additions for filter tabs, intelligent panel, and builder step labels

🧪 Tests

• 16 New Test Suites: Added comprehensive test coverage including:
  • combo-builder-draft.test.mjs (186 lines) — Builder draft step construction and validation
  • combo-builder-options-route.test.mjs (228 lines) — Builder options API endpoint
  • combo-health-route.test.mjs (266 lines) — Combo health analytics with per-target metrics
  • combo-routes-composite-tiers.test.mjs (157 lines) — Composite tiers API integration
  • composite-tiers-validation.test.mjs (131 lines) — Composite tier validation rules
  • db-combos-crud.test.mjs — Combo CRUD with step normalization
  • db-core-init.test.mjs (129 lines) — DB initialization and column migrations
  • model-capabilities-registry.test.mjs (105 lines) — Model capabilities resolution
  • observability-payloads.test.mjs (165 lines) — Health/telemetry payload construction
  • openapi-spec-route.test.mjs — OpenAPI spec generation
  • proxy-e2e-mode.test.mjs (74 lines) — E2E mode auth bypass
  • quota-monitor.test.mjs — Quota monitor lifecycle state
  • run-next-playwright.test.mjs (119 lines) — Playwright runner script
  • sse-auth.test.mjs (154 lines) — P2C credential selection and quota preflight
  • telemetry-summary-route.test.mjs (35 lines) — Telemetry summary endpoint
  • Plus updates to 12 existing test files for compatibility with new step architecture
• Auto-Combo Unification Tests:
  • autocombo-unification.test.mjs (156 lines) — Strategy categorization, schema deduplication, sidebar cleanup, and routing strategies metadata validation
  • combo-unification.spec.ts (189 lines) — Playwright E2E tests for filter tabs, intelligent panel rendering, redirect from old route, sidebar entry removal, and Builder v2 intelligent step flow
  • 3 new LKGP standalone tests in combo-routing-engine.test.mjs — Validates LKGP provider prioritization, fallback to priority when no state exists, and LKGP state persistence after successful requests
  • Updated combo-builder-draft.test.mjs with intelligent stage navigation tests
  • Updated sidebar-visibility.test.mjs to reflect auto-combo removal

---

[3.6.3] — 2026-04-11

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• OpenAI-Compatible Loose Validation: Empty API keys can now be naturally submitted and saved for any openai-compatible-* providers (e.g. Pollinations, localized routes) directly in the UI instead of blocking save actions (#1152)

• Cloudflare Configuration: Updated the provider schema and UI integration for Cloudflare AI to officially expose and support the backend accountId field securely without overrides (#1150)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Vertex JSON Validation Crash: Prevented invalid character in header crashes inside the /validate endpoint by creating a native authentication parser that correctly handles Google Identity Service Account JSON flows prior to pinging endpoints (#1153)

• Extraneous Payload Rejection: Globally prevented upstream 400 Bad Request execution crashes by stripping the non-standard prompt_cache_retention attribute forcibly attached by Cursor/Cline IDE engines when targeting strict OpenAI/Anthropic routes (#1154)

• Reasoning Content Drop: Prevented pure reasoning packets, common in advanced fallback models like DeepSeek, from being aborted mid-stream by explicitly adjusting the Empty Content (502) circuit breakers to acknowledge reasoning_content states as valid (#1155)

• Desktop Windows Build Crash: Fixed better_sqlite3.node is not a valid Win32 application preventing OmniRoute Desktop from launching on Windows by properly removing the ABI-mismatched sqlite cache from Next.js standalone and falling back to the cross-compiled Electron equivalent during packager build steps (#1163)

• Login Visual Security: Removed the raw fallback hash dump that artificially rendered underneath the login modal in Docker instances missing OMNIROUTE_API_KEY_BASE64 flags (#1148)

🔧 Maintenance & Dependencies

• Dependabot Updates: Safely bumped GitHub Actions docker/build-push-action to v7 and actions/download-artifact to v8
• Electron Updates: Upgraded desktop wrapper core to Electron 41.2.0 and electron-builder to 26.8.1, incorporating essential V8/Chromium security patches
• NPM Package Groups: Updated production and development NPM groups to securely handle minor audit warnings and keep toolchains modern
• CI/CD Reliability: Fixed persistent Snyk token-absence failures on automated pull requests by appropriately bypassing on dependabot actions

[3.6.2] — 2026-04-11

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• 33 New API Key Providers: Massive provider expansion adding DeepInfra, Vercel AI Gateway, Lambda AI, SambaNova, nScale, OVHcloud AI, Baseten, PublicAI, Moonshot AI, Meta Llama API, v0 (Vercel), Morph, Featherless AI, FriendliAI, LlamaGate, Galadriel, Weights & Biases Inference, Volcengine, AI21 Labs, Venice.ai, Codestral, Upstage, Maritalk, Xiaomi MiMo, Inference.net, NanoGPT, Predibase, Bytez, Heroku AI, Databricks, Snowflake Cortex, and GigaChat (Sber). OmniRoute now supports 100+ providers (4 Free + 8 OAuth + 91 API Key + Custom compatible)

• Global Email Privacy Toggle: Added a persistent eye-icon toggle button across all dashboard pages (Providers, Usage Limits, Playground) that reveals or hides masked email addresses. Toggle state is stored in localStorage and synced globally via Zustand store

• Documentation Refresh: Updated README, ARCHITECTURE, FEATURES, AGENTS.md, and API_REFERENCE for v3.6.2 with accurate provider counts (100+), new executor list, and system API documentation

• Uninstall Guide: Created comprehensive docs/guides/UNINSTALL.md covering clean uninstallation for all deployment methods (npm, Docker, Electron, source)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• PDF Attachments: Unlocked deep string object parsing (geminiHelper) ensuring Gemini translation successfully passes complex PDF payloads from OpenAI-compatible streams without dropping them silently (#993)

• SkillsMP Engine: Corrected object extraction path mappings inside the API router to fix UI marketplace rendering under Docker/Standalone Node isolated deployments (#988)

---

[3.6.1] — 2026-04-10

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• OAuth Env Repair Action: Added a "Repair env" button to the OAuth Providers dashboard that detects and restores missing OAuth client IDs from .env.example — with timestamped backup and append-only safety. Includes full 33-language i18n support and sanitized API responses (#1116, by @yart)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• i18n: Missing Provider Keys: Added missing filterModels, modelsActive, showModel, hideModel keys across all 32 locale files, fixing runtime MISSING_MESSAGE errors in the providers UI. Also cleaned up duplicate keys in en.json (#1111, by @rilham97)

• GPT-5.4 Routing: Added missing targetFormat: "openai-responses" to gpt-5.4 and gpt-5.4-mini models in both the Codex and GitHub Copilot providers, fixing [400]: model not accessible via /chat/completions errors (#1114, by @ask33r)

---

[3.6.0] — 2026-04-10

✨ New Features & Analytics

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Combo Smoke Test: Raised the default token budget to 2048 to prevent truncation of thinking models during preflight checks, and fully randomized the arithmetic probe prompt to bypass deterministic caching from upstream relays (#1105)

🐛 Bug Fixes & Compliance

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• DB Bloat / Row Limits: Added CALL_LOGS_TABLE_MAX_ROWS and PROXY_LOGS_TABLE_MAX_ROWS (default: 100,000) to the backend DB compliance cleaner to prevent runaway SQLite growth. Limits are enforced automatically on the TTL cycle (#1104, fixes #1101)

• HTML Error Handling: The router now correctly identifies unexpected HTML responses (e.g. <!DOCTYPE html>) sent by upstream providers (like Azure/Copilot) instead of throwing obscure Unexpected token '<' JSON parse errors, bubbling up a clean 502 Bad Gateway (#1104, fixes #1066)

• Android/Termux SQLite Native Support: better-sqlite3 is now correctly built from source with cross-compilation flags in ARM64 local Termux deployments without failing on missing prebuilt binaries (#1107)

---

[3.5.9] — 2026-04-09

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Persistent Combo Ordering: Drag combo cards by handle to reorder them in the dashboard; order is persisted to SQLite via a new sort_order column and POST /api/combos/reorder endpoint. Includes DB migration 020_combo_sort_order.sql and JSON import preservation (#1095)

• Sidebar Group Reorder: Moved "Logs" before "Health" in the System section and "Limits & Quotas" after "Cache" in the Primary section for a more logical navigation flow (#1095)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Stream Failure Surfacing: Upstream response.failed events (e.g. Codex rate-limit errors) are now properly surfaced as non-200 errors instead of being silently swallowed as empty 200 OK streams. Rate-limit failures return HTTP 429 (#1098, closes #1093)

• Upstream Model Preservation: The Responses-to-OpenAI stream translator now preserves the actual upstream model (e.g. gpt-5.4) instead of hardcoding a gpt-4 fallback (#1098, closes #1094)

• Docker EXDEV Fix: build-next-isolated.mjs now falls back from fs.rename() to cp/rm when Docker buildx raises EXDEV (cross-device link), unblocking the Docker image publish workflow (#1097)

• macOS CLI Path Resolution: cliRuntime.ts resolves symlink parents with fs.realpath() to handle macOS /var → /private/var chains, preventing false symlink_escape rejections (#1097)

• Request Log Token Layout: Split token badges into separate Input (Total In, Cache Read, Cache Write) and Output (Total Out, Reasoning) groups for clearer readability; renamed "Time" label to "Completed Time" (#1096)

---

[3.5.8] — 2026-04-09

✨ New Features & Analytics

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Analytics Layout Redesign: Replaced flat metrics with a responsive CompactStatGrid, grouping data visually across sections (#1089)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Build Core: Force Turbopack cleanup via Prepbulish script to prevent Next.js 16 app/ routing conflicts on runtime.

• Provider Quarantine: Introduces model/provider circuit-breakers with adaptive TTL exponential backoff for recurring upstream errors (#1090)

• Oauth Keep-Alive: Safely protects authenticated active accounts against spontaneous dropping from router due to transient token refresh failures (#1085)

🔒 Security & Maintenance

• Dependabot: bumped axios from 1.14.0 to 1.15.0 addressing SSRF flags (#1088)

---

[3.5.7] — 2026-04-09

🐛 Bug Fixes & Security

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Turbopack Standalone Chunks: Fixed a critical bug in scripts/prepublish.mjs where Turbopack chunks missing from the .next/standalone trace resulted in a 500 ChunkLoadError (e.g., _not-found page crash) during production deployments via NPM or Docker. Standalone chunks are now explicitly copied and correctly stripped of Turbopack hashes.

---

[3.5.6] — 2026-04-09

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Email Privacy Masking: OAuth account emails are now masked in the provider dashboard (e.g. di*****@g****.com) to prevent accidental exposure when sharing screenshots. Full address visible on hover via title attribute (#1025).

• OpenRouter & GitHub in Embedding/Image Registries: OpenRouter (3 embedding models, 4 image models) and GitHub Models (2 embedding models via Azure inference) are now first-class entries in the provider registries, enabling their use for /v1/embeddings and /v1/images/generations (#960).

• Model Visibility Toggle & Search Filter: The provider page model list now includes a real-time search/filter bar and a per-model visibility toggle (👁 icon). Hidden models are grayed out and excluded from the /v1/models catalog. An active-count badge (N/M active) shows at a glance how many models are enabled (#750).

• Chinese Localization (zh-CN): Added missing translations for Context Relay, Memory, LKGP, and Models.dev sync features, while standardizing terminology across the application (#1079).

• Environment Auto-Sync: Added sync-env.mjs to auto-generate and append .env from .env.example during installation, automatically generating cryptographic secrets on first run.

• Source Mode Dashboard Update: Fixed real-time Source (git-checkout) updating in the dashboard, enabling secure, real-time update pipelines for non-NPM installations.

🐛 Bug Fixes & Security

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Hardcoded Secret Cleanup: Removed 12 hardcoded OAuth credential fallbacks from the source code, forcing secure reliance on environment variables and resolving static analysis security alerts.

• Next.js Security Patch: Bumped next from 16.2.2 to 16.2.3 to resolve critical RSC deserialization RCE vulnerability (SNYK-JS-NEXT-15954202).

• Memory/Cache UI Crash: Added null-safety guards (?? 0) to .toLocaleString() calls in Memory and Cache dashboard pages, preventing TypeError crashes when database tables are empty or contain null numeric values (#1083).

• WebSearch tool_choice Translation: Fixed OpenAI-to-Claude translator dropping tool_choice objects with type: "function" as-is, which Claude rejects. Now properly maps all OpenAI tool_choice variants (function, required, none) to Claude-compatible format (tool, any, auto), fixing "Did 0 searches" in Claude Code WebSearch (#1072).

• Provider Validation baseUrl Override: Added baseUrl passthrough from frontend validation requests to the backend validation endpoint. Chinese-site users of Alibaba Coding Plan (bailian-coding-plan) can now validate API keys against their custom Base URL instead of always hitting the international endpoint (#1078).

• Minimax Auth Header: Switched Minimax provider from x-api-key to Authorization: Bearer header format, matching the current API spec (#1076).

• Native Fetch Fallback: Added graceful fallback to native fetch when the undici dispatcher fails, improving resilience in environments where undici is unavailable (#1054).

• EPIPE Flood Fix: Added circuit-breaker logic to prevent EPIPE errors from creating a feedback loop that fills logs at GB/s (#1006).

• Qoder PAT Validation: Improved Qoder Personal Access Token validation with actionable error messages that guide users to the correct token format (#966).

• CI/CD Pipeline: Fixed check:docs-sync failure by syncing OpenAPI version to 3.5.6 and finalizing CHANGELOG release heading. Commented out DATA_DIR in .env.example to prevent E2E test failures in CI runners lacking root permissions.

🌍 i18n

• Auto Language Generation (CI): Added CI pipeline to auto-generate missing language files and strings via feat(CI,i18n) workflow, covering 30+ locales (#1071).

---

[3.5.5] — 2026-04-08

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Node.js 24 Compatibility Warning: Added a proactive version incompatibility warning on the login page to guide users to the stable Node.js 22 LTS, preventing native sqlite binding crashes.

• Context Relay Combo Strategy: Added the new context-relay combo strategy with priority-style routing, structured handoff summary generation once quota usage reaches the warning threshold, and handoff injection after the next real account switch.

• Global Context Relay Defaults: Added global Settings defaults plus combo-level configuration for handoffThreshold, handoffModel, and handoffProviders, so new or unconfigured combos can inherit the feature consistently.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Proxy Connection Healthchecks: Applied proxy resolution per connection in the sweeping loop (tokenHealthCheck.ts) and global provider validation sweeps, resolving Node 22 bypass and improving proxy stability (#1051, #1056, #1061).

• Security Vulnerability Remediation: Resolved multiple CodeQL scanning alerts including SSRF in model sync, insecure randomness in web crypto (generateSessionId), and incomplete URL sanitization.

• Context Relay Typing & Synchronization: Reverted out-of-scope test breakages and resolved handoffProvider and response input extraction payload typing.

• Legacy OpenAI-Compatible Responses Routing: Fixed legacy/imported OpenAI-compatible providers (for example openai-compatible-sp-openai) incorrectly routing Chat Completions traffic to /chat/completions when the real provider node was configured as apiType: "responses". OmniRoute now treats providerSpecificData.apiType as authoritative across routing, executors, and translator tools, avoiding false empty-content failures during combo/provider smoke tests (#1069).

• Gemini PDF Attachment Integration: Fixed payload generation and format for parsing inline_data and generic base64 sources for deep Gemini PDF routing (#993, #1021).

• Vercel AI SDK Fallbacks: Mapped max_output_tokens to max_tokens for strict OpenAI-compatible providers, resolving errors from standard AI agents and frameworks (#994).

• External Auth & UI Reliability: Handled null state failures in Cline OAuth exchange (#1016), added 3rd-party 400 error patterns to combo fallback (#1024), and resolved desktop sidebar layout and popover overflows (#1039, #1001).

• Context Relay In-Flight Deduplication: Prevented duplicate handoff generation for the same session/combo while an earlier summary request is still in flight.

• Context Relay Provider Gating: Aligned runtime behavior with configuration so explicit handoffProviders exclusions, including an empty array, now disable handoff generation as expected.

🛠️ Maintenance & Dependabot

• Updated Sub-dependencies: Bumped hono to 4.12.12 and @hono/node-server to 1.19.13 to patch critical security gaps (#1063, #1064, #1067, #1068).

📚 Documentation

• Documentation Synchronization: Updated system documentation (README, Architecture, Features, Tools, Troubleshooting) and synced i18n configurations to match the v3.5.5 context relay patterns and proxy troubleshooting steps.
• Context Relay Delivery Notes: Documented the current architecture, runtime flow, and Codex-focused scope in the feature docs, changelog, and agent guidance.

---

[3.5.4] — 2026-04-07

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Detailed Token Tracking: Added granular token breakdown columns (cache read, cache write, reasoning) to call logs with proper null vs zero distinction. Includes DB migration 018 and 5-label UI display per provider capability (#1017 — thanks @rdself).

• Legacy JSON Config Import/Export: Restored JSON-based settings export and import for migration from legacy configurations. Security-hardened with Zero-Trust redaction of passwords and requireLogin fields, and automatic pre-import database backups (#1012 — thanks @luandiasrj).

• Non-Stream Aliases: Added API support for explicit non-streaming aliases (non_stream, disable_stream, disable_streaming, streaming=false), normalized at the boundary before provider translation (#1036 — thanks @wlfonseca).

• Russian Dashboard Localization: Comprehensive Russian translation for the dashboard UI, including fixes for 2 Ukrainian locale keys (#1003 — thanks @mercs2910).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Anthropic Streaming Input Undercount: Fixed a critical bug where Anthropic streaming prompt_tokens only reported non-cached tokens (e.g., in=3 when actual total was 113,616). Cache tokens are now summed into prompt_tokens during streaming (#1017).

• Built-in Responses API Tool Types: Preserved built-in Responses API tools (web_search, file_search, computer, code_interpreter, image_generation) from being silently stripped by the empty-name tool filter — these tools carry no .name field (#1014 — thanks @rdself).

• Cursor/Codex Responses Compatibility: Fixed empty output in Cursor when using Codex models by hoisting system input items to instructions, sanitizing invalid tool names, and detecting Responses-format payloads on chat/completions endpoint (#1002 — thanks @mercs2910).

• OAuth Token Expiry Display: Fixed OAuth connections showing "expired" badge even with valid tokens by reading tokenExpiresAt (updated on refresh) instead of expiresAt (original grant timestamp) (#1032 — thanks @tombii).

• Codex Fast-Tier Copy: Corrected dashboard settings copy from service_tier=fast to service_tier=priority, matching the actual Codex wire format (#1045 — thanks @kfiramar).

• macOS Desktop App Startup: Stabilized packaged macOS app launch by excluding desktop artifacts from the standalone bundle and improving launch path detection (#1004 — thanks @mercs2910).

• macOS Sidebar Layout: Fixed macOS traffic light overlap, sidebar spacing, and button overflow in the Electron desktop app (#1001 — thanks @mercs2910).

⚡ Performance

• Analytics Page Load: Dramatically reduced analytics page load times (30s→1-2s for 50K entries) via date-filtered DB queries, parallel Promise.all() cost calculations, and merged 6 COUNT queries into a single CASE WHEN aggregate (#1038 — thanks @oyi77).

🔒 Security & Dependencies

• Node Base Image: Upgraded Docker base from 22-bookworm-slim to 22.22.2-trixie-slim (#1011 — Snyk).
• Production Dependencies: Bumped 5 production dependencies (#1044 — Dependabot).
• Vite: Bumped from 8.0.3 to 8.0.5 (#1031 — Dependabot).
• Development Dependencies: Bumped 4 development dependencies (#1030 — Dependabot).

🧪 Tests

• Token Accounting Tests: Added 18 new unit tests covering detailed token breakdown, null vs zero semantics, per-provider token extraction, and Anthropic streaming input fix (#1017).
• Built-in Tool Tests: Added 3 new test cases for built-in Responses API tool type preservation (#1014).
• ChatCore Sanitization: Updated sanitization tests to accommodate Responses format detection (PR #1002) and built-in tool preservation (PR #1014).

🛠️ Maintenance

• PR Workflow: Updated /review-prs workflow to merge PRs into the release branch (release/vX.Y.Z) instead of directly into main, ensuring proper pre-release staging.

Coverage

• 2537 tests, 2532 passing — Statement coverage: 91.95%, Branch coverage: 78.79%, Function coverage: 93.19%

[3.5.3] - 2026-04-07

Security

• Vulnerabilities: Fully remediated 12 High-Severity CodeQL vulnerabilities by migrating from Math.random to crypto.randomUUID(), wrapping SSE injection points with aggressive backslash escaping, sanitizing trailing HTTP fragments, and enforcing rigid SSRF HTTP verification schemes across internal routes.
• Dependencies: Upgraded Next.js to ^16.2.2 and Vite to >=8.0.5 resolving critical DoS, arbitrary file reads and CSRF vectors in the build/server environments.

Fixed

• E2E Stability: Eliminated extreme CI unreliability and transient test timeouts (Playwright) by propagating internal standalone _next/static assets properly and refactoring deep UI interactions inside defensive expect().toPass() loops.
• Middleware: Resolved infinite redirect loop on dashboard for fresh instances when requireLogin is disabled.
• Core Fallbacks: Preserved primary failure contexts and enhanced Edge-case error handling pipelines across chat and fallback loops.
• Proxy/Hooks: Optimized local git hooks, normalized token coverage endpoints into /coverage, and guarded GLM region lookups.

🛠️ Maintenance

• CI/CD Stabilization: Prevented random GitHub Runner freezes by decoupling sharded processes, adjusting test concurrencies, unref-ing active connections on server teardown, and strictly capping job timeout durations.

Documentation

• I18n Engine: Synchronized and pushed deep Machine Translation updates across all 32 natively-supported languages (682 translation nodes aligned).

Coverage

• Testing: Consolidated the workspace test coverage framework hitting 92.1% statement line coverage, with new rigid unit-tests matching API key policies and tool scopes.

---

[3.5.2] — 2026-04-05

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Qoder API Native Integration: Completely refactored the Qoder Executor to bypass the legacy COSY AES/RSA encryption algorithm, routing directly into the native DashScope OpenAi-compatible URL. Eliminates complex dependencies on Node crypto modules while improving stream fidelity.

• Resilience Engine Overhaul: Integrated context overflow graceful fallbacks, proactive OAuth token detection, and empty-content emission prevention (#990).

• Context-Optimized Routing Strategy: Added new intelligent routing capability to natively maximize context windows in automated combo deployments (#990).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Responses API Stream Corruption: Fixed deep-cloning corruption where Anthropic/OpenAI translation boundaries stripped response. specific SSE prefixes from streaming boundaries (#992).

• Claude Cache Passthrough Alignment: Aligned CC-Compatible cache markers consistently with upstream Client Pass-Through mode preserving prompt caching.

• Turbopack Memory Leak: Pinned Next.js to strict 16.0.10 preventing memory leaks and build staleness from recent upstream Turbopack hashed module regressions (#987).

---

[3.5.1] — 2026-04-04

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Models.dev Integration: Integrated models.dev as the authoritative runtime source for model pricing, capabilities, and specifications, overriding hardcoded prices. Includes a settings UI to manage sync intervals, translation strings for all 30 languages, and robust test coverage.

• Provider Native Capabilities: Added support for declaring and checking native API features (e.g. systemInstructions_supported) preventing failures by sanitizing invalid roles. Currently configured for Gemini Base and Antigravity OAuth providers.

• API Provider Advanced Settings: Added per-connection custom User-Agent overrides for API-key provider connections. The override is stored in providerSpecificData.customUserAgent and now applies to validation probes and upstream execution requests.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Qwen OAuth Reliability: Resolved a series of OAuth integration issues including a 400 Bad Request blocker on expired tokens, fallback generation for parsing OIDC access_token properties when id_token is omitted, model catalog discovery errors, and strict filtering of X-Dashscope-* headers to avoid 400 rejection from OpenAI-compatible endpoints.

[3.5.0] — 2026-04-03

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Auto-Combo & Routing: Completed native CRUD lifecycle integration for the advanced Auto-Combo engine (#955).

• Core Operations: Fixed missing translations for new native Auto-Combos options (#955).

• Security Validation: Disabled SQLite auto-backup tasks natively during unit test CI execution to explicitly resolve Node 22 Event Loop hanging memory leaks (#956).

• Ecosystem Proxies: Completed explicit integration mapping model synchronization schedulers, OAuth cycles, and Token Check refreshes safely through OmniRoute's native system upstream proxies (#953).

• MCP Extensibility: Added and successfully registered the new omniroute_web_search MCP framework tool out of beta into production schemas (#951).

• Tokens Buffer Logic: Added runtime configuration limits extending configurable input/output token buffers for precise Usage Tracking metrics (#959).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• CodeQL Remediation: Fully resolved and secured critical string indexing operations preventing Server-Side Request Forgery (SSRF) arrays indexing heuristics alongside polynomial algorithmic backtracking (ReDoS) inside deep proxy dispatcher modules.

• Crypto Hashes: Replaced weak unverified legacy OAuth 1.0 hashes with robust HMAC-SHA-256 standard validation primitives ensuring tight access controls.

• API Boundary Protection: Correctly verified and mapped structural route protections enforcing strict isAuthenticated() middleware logic covering newer dynamic endpoints targeting settings manipulation and native skills loading.

• CLI Ecosystem Compat: Resolved broken native runtime parser bindings crashing where environment detectors strictly over .cmd/.exe edge cases gracefully for external plugins (#969).

• Cache Architecture: Refactored exact Analytics and System Settings dashboard parameters layout structure caching to maintain stable re-hydration persistence cycles resolving visual unaligned state flashes (#952).

• Claude Caching Standards: Normalized and accurately strictly preserved critical ephemeral block markers ephemeral caching TTL orders for downstream nodes enforcing standard compatible CC requests mapping cleanly without dropped metrics (#948).

• Internal Aliases Auth: Simplified internal runtime mappings normalizing Codex credential payload lookups inside global translation parameters resolving 401 unauthenticated drops (#958).

🛠️ Maintenance

• UI Discoverability: Correctly adjusted layout categorizations explicitly separating free tier providers logic improving UX sorting flows inside the general API registry pages (#950).
• Deployment Topology: Unified Docker deployment artifacts ensuring the root fly.toml matches expected cloud instance parameters out-of-the-box natively handling automated deployments scaling properly.
• Development Tooling: Decoupled LKGP runtime parameters into explicit DB layer abstraction caching utilities ensuring strict test isolation coverage for core caching layers safely.

---

[3.4.9] — 2026-04-03

Features & Refactoring

• Dashboard Auto-Combo Panel: Completely refactored the /dashboard/auto-combo UI to seamlessly integrate with native Dashboard Cards and standardized visual padding/headers. Added dynamic visual progress bars mapping model selection weight mechanisms.
• Settings Routing Sync: Fully exposed advanced routing priority and weighted schema targets internally inside global settings fallback lists.

Bug Fixes

• Memory & Skills Locale Nodes: Resolved empty rendering tags for Memory and Skills options directly inside global settings views by wiring all settings.* mapping values internally into en.json (also mapped implicitly for cross-translation tools).

Internal Integrations

• Integrated PR #946 — fix: preserve Claude Code compatibility in responses conversion
• Integrated PR #944 — fix(gemini): preserve thought signatures across antigravity tool calls
• Integrated PR #943 — fix: restore GitHub Copilot body
• Integrated PR #942 — Fix cc-compatible cache markers
• Integrated PR #941 — refactor(auth): improve NVIDIA alias lookup + add LKGP error logging
• Integrated PR #939 — Restore Claude OAuth localhost callback handling
• (Note: PR #934 was omitted from 3.4.9 cycle to prevent core conflict regressions)

---

[3.4.8] — 2026-04-03

Security

• Fully remediated all outstanding Github Advanced Security (CodeQL) findings and Dependabot alerts.
• Fixed insecure randomness vulnerabilities by migrating from Math.random to crypto.randomUUID().
• Secured shell commands in automated scripts from string injection.
• Migrated vulnerable catastrophic backtracking RegEx parsing patterns in chat/translation pipelines.
• Enhanced output sanitization controls inside React UI components and Server Sent Events (SSE) tag injection.

---

[3.4.7] — 2026-04-03

Features

• Added Cryptography node to Monitoring and MCP health checks (#798)
• Hardened model-catalog route permissions mapping (/models) (#781)

Bug Fixes

• Fixed Claude OAuth token refreshes failing to preserve cache contexts (#937)
• Fixed CC-Compatible provider errors rendering cached models unreachable (#937)
• Fixed GitHub Executor errors related to invalid context arrays (#937)
• Fixed NPM-installed CLI tools healthcheck failures on Windows (#935)
• Fixed payload translation dropping valid content due to invalid API fields (#927)
• Fixed runtime crash in Node 25 regarding API key execution (#867)
• Fixed MCP standalone module-resolution (ERR_MODULE_NOT_FOUND) via esbuild (#936)
• Fixed NVIDIA NIM routing credential resolution alias mismatch (#931)

Security

• Added safe strict input boundary protection against raw shell: true remote-code execution injections.

---

[3.4.6] - 2026-04-02

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Providers: Registered new image, video, and audio generation providers from the community-requested list (#926).

• Dashboard UI: Added standalone sidebar navigation for the new Memory and Skills modules (#926).

• i18n: Added translation strings and layout mappings across 30 languages for the Memory and Skills namespaces.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Resilience: Prevented the proxy Circuit Breaker from becoming stuck in an OPEN state indefinitely by handling direct transitions to CLOSED state inside fallback combo paths (#930).

• Protocol Translation: Patched the streaming transformer to sanitize response blocks based on the expected source protocol rather than the provider target protocol, fixing Anthropics models wrapped in OpenAI payloads crashing Claude Code (#929).

• API Specs & Gemini: Fixed thought_signature parsing in openai-to-gemini and claude-to-gemini translators, preventing HTTP 400 errors across all Gemini 3 API tool-calls.

• Providers: Cleaned up non-OpenAI-compatible endpoints preventing valid upstream connections (#926).

• Cache Trends: Fixed an invalid property mapping data mismatch causing Cache Trends UI charts to crash, and extracted redundant cache metric widgets (#926).

---

[3.4.5] - 2026-04-02

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• CLIProxyAPI Ecosystem Integration: Added the cliproxyapi executor with built-in module-level caching and proxy routing. Introduced a comprehensive Version Manager service to automatically test health, download binaries from GitHub, spawn isolated background processes, and cleanly manage the lifecycle of external CLI tools directly through the UI. Includes DB tables for proxy configuration to enable automatic SSRF-gated cross-routing of external OpenAI requests via the local CLI tool layer (#914, #915, #916).

• Qoder PAT Support: Integrated Personal Access Tokens (PAT) support directly via the local qodercli transport instead of legacy remote .cn browser configurations (#913).

• Gemini 3.1 Pro Preview (GitHub): Added gemini-3.1-pro-preview canonical explicit model support natively into the GitHub Copilot provider while preserving older routing aliases (#924).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• GitHub Copilot Token Stability: Repaired the Copilot token refresh loop where stale tokens weren't deep-merged into DB, and removed reasoning_text fields that were fatally breaking downstream Anthropic block conversions for multi-turn chats (#923).

• Global Timeout Matrix: Centralized and parameterized request timeouts explicitly from REQUEST_TIMEOUT_MS to prevent hidden (~300s) default fetch buffers prematurely cutting off long-lived SSE streaming responses from heavy reasoning models (#918).

• Cloudflare Quick Tunnels State: Fixed a severe state inconsistency where restarted OmniRoute instances erroneously showed destroyed tunnels as active, and defaulted cloudflared tunneling to HTTP/2 to eliminate UDP receive buffer log spam (#925).

• i18n Translation Overhaul (Czech & Hindi): Fixed Hindi code from DEPRECATED in.json to canonical hi.json, overhauled Czech text mappings, extracted untranslatable-keys.json to fix CI/CD false-positive validations, and generated comprehensive I18N.md docs to guide translators (#912).

• Tokens Provider Recovery: Fixed Qwen losing specific resourceUrl endpoints after automatic health-check token refreshes because of missing DB deep merges (#917).

• CC Compatible UX & Streaming: Unified the Add CC/OpenAI/Anthropic compatible actions around the Anthropic UI treatment, forced CC-compatible upstream requests to use SSE while still returning streaming or non-streaming responses based on the client request, removed CC model-list configuration/import support in favor of an explicit unsupported-model-listing error, and made CC-compatible Available Models mirror the OAuth Claude Code registry list (#921).

---

[3.4.4] - 2026-04-02

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Responses API Token Reporting: Emit response.completed with correct input_tokens/output_tokens fields for Codex CLI clients, fixing token usage display (#909 — thanks @christopher-s).

• SQLite WAL Checkpoint on Shutdown: Flush WAL changes into the primary database file during graceful shutdown/restart, preventing data loss on Docker container stops (#905 — thanks @rdself).

• Graceful Shutdown Signal: Changed /api/restart and /api/shutdown routes from process.exit(0) to process.kill(SIGTERM), ensuring the shutdown handler runs before exit.

• Docker Stop Grace Period: Added stop_grace_period: 40s to Docker Compose files and --stop-timeout 40 to Docker run examples.

🛠️ Maintenance

• Closed 5 resolved/not-a-bug issues (#872, #814, #816, #890, #877).
• Triaged 6 issues with needs-info requests (#892, #887, #886, #865, #895, #870).
• Responded to CLI detection tracking issue (#863) with contributor guidance.

---

[3.4.3] - 2026-04-02

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Antigravity Memory & Skills: Completed remote memory and skills injection for the Antigravity provider at the proxy network level.

• Claude Code Compatibility: Built a natively hidden compatibility bridge for Claude Code, passing tools and formatting through cleanly.

• Web Search MCP: Added the omniroute_web_search tool with the execute:search scope.

• Cache Components: Implemented dynamic cache components utilizing TDD.

• UI & Customization: Added custom favicon support, appearance tabs, wired whitelabeling to the sidebar, and added Windsurf guide steps across all 33 languages.

• Log Retention: Unified request log retention and artifacts natively.

• Model Enhancements: Added explicit contextLength for all opencode-zen models.

• i18n & translations: Integrated 33 language translations natively, including placeholder CI validations and Chinese documentation updates (#873, #869).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Qwen OAuth Mapping: Reverted id_token reliance to access_token and enabled dynamic resource_url API endpoint injection for proper regional routing (#900).

• Model Sync Engine: Stored the strict internal Provider ID in getCustomModels() sync routines instead of the UI Channel Alias format, preventing SQLite catalog insertion failures (#903).

• Claude Code & Codex: Standardized non-streaming blank responses to Anthropic-formatted (empty response) to prevent CLI proxy crashes (#866).

• CC Compatible Routing: Resolved duplicate /v1 endpoint collision during path concatenation for generic Claude Code gateways (#904).

• Antigravity Dashboards: Blocked unlimited quota models from falsely registering as exhausted 100% Usage limit states in the Provider Usage UI (#857).

• Claude Image Passthrough: Fixed Claude models missing image block passthroughs (#898).

• Gemini CLI Routing: Resolved 403 authorization lockouts and content accumulation issues by refreshing the project ID via loadCodeAssist (#868).

• Antigravity Stability: Corrected model access lists, enforced 404 lockouts, fixed 429 cascades locking out standard connections, and capped gemini-3.1-pro output tokens (#885).

• Provider Sync Cadence: Repaired the provider limits synchronization cadence via the internal scheduler (#888).

• Dashboard Optimization: Resolved /dashboard/limits UI freezing when processing 70+ accounts via chunk parallelization (#784).

• SSRF Hardening: Enforced strict SSRF IP range filtering and blocked the ::1 loopback interface.

• MIME Types: Standardized mime_type to snake_case to match Gemini API specifications.

• CI Stabilization: Fixed failing analytics/settings Playwright selectors and request assertions so GitHub Actions E2E runs pass reliably across localized UIs and switch-based controls.

• Deterministic Tests: Removed date-sensitive quota fixtures from Copilot usage tests and aligned idempotency/model catalog tests with the merged runtime behavior.

• MCP Type Hardening: Removed zero-budget explicit any regressions from the MCP server tool registration path.

• Model Sync Engine: Bypassed destructive replace overrides when the provider's auto-sync yields an empty model list, maintaining stability for dynamic catalogs (#899).

🛠️ Maintenance

• Pipeline Logging: Refined pipeline logging artifacts and enforce retention caps (#880).
• AGENTS.md Overhaul: Condensed from 297→153 lines. Added build/test/style guidelines, code workflows (Prettier, TypeScript, ESLint), and trimmed verbose tables (#882).
• Release Branch Integration: Consolidated the active feature branches into release/v3.4.2 on top of current main and validated the branch with lint, unit, coverage, build, and CI-mode E2E runs.
• Testing: Added vitest configuration for component testing and Playwright specs for settings toggles.
• Doc Updates: Expanded root readmes, translated chinese documents natively, and cleaned up obsolete files.

[3.4.1] - 2026-03-31

Warning

BREAKING CHANGE: request logging, retention, and logging environment variables have been redesigned. On the first startup after upgrading, OmniRoute archives legacy request logs from DATA_DIR/logs/, legacy DATA_DIR/call_logs/, and DATA_DIR/log.txt into DATA_DIR/log_archives/*.zip, then removes the deprecated layout and switches to the new unified artifact format under DATA_DIR/call_logs/.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• .ENV Migration Utility: Included scripts/migrate-env.mjs to seamlessly migrate <v3.3 configurations to v3.4.x strict security validation constraints (FASE-01), repairing startup crashes caused by short JWT_SECRET instances.

• Kiro AI Cache Optimization: Implemented deterministic conversationId generation (uuidv5) to enable AWS Builder ID Prompt Caching properly across invocations (#814).

• Dashboard UI Restoration & Consolidation: Resolved sidebar logic omitting the Debug section, and cleared Nextjs routing warnings by moving standalone /dashboard/mcp and /dashboard/a2a pages explicitly into embedded Endpoint Proxy UI components.

• Unified Request Log Artifacts: Request logging now stores one SQLite index row plus one JSON artifact per request under DATA_DIR/call_logs/, with optional pipeline capture embedded in the same file.

• Language: Improved the Chinese translation (#855)

• Opencode-Zen Models: Added 4 free models to opencode-zen registry (#854)

• Tests: Added unit and E2E tests for settings toggles and bug fixes (#850)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• 429 Quota Parsing: Parsed long quota reset times from error bodies to honor correct backoffs and prevent rate-limited account bans (#859)

• Prompt Caching: Preserved client cache_control headers for all Claude-protocol providers (like Minimax, GLM, and Bailian), correctly recognizing caching support (#856)

• Model Sync Logs: Reduced log spam by recording sync-models only when the channel actually modifies the list (#853)

• Provider Quota & Token Parsing: Switched Antigravity limits to use retrieveUserQuota natively and correctly mapped Claude token refresh payloads to URL-encoded forms (#862)

• Rate-Limiting Stability: Universalized the 429 Retry-After parsing architecture to cap provider-induced cooldowns at 24 hours max (#862)

• Dashboard Limit Rendering: Re-architected /dashboard/limits quota mapping to render immediately inside chunks, fixing a major UI freezing delay on accounts exceeding 70 active connections (#784)

• QWEN OAuth Authorization: Mapped the OIDC id_token as the primary API Bearer token for Dashscope requests, fixing immediate 401 Unauthorized errors after connecting accounts or refreshing tokens (#864)

• ZAI API Stability: Hardened Server-Sent Events compiler to gracefully fallback to empty strings when DeepSeek providers stream mathematically null content during reasoning phases (#871)

• Claude Code/Codex Translations: Protected non-streaming payload conversions against empty responses from upstream Codex tools, avoiding catastrophic TypeErrors (#866)

• NVIDIA NIM Rendering: Conditionally stripped identical provider prefixes dynamically pushed by audio models, eliminating duplicate nim/nim tag structures throwing 404 on the Media Playground (#872)

⚠️ Breaking Changes

• Request Log Layout: Removed the old multi-file DATA_DIR/logs/ request log sessions and the DATA_DIR/log.txt summary file. New requests are written as single JSON artifacts in DATA_DIR/call_logs/YYYY-MM-DD/.
• Logging Environment Variables: Replaced LOG_*, ENABLE_REQUEST_LOGS, CALL_LOGS_MAX, CALL_LOG_PAYLOAD_MODE, and PROXY_LOG_MAX_ENTRIES with the new APP_LOG_* and CALL_LOG_RETENTION_DAYS configuration model.
• Pipeline Toggle Setting: Replaced the legacy detailed_logs_enabled setting with call_log_pipeline_enabled. New pipeline details are embedded inside the request artifact instead of being stored as separate request_detail_logs records.

🛠️ Maintenance

• Legacy Request Log Upgrade Backup: Upgrades now archive old data/logs/, legacy data/call_logs/, and data/log.txt layouts into DATA_DIR/log_archives/*.zip before removing the deprecated structure.
• Streaming Usage Persistence: Streaming requests now write a single usage_history row on completion instead of emitting a duplicate in-progress usage row with empty status metadata.
• Logging Follow-up Cleanup: Pipeline logs no longer capture SOURCE REQUEST, request artifact entries now honor CALL_LOG_MAX_ENTRIES, and application log archives now honor APP_LOG_MAX_FILES.

---

[3.4.0] - 2026-03-31

🚀 Features

• Subscription Utilization Analytics: Added quota snapshot time-series tracking, Provider Utilization and Combo Health tabs with recharts visualizations, and corresponding API endpoints (#847)
• SQLite Backup Control: New OMNIROUTE_DISABLE_AUTO_BACKUP env flag to disable automatic SQLite backups (#846)
• Model Registry Update: Injected gpt-5.4-mini into the Codex provider's array of models (#756)
• Provider Limit Tracking: Track and display when provider rate limits were last refreshed per account (#843)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Qwen Auth Routing: Re-routed Qwen OAuth completions from the DashScope API to the Web Inference API (chat.qwen.ai), resolving authorization failures (#844, #807, #832)

• Qwen Auto-Retry Loop: Added targeted 429 Quota Exceeded backoff handling inside chatCore protecting burst requests

• Codex OAuth Fallback: Modern browser popup blocking no longer traps the user; it automatically falls back to manual URL entry (#808)

• Claude Token Refresh: Anthropic's strict application/json boundaries are now respected during token generation instead of encoded URLs (#836)

• Codex Messages Schema: Stripped purist messages injects from native passthrough requests to avoid structural rejections from the ChatGPT upstream (#806)

• CLI Detection Size Limit: Safely bumped the Node binary scanning upper bound from 100MB to 350MB, allowing heavy standalone tools like Claude Code (229MB) and OpenCode (153MB) to be correctly detected by the VPS runtime (#809)

• CLI Runtime Environment: Restored ability for CLI configurations to respect user override paths (CLI_{PROVIDER}_BIN) bypassing strict path-bound discovery rules

• Nvidia Header Conflicts: Removed prompt_cache_key properties from upstream headers when calling non-Anthropic providers (#848)

• Codex Fast Tier Toggle: Restored Codex service tier toggle contrast in light mode (#842)

• Test Infrastructure: Updated t28-model-catalog-updates test that incorrectly expected the outdated DashScope endpoint for the Qwen native registry

---

[3.3.9] - 2026-03-31

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Custom Provider Rotation: Integrated getRotatingApiKey internally inside DefaultExecutor, ensuring extraApiKeys rotation triggers correctly for custom and compatible upstream providers (#815)

---

[3.3.8] - 2026-03-30

🚀 Features

• Models API Filtering: Endpoint /v1/models now dynamically filters its list based on the permissions tied to the Authorization: Bearer <token> when restricted access is on (#781)
• Qoder Integration: Native integration for Qoder AI natively replacing the legacy iFlow platform mappings (#660)
• Prompt Cache Tracking: Added tracking capabilities and frontend visualization (Stats card) for semantic and prompt caching in the Dashboard UI

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Cache Dashboard Sizing: Improved the UI layout sizes and context headers for the advanced cache pages (#835)

• Debug Sidebar Visibility: Fixed an issue where the debug toggle wouldn't correctly show/hide sidebar debug details (#834)

• Gemini Model Prefixing: Modified the namespace fallback to properly route via gemini-cli/ instead of gc/ to respect upstream specs (#831)

• OpenRouter Sync: Improved compatibility synchronization to automatically ingest the available models catalog correctly from OpenRouter (#830)

• Streaming Payloads Mapping: Reserialization of reasoning fields natively resolves conflict alias paths when output is streaming to edge devices

---

[3.3.7] - 2026-03-30

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• OpenCode Config: Restructured generated opencode.json to use the @ai-sdk/openai-compatible record-based schema with options and models as object maps instead of flat arrays, fixing config validation failures (#816)

• i18n Missing Keys: Added missing cloudflaredUrlNotice translation key across all 30 language files to prevent MISSING_MESSAGE console errors in the Endpoint page (#823)

---

[3.3.6] - 2026-03-30

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Token Accounting: Included prompt cache tokens safely in historical usage inputs calculations for correct quota deductions (PR #822)

• Combo Test Probes: Fixed combo testing logic false negatives by resolving parsing for reasoning-only responses and enabled massive parallelization via Promise.all (PR #828)

• Docker Quick Tunnels: Embedded required ca-certificates inside the base runtime container to resolve Cloudflared TLS startup failures, and surfaced stdout network errors replacing generic exit codes (PR #829)

---

[3.3.5] - 2026-03-30

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Gemini Quota Tracking: Added real-time Gemini CLI quota tracking via the retrieveUserQuota API (PR #825)

• Cache Dashboard: Enhanced the Cache Dashboard to display prompt cache metrics, 24h trends, and estimated cost savings (PR #824)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• User Experience: Removed invasive auto-opening OAuth modal loops on barren provider detailed pages (PR #820)

• Dependency Updates: Bumped and locked down dependencies for development and production trees including Next.js 16.2.1, Recharts, and TailwindCSS 4.2.2 (PR #826, #827)

---

[3.3.4] - 2026-03-30

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• A2A Workflows: Added deterministic FSM orchestrator for multi-step agent workflows.

• Graceful Degradation: Added a new multi-layer fallback framework to preserve core functionality during partial system outages.

• Config Audit: Added an audit trail with diff detection to track changes and enable configuration rollbacks.

• Provider Health: Added provider expiration tracking with proactive UI alerts for expiring API keys.

• Adaptive Routing: Added an adaptive volume and complexity detector to override routing strategies dynamically based on load.

• Provider Diversity: Implemented provider diversity scoring via Shannon entropy to improve load distribution.

• Auto-Disable Bounds: Added an Auto-Disable Banned Accounts setting toggle to the Resilience dashboard.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Codex & Claude Compatibility: Fixed UI fallbacks, patched Codex non-streaming integration issues, and resolved CLI runtime detection on Windows.

• Release Automation: Expanded permissions required for the Electron App build in GitHub Actions.

• Cloudflare Runtime: Addressed correct runtime isolation exit codes for Cloudflared tunnel components.

🧪 Tests

• Test Suite Updates: Expanded test coverage for volume detectors, provider diversity, configuration audit, and FSM.

---

[3.3.3] - 2026-03-29

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• CI/CD Reliability: Patched GitHub Actions to stable dependency versions (actions/checkout@v4, actions/upload-artifact@v4) to mitigate unannounced builder environment deprecations.

• Image Fallbacks: Replaced arbitrary fallback chains in ProviderIcon.tsx with explicit asset validation to prevent UI loading <Image> components for files that don't exist, eliminating 404 errors in dashboard console logs (#745).

• Admin Updater: Dynamic source-installation detection for the dashboard Updater. Safely disables the Update Now button when OmniRoute is built locally rather than through npm, prompting for git pull (#743).

• Update ERESOLVE Error: Injected package.json overrides for react/react-dom and enabled --legacy-peer-deps within the internal automatic updater scripts to resolve breaking dependency tree conflicts with @lobehub/ui.

---

[3.3.2] - 2026-03-29

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Cloudflare Tunnels: Cloudflare Quick Tunnel integration with dashboard controls (PR #772).

• Diagnostics: Semantic cache bypass for combo live tests (PR #773).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Streaming Stability: Apply FETCH_TIMEOUT_MS to streaming requests' initial fetch() call to prevent 300s Node.js TCP timeout causing silent task failures (#769).

• i18n: Add missing windsurf and copilot entries to toolDescriptions across all 33 locale files (#748).

• GLM Coding Audit: Complete provider audit fixing ReDoS vulnerabilities, context window sizing (128k/16k), and model registry syncing (PR #778).

---

[3.3.1] - 2026-03-29

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• OpenAI Codex: Fallback processing fix for type: "text" elements carrying null or empty datasets that caused 400 rejection (#742).

• Opencode: Update schema alignment to singular provider to match official spec (#774).

• Gemini CLI: Inject missing end-user quota headers preventing 403 authorization lockouts (#775).

• DB Recovery: Refactor multipart payload imports into raw binary buffered arrays to bypass reverse proxy max body limits (#770).

---

[3.3.0] - 2026-03-29

✨ Enhancements & Refactoring

• Release Stabilization — Finalized v3.2.9 release (combo diagnostics, quality gates, Gemini tool fix) and created missing git tag. Consolidated all staged changes into a single atomic release commit.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Auto-Update Test — Fixed buildDockerComposeUpdateScript test assertion to match unexpanded shell variable references ($TARGET_TAG, ${TARGET_TAG#v}) in the generated deploy script, aligning with the refactored template from v3.2.8.

• Circuit Breaker Test — Hardened combo-circuit-breaker.test.mjs by injecting maxRetries: 0 to prevent retry inflation from skewing failure count assertions during breaker state transitions.

---

[3.2.9] - 2026-03-29

✨ Enhancements & Refactoring

• Combo Diagnostics — Introduced a live test bypass flag (forceLiveComboTest) allowing administrators to execute real upstream health checks that bypass all local circuit-breaker and cooldown state mechanisms, enabling precise diagnostics during rolling outages (PR #759)
• Quality Gates — Added automated response quality validation for combos and officially integrated claude-4.6 model support into the core routing schemas (PR #762)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Tool Definition Validation — Repaired Gemini API integration by normalizing enum types inside tool definitions, preventing upstream HTTP 400 parameter errors (PR #760)

---

[3.2.8] - 2026-03-29

✨ Enhancements & Refactoring

• Docker Auto-Update UI — Integrated a detached background update process for Docker Compose deployments. The Dashboard UI now seamlessly tracks update lifecycle events combining JSON REST responses with SSE streaming progress overlays for robust cross-environment reliability.
• Cache Analytics — Repaired zero-metrics visualization mapping by migrating Semantic Cache telemetry logs directly into the centralized tracking SQLite module.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Authentication Logic — Fixed a bug where saving dashboard settings or adding models failed with a 401 Unauthorized error when requireLogin was disabled. API endpoints now correctly evaluate the global authentication toggle. Resolved global redirection by reactivating src/middleware.ts.

• CLI Tool Detection (Windows) — Prevented fatal initialization exceptions during CLI environment detection by catching cross-spawn ENOENT errors correctly. Adds explicit detection paths for \AppData\Local\droid\droid.exe.

• Codex Native Passthrough — Normalized model translation parameters preventing context poisoning in proxy pass-through mode, enforcing generic store: false constraints explicitly for all Codex-originated requests.

• SSE Token Reporting — Normalized provider tool-call chunk finish_reason detection, fixing 0% Usage analytics for stream-only responses missing strict <DONE> indicators.

• DeepSeek Tags — Implemented an explicit <think> extraction mapping inside responsesHandler.ts, ensuring DeepSeek reasoning streams map equivalently to native Anthropic <thinking> structures.

---

[3.2.7] - 2026-03-29

Fixed

• Seamless UI Updates: The "Update Now" feature on the Dashboard now provides live, transparent feedback using Server-Sent Events (SSE). It performs package installation, native module rebuilds (better-sqlite3), and PM2 restarts reliably while showing real-time loaders instead of silently hanging.

---

[3.2.6] — 2026-03-29

✨ Enhancements & Refactoring

• API Key Reveal (#740) — Added a scoped API key copy flow in the Api Manager, protected by the ALLOW_API_KEY_REVEAL environment variable.
• Sidebar Visibility Controls (#739) — Admins can now hide any sidebar navigation link via the Appearance settings to reduce visual clutter.
• Strict Combo Testing (#735) — Hardened the combo health check endpoint to require live text responses from models instead of just soft reachability signals.
• Streamed Detailed Logs (#734) — Switched detailed request logging for SSE streams to reconstruct the final payload, saving immense amounts of SQLite database size and significantly cleaning up the UI.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• OpenCode Go MiniMax Auth (#733) — Corrected the authentication header logic for minimax models on OpenCode Go to use x-api-key instead of standard bearer tokens across the /messages protocol.

---

[3.2.5] — 2026-03-29

✨ Enhancements & Refactoring

• Void Linux Deployment Support (#732) — Integrated xbps-src packaging template and instructions to natively compile and install OmniRoute with better-sqlite3 bindings via cross-compilation target.

[3.2.4] — 2026-03-29

✨ Enhancements & Refactoring

• Qoder AI Migration (#660) — Completely migrated the legacy iFlow core provider onto Qoder AI maintaining stable API routing capabilities.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Gemini Tools HTTP 400 Payload Invalid Argument (#731) — Prevented thoughtSignature array injections inside standard Gemini functionCall sequences blocking agentic routing flows.

---

[3.2.3] — 2026-03-29

✨ Enhancements & Refactoring

• Provider Limits Quota UI (#728) — Normalized quota limit logic and data labeling inside the Limits interface.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Core Routing Schemas & Leaks — Expanded comboStrategySchema to natively support fill-first and p2c strategies to unblock complex combo editing natively.

• Thinking Tags Extraction (CLI) — Restructured CLI token responses sanitizer RegEx capturing model reasoning structures inside streams avoiding broken <thinking> extractions breaking response text output format.

• Strict Format Enforcements — Hardened pipeline sanitization execution making it universally apply to translation mode targets.

---

[3.2.2] — 2026-03-29

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Four-Stage Request Log Pipeline (#705) — Refactored log persistence to save comprehensive payloads at four distinct pipeline stages: Client Request, Translated Provider Request, Provider Response, and Translated Client Response. Introduced streamPayloadCollector for robust SSE stream truncation and payload serialization.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Mobile UI Fixes (#659) — Prevented table components on the dashboard from breaking the layout on narrow viewports by adding proper horizontal scrolling and overflow containment to DashboardLayout.

• Claude Prompt Cache Fixes (#708) — Ensured cache_control blocks in Claude-to-Claude fallback loops are faithfully preserved and passed safely back to Anthropic models.

• Gemini Tool Definitions (#725) — Fixed schema translation errors when declaring simple object parameter types for Gemini function calling.

[3.2.1] — 2026-03-29

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Global Fallback Provider (#689) — When all combo models are exhausted (502/503), OmniRoute now attempts a configurable global fallback model before returning the error. Set globalFallbackModel in settings to enable.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Fix #721 — Fixed context pinning bypass during tool-call responses. Non-streaming tagging used wrong JSON path (json.messages → json.choices[0].message). Streaming injection now triggers on finish_reason chunks for tool-call-only streams. injectModelTag() now appends synthetic pin messages for non-string content.

• Fix #709 — Confirmed already fixed (v3.1.9) — system-info.mjs creates directories recursively. Closed.

• Fix #707 — Confirmed already fixed (v3.1.9) — empty tool name sanitization in chatCore.ts. Closed.

🧪 Tests

• Added 6 unit tests for context pinning with tool-call responses (null content, array content, roundtrip, re-injection)

[3.2.0] — 2026-03-28

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Cache Management UI — Added a dedicated semantic caching dashboard at `/dashboard/cache` with targeted API invalidation and 31-language i18n support (PR #701 by @oyi77)

• GLM Quota Tracking — Added real-time usage and session quota tracking for the GLM Coding (Z.AI) provider (PR #698 by @christopher-s)

• Detailed Log Payloads — Wired full four-stage pipeline payload capturing (original, translated, provider-response, streamed-deltas) directly into the UI (PR #705 by @rdself)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Fix #708 — Prevented token bleeding for Claude Code users routing through OmniRoute by correctly preserving native `cache_control` headers during Claude-to-Claude passthrough (PR #708 by @tombii)

• Fix #719 — Setup internal auth boundaries for `ModelSyncScheduler` to prevent unauthenticated daemon failures on startup (PR #719 by @rdself)

• Fix #718 — Rebuilt badge rendering in Provider Limits UI preventing bad quota boundaries overlap (PR #718 by @rdself)

• Fix #704 — Fixed Combo Fallbacks breaking on HTTP 400 content-policy errors preventing model-rotation dead-routing (PR #704 by @rdself)

🔒 Security & Dependencies

• Bumped `path-to-regexp` to `8.4.0` resolving dependabot vulnerabilities (PR #715)

[3.1.10] — 2026-03-28

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Fix #706 — Fixed icon fallback rendering caused by Tailwind V4 font-sans override by applying !important to .material-symbols-outlined.

• Fix #703 — Fixed GitHub Copilot broken streams by enabling responses to openai format translation for any custom models leveraging apiFormat: "responses".

• Fix #702 — Replaced flat-rate usage tracking with accurate DB pricing calculations for both streaming and non-streaming responses.

• Fix #716 — Cleaned up Claude tool-call translation state, correctly parsing streaming arguments and preventing OpenAI tool_calls chunks from repeating the id field.

[3.1.9] — 2026-03-28

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Schema Coercion — Auto-coerce string-encoded numeric JSON Schema constraints (e.g. "minimum": "1") to proper types, preventing 400 errors from Cursor, Cline, and other clients sending malformed tool schemas.

• Tool Description Sanitization — Ensure tool descriptions are always strings; converts null, undefined, or numeric descriptions to empty strings before sending to providers.

• Clear All Models Button — Added i18n translations for the "Clear All Models" provider action across all 30 languages.

• Codex Auth Export — Added Codex auth.json export and apply-local buttons for seamless CLI integration.

• Windsurf BYOK Notes — Added official limitation warnings to the Windsurf CLI tool card documenting BYOK constraints.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Fix #709 — system-info.mjs no longer crashes when the output directory doesn't exist (added mkdirSync with recursive flag).

• Fix #710 — A2A TaskManager singleton now uses globalThis to prevent state leakage across Next.js API route recompilations in dev mode. E2E test suite updated to handle 401 gracefully.

• Fix #711 — Added provider-specific max_tokens cap enforcement for upstream requests.

• Fix #605 / #592 — Strip proxy_ prefix from tool names in non-streaming Claude responses; fixed LongCat validation URL.

• Call Logs Max Cap — Upgraded getMaxCallLogs() with caching layer, env var support (CALL_LOGS_MAX), and DB settings integration.

🧪 Tests

• Test suite expanded from 964 → 1027 tests (63 new tests)
• Added schema-coercion.test.mjs — 9 tests for numeric field coercion and tool description sanitization
• Added t40-opencode-cli-tools-integration.test.mjs — OpenCode/Windsurf CLI integration tests
• Enhanced feature-tests branch with comprehensive coverage tooling

📁 New Files

File	Purpose
open-sse/translator/helpers/schemaCoercion.ts	Schema coercion and tool description sanitization utilities
tests/unit/schema-coercion.test.mjs	Unit tests for schema coercion
tests/unit/t40-opencode-cli-tools-integration.test.mjs	CLI tool integration tests
COVERAGE_PLAN.md	Test coverage planning document

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Claude Prompt Caching Passthrough — Fixed cache_control markers being stripped in Claude passthrough mode (Claude → OmniRoute → Claude), which caused Claude Code users to deplete their Anthropic API quota 5-10x faster than direct connections. OmniRoute now preserves client's cache_control markers when sourceFormat and targetFormat are both Claude, ensuring prompt caching works correctly and dramatically reducing token consumption.

[3.1.8] - 2026-03-27

🐛 Bug Fixes & Features

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Platform Core: Implemented global state handling for Hidden Models & Combos preventing them from cluttering the catalog or leaking into connected MCP agents (#681).

• Stability: Patched streaming crashes related to the native Antigravity provider integration failing due to unhandled undefined state arrays (#684).

• Localization Sync: Deployed a fully overhauled i18n synchronizer detecting missing nested JSON properties and retro-fitting 30 locales sequentially (#685).## [3.1.7] - 2026-03-27

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Streaming Stability: Fixed hasValuableContent returning undefined for empty chunks in SSE streams (#676).

• Tool Calling: Fixed an issue in sseParser.ts where non-streaming Claude responses with multiple tool calls dropped the id of subsequent tool calls due to incorrect index-based deduplication (#671).

---

[3.1.6] — 2026-03-27

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Claude Native Tool Name Restoration — Tool names like TodoWrite are no longer prefixed with proxy_ in Claude passthrough responses (both streaming and non-streaming). Includes unit test coverage (PR #663 by @coobabm)

• Clear All Models Alias Cleanup — "Clear All Models" button now also removes associated model aliases, preventing ghost models in the UI (PR #664 by @rdself)

---

[3.1.5] — 2026-03-27

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Backoff Auto-Decay — Rate-limited accounts now auto-recover when their cooldown window expires, fixing a deadlock where high backoffLevel permanently deprioritized accounts (PR #657 by @brendandebeasi)

🌍 i18n

• Chinese translation overhaul — Comprehensive rewrite of zh-CN.json with improved accuracy (PR #658 by @only4copilot)

---

[3.1.4] — 2026-03-27

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Streaming Override Fix — Explicit stream: true in request body now takes priority over Accept: application/json header. Clients sending both will correctly receive SSE streaming responses (#656)

🌍 i18n

• Czech string improvements — Refined terminology across cs.json (PR #655 by @zen0bit)

---

[3.1.3] — 2026-03-26

🌍 i18n & Community

• ~70 missing translation keys added to en.json and 12 languages (PR #652 by @zen0bit)
• Czech documentation updated — CLI-TOOLS, API_REFERENCE, VM_DEPLOYMENT guides (PR #652)
• Translation validation scripts — check_translations.py and validate_translation.py for CI/QA (PR #651 by @zen0bit)

---

[3.1.2] — 2026-03-26

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Critical: Tool Calling Regression — Fixed proxy_Bash errors by disabling the proxy_ tool name prefix in the Claude passthrough path. Tools like Bash, Read, Write were being renamed to proxy_Bash, proxy_Read, etc., causing Claude to reject them (#618)

• Kiro Account Ban Documentation — Documented as upstream AWS anti-fraud false positive, not an OmniRoute issue (#649)

🧪 Tests

• 936 tests, 0 failures

---

[3.1.1] — 2026-03-26

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Vision Capability Metadata: Added capabilities.vision, input_modalities, and output_modalities to /v1/models entries for vision-capable models (PR #646)

• Gemini 3.1 Models: Added gemini-3.1-pro-preview and gemini-3.1-flash-lite-preview to the Antigravity provider (#645)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Ollama Cloud 401 Error: Fixed incorrect API base URL — changed from api.ollama.com to official ollama.com/v1/chat/completions (#643)

• Expired Token Retry: Added bounded retry with exponential backoff (5→10→20 min) for expired OAuth connections instead of permanently skipping them (PR #647)

🧪 Tests

• 936 tests, 0 failures

---

[3.1.0] — 2026-03-26

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• GitHub Issue Templates: Added standardized bug report, feature request, and config/proxy issue templates (#641)

• Clear All Models: Added a "Clear All Models" button to the provider detail page with i18n support in 29 languages (#634)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Locale Conflict (in.json): Renamed the Hindi locale file from in.json (Indonesian ISO code) to hi.json to fix translation conflicts in Weblate (#642)

• Codex Empty Tool Names: Moved tool name sanitization before the native Codex passthrough, fixing 400 errors from upstream providers when tools had empty names (#637)

• Streaming Newline Artifacts: Added collapseExcessiveNewlines to the response sanitizer, collapsing runs of 3+ consecutive newlines from thinking models into a standard double newline (#638)

• Claude Reasoning Effort: Converted OpenAI reasoning_effort param to Claude's native thinking budget block across all request paths, including automatic max_tokens adjustment (#627)

• Qwen Token Refresh: Implemented proactive pre-expiry OAuth token refreshes (5-minute buffer) to prevent requests from failing when using short-lived tokens (#631)

🧪 Tests

• 936 tests, 0 failures (+10 tests since 3.0.9)

---

[3.0.9] — 2026-03-26

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• NaN tokens in Claude Code / client responses (#617):

  • sanitizeUsage() now cross-maps input_tokens→prompt_tokens and output_tokens→completion_tokens before the whitelist filter, fixing responses showing NaN/0 token counts when providers return Claude-style usage field names

🔒 Security

• Updated yaml package to fix stack overflow vulnerability (GHSA-48c2-rrv3-qjmp)

📋 Issue Triage

• Closed #613 (Codestral — resolved with Custom Provider workaround)
• Commented on #615 (OpenCode dual-endpoint — workaround provided, tracked as feature request)
• Commented on #618 (tool call visibility — requesting v3.0.9 test)
• Commented on #627 (effort level — already supported)

---

[3.0.8] — 2026-03-25

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Translation Failures for OpenAI-format Providers in Claude CLI (#632):

  • Handle reasoning_details[] array format from StepFun/OpenRouter — converts to reasoning_content
  • Handle reasoning field alias from some providers → normalized to reasoning_content
  • Cross-map usage field names: input_tokens↔prompt_tokens, output_tokens↔completion_tokens in filterUsageForFormat
  • Fix extractUsage to accept both input_tokens/output_tokens and prompt_tokens/completion_tokens as valid usage fields
  • Applied to both streaming (sanitizeStreamingChunk, openai-to-claude.ts translator) and non-streaming (sanitizeMessage) paths

---

[3.0.7] — 2026-03-25

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Antigravity Token Refresh: Fixed client_secret is missing error for npm-installed users — the clientSecretDefault was empty in providerRegistry, causing Google to reject token refresh requests (#588)

• OpenCode Zen Models: Added modelsUrl to the OpenCode Zen registry entry so "Import from /models" works correctly (#612)

• Streaming Artifacts: Fixed excessive newlines left in responses after thinking-tag signature stripping (#626)

• Proxy Fallback: Added automatic retry without proxy when SOCKS5 relay fails

• Proxy Test: Test endpoint now resolves real credentials from DB via proxyId

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Playground Account/Key Selector: Persistent, always-visible dropdown to select specific provider accounts/keys for testing — fetches all connections at startup and filters by selected provider

• CLI Tools Dynamic Models: Model selection now dynamically fetches from /v1/models API — providers like Kiro now show their full model catalog

• Antigravity Model List: Updated with Claude Sonnet 4.5, Claude Sonnet 4, GPT 5, GPT 5 Mini; enabled passthroughModels for dynamic model access (#628)

🔧 Maintenance

• Merged PR #625 — Provider Limits light mode background fix

---

[3.0.6] — 2026-03-25

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Limits/Proxy: Fixed Codex limit fetching for accounts behind SOCKS5 proxies — token refresh now runs inside proxy context

• CI: Fixed integration test v1/models assertion failure in CI environments without provider connections

• Settings: Proxy test button now shows success/failure results immediately (previously hidden behind health data)

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Playground: Added Account selector dropdown — test specific connections individually when a provider has multiple accounts

🔧 Maintenance

• Merged PR #623 — LongCat API base URL path correction

---

[3.0.5] — 2026-03-25

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Limits UI: Added tag grouping feature to the connections dashboard to improve visual organization for accounts with custom tags.

---

[3.0.4] — 2026-03-25

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Streaming: Fixed TextDecoder state corruption inside combo sanitize TransformStream which caused SSE garbled output matching multibyte characters (PR #614)

• Providers UI: Safely render HTML tags inside provider connection error tooltips using dangerouslySetInnerHTML

• Proxy Settings: Added missing username and password payload body properties allowing authenticated proxies to be successfully verified from the Dashboard.

• Provider API: Bound soft exception returns to getCodexUsage preventing API HTTP 500 failures when token fetch fails

---

[3.0.3] — 2026-03-25

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Auto-Sync Models: Added a UI toggle and sync-models endpoint to automatically synchronise model lists per provider using a scheduled interval scheduler (PR #597)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Timeouts: Elevated default proxies FETCH_TIMEOUT_MS and STREAM_IDLE_TIMEOUT_MS to 10 minutes to properly support deep reasoning models (like o1) without aborting requests (Fixes #609)

• CLI Tool Detection: Improved cross-platform detection handling NVM paths, Windows PATHEXT (preventing .cmd wrappers issue), and custom NPM prefixes (PR #598)

• Streaming Logs: Implemented tool_calls delta accumulation in streaming response logs so function calls are tracked and persisted accurately in DB (PR #603)

• Model Catalog: Removed auth exemption, properly hiding comfyui and sdwebui models when no provider is explicitly configured (PR #599)

🌐 Translations

• cs: Improved Czech translation strings across the app (PR #601)

[3.0.2] — 2026-03-25

🚀 Enhancements & Features

feat(ui): Connection Tag Grouping

• Added a Tag/Group field to EditConnectionModal (stored in providerSpecificData.tag) without requiring DB schema migrations.
• Connections in the provider view now dynamically group by tag with visual dividers.
• Untagged connections appear first without a header, followed by tagged groups in alphabetical order.
• The tag grouping automatically applies to the Codex/Copilot/Antigravity Limits section since toggles exist inside connection rows.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.
• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.
• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.
• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.
• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.
• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.
• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.
• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.
• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

fix(ui): Proxy Management UI Stabilization

• Missing badges on connection cards: Fixed by using resolveProxyForConnection() rather than static mapping.
• Test Connection disabled in saved mode: Enabled the Test button by resolving proxy config from the saved list.
• Config Modal freezing: Added onClose() calls after save/clear to prevent the UI from freezing.
• Double usage counting: ProxyRegistryManager now loads usage eagerly on mount with deduplication by scope + scopeId. Usage counts were replaced with a Test button displaying IP/latency inline.

fix(translator): function_call prefix stripping

• Repaired an incomplete fix from PR #607 where only tool_use blocks stripped Claude's proxy_ tool prefix. Now, clients using the OpenAI Responses API format will also correctly receive tool tools without the proxy_ prefix.

---

[3.0.1] — 2026-03-25

🔧 Hotfix Patch — Critical Bug Fixes

Three critical regressions reported by users after the v3.0.0 launch have been resolved.

fix(translator): strip proxy_ prefix in non-streaming Claude responses (#605)

The proxy_ prefix added by Claude OAuth was only stripped from streaming responses. In non-streaming mode, translateNonStreamingResponse had no access to the toolNameMap, causing clients to receive mangled tool names like proxy_read_file instead of read_file.

Fix: Added optional toolNameMap parameter to translateNonStreamingResponse and applied prefix stripping in the Claude tool_use block handler. chatCore.ts now passes the map through.

fix(validation): add LongCat specialty validator to skip /models probe (#592)

LongCat AI does not expose GET /v1/models. The generic validateOpenAICompatibleProvider validator fell through to a chat-completions fallback only if validationModelId was set, which LongCat doesn't configure. This caused provider validation to fail with a misleading error on add/save.

Fix: Added longcat to the specialty validators map, probing /chat/completions directly and treating any non-auth response as a pass.

fix(translator): normalize object tool schemas for Anthropic (#595)

MCP tools (e.g. pencil, computer_use) forward tool definitions with {type:"object"} but without a properties field. Anthropic's API rejects these with: object schema missing properties.

Fix: In openai-to-claude.ts, inject properties: {} as a safe default when type is "object" and properties is absent.

---

🔀 Community PRs Merged (2)

PR	Author	Summary
#589	@flobo3	docs(i18n): fix Russian translation for Playground and Testbed
#591	@rdself	fix(ui): improve Provider Limits light mode contrast and plan tier display

---

✅ Issues Resolved

#592 #595 #605

---

🧪 Tests

• 926 tests, 0 failures (unchanged from v3.0.0)

---

[3.0.0] — 2026-03-24

🎉 OmniRoute v3.0.0 — The Free AI Gateway, Now with 67+ Providers

The biggest release ever. From 36 providers in v2.9.5 to 67+ providers in v3.0.0 — with MCP Server, A2A Protocol, auto-combo engine, Provider Icons, Registered Keys API, 926 tests, and contributions from 12 community members across 10 merged PRs.

Consolidated from v3.0.0-rc.1 through rc.17 (17 release candidates over 3 days of intense development).

---

🆕 New Providers (+31 since v2.9.5)

Provider	Alias	Tier	Notes
OpenCode Zen	opencode-zen	Free	3 models via opencode.ai/zen/v1 (PR #530 by @kang-heewon)
OpenCode Go	opencode-go	Paid	4 models via opencode.ai/zen/go/v1 (PR #530 by @kang-heewon)
LongCat AI	lc	Free	50M tokens/day (Flash-Lite) + 500K/day (Chat/Thinking) during public beta
Pollinations AI	pol	Free	No API key needed — GPT-5, Claude, Gemini, DeepSeek V3, Llama 4 (1 req/15s)
Cloudflare Workers AI	cf	Free	10K Neurons/day — ~150 LLM responses or 500s Whisper audio, edge inference
Scaleway AI	scw	Free	1M free tokens for new accounts — EU/GDPR compliant (Paris)
AI/ML API	aiml	Free	$0.025/day free credits — 200+ models via single endpoint
Puter AI	pu	Free	500+ models (GPT-5, Claude Opus 4, Gemini 3 Pro, Grok 4, DeepSeek V3)
Alibaba Cloud (DashScope)	ali	Paid	International + China endpoints via alicode/alicode-intl
Alibaba Coding Plan	bcp	Paid	Alibaba Model Studio with Anthropic-compatible API
Kimi Coding (API Key)	kmca	Paid	Dedicated API-key-based Kimi access (separate from OAuth)
MiniMax Coding	minimax	Paid	International endpoint
MiniMax (China)	minimax-cn	Paid	China-specific endpoint
Z.AI (GLM-5)	zai	Paid	Zhipu AI next-gen GLM models
Vertex AI	vertex	Paid	Google Cloud — Service Account JSON or OAuth access_token
Ollama Cloud	ollamacloud	Paid	Ollama's hosted API service
Synthetic	synthetic	Paid	Passthrough models gateway
Kilo Gateway	kg	Paid	Passthrough models gateway
Perplexity Search	pplx-search	Paid	Dedicated search-grounded endpoint
Serper Search	serper-search	Paid	Web search API integration
Brave Search	brave-search	Paid	Brave Search API integration
Exa Search	exa-search	Paid	Neural search API integration
Tavily Search	tavily-search	Paid	AI search API integration
NanoBanana	nb	Paid	Image generation API
ElevenLabs	el	Paid	Text-to-speech voice synthesis
Cartesia	cartesia	Paid	Ultra-fast TTS voice synthesis
PlayHT	playht	Paid	Voice cloning and TTS
Inworld	inworld	Paid	AI character voice chat
SD WebUI	sdwebui	Self-hosted	Stable Diffusion local image generation
ComfyUI	comfyui	Self-hosted	ComfyUI local workflow node-based generation
GLM Coding	glm	Paid	BigModel/Zhipu coding-specific endpoint

Total: 67+ providers (4 Free, 8 OAuth, 55 API Key) + unlimited OpenAI/Anthropic-Compatible custom providers.

---

✨ Major Features

🔑 Registered Keys Provisioning API (#464)

Auto-generate and issue OmniRoute API keys programmatically with per-provider and per-account quota enforcement.

Endpoint	Method	Description
/api/v1/registered-keys	POST	Issue a new key — raw key returned once only
/api/v1/registered-keys	GET	List registered keys (masked)
/api/v1/registered-keys/{id}	GET/DELETE	Get metadata / Revoke
/api/v1/quotas/check	GET	Pre-validate quota before issuing
/api/v1/providers/{id}/limits	GET/PUT	Configure per-provider issuance limits
/api/v1/accounts/{id}/limits	GET/PUT	Configure per-account issuance limits
/api/v1/issues/report	POST	Report quota events to GitHub Issues

Security: Keys stored as SHA-256 hashes. Raw key shown once on creation, never retrievable again.

🎨 Provider Icons via @lobehub/icons (#529)

130+ provider logos using @lobehub/icons React components (SVG). Fallback chain: Lobehub SVG → existing PNG → generic icon. Applied across Dashboard, Providers, and Agents pages with standardized ProviderIcon component.

🔄 Model Auto-Sync Scheduler (#488)

Auto-refreshes model lists for connected providers every 24 hours. Runs on server startup. Configurable via MODEL_SYNC_INTERVAL_HOURS.

🔀 Per-Model Combo Routing (#563)

Map model name patterns (glob) to specific combos for automatic routing:

• claude-sonnet* → code-combo, gpt-4o* → openai-combo, gemini-* → google-combo
• New model_combo_mappings table with glob-to-regex matching
• Dashboard UI section: "Model Routing Rules" with inline add/edit/toggle/delete

🧭 API Endpoints Dashboard

Interactive catalog, webhooks management, OpenAPI viewer — all in one tabbed page at /dashboard/endpoint.

🔍 Web Search Providers

5 new search provider integrations: Perplexity Search, Serper, Brave Search, Exa, Tavily — enabling grounded AI responses with real-time web data.

📊 Search Analytics

New tab in /dashboard/analytics — provider breakdown, cache hit rate, cost tracking. API: GET /api/v1/search/analytics.

🛡️ Per-API-Key Rate Limits (#452)

max_requests_per_day and max_requests_per_minute columns with in-memory sliding-window enforcement returning HTTP 429.

🎵 Media Playground

Full media generation playground at /dashboard/media: Image Generation, Video, Music, Audio Transcription (2GB upload limit), and Text-to-Speech.

---

🔒 Security & CI/CD

• CodeQL remediation — Fixed 10+ alerts: 6 polynomial-redos, 1 insecure-randomness (Math.random() → crypto.randomUUID()), 1 shell-command-injection
• Route validation — Zod schemas + validateBody() on 176/176 API routes — CI enforced
• CVE fix — dompurify XSS vulnerability (GHSA-v2wj-7wpq-c8vv) resolved via npm overrides
• Flatted — Bumped 3.3.3 → 3.4.2 (CWE-1321 prototype pollution)
• Docker — Upgraded docker/setup-buildx-action v3 → v4

---

🐛 Bug Fixes (40+)

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.
• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.
• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.
• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.
• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.
• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.
• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.
• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.
• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

OAuth & Auth

• #537 — Gemini CLI OAuth: clear actionable error when GEMINI_OAUTH_CLIENT_SECRET missing in Docker
• #549 — CLI settings routes now resolve real API key from keyId (not masked strings)
• #574 — Login no longer freezes after skipping wizard password setup
• #506 — Cross-platform machineId rewritten (Windows REG.exe → macOS ioreg → Linux → hostname fallback)

Providers & Routing

• #536 — LongCat AI: fixed baseUrl and authHeader
• #535 — Pinned model override: body.model correctly set to pinnedModel
• #570 — Unprefixed Claude models now resolve to Anthropic provider
• #585 — <omniModel> internal tags no longer leak to clients in SSE streaming
• #493 — Custom provider model naming no longer mangled by prefix stripping
• #490 — Streaming + context cache protection via TransformStream injection
• #511 — <omniModel> tag injected into first content chunk (not after [DONE])

CLI & Tools

• #527 — Claude Code + Codex loop: tool_result blocks now converted to text
• #524 — OpenCode config saved correctly (XDG_CONFIG_HOME, TOML format)
• #522 — API Manager: removed misleading "Copy masked key" button
• #546 — --version returning unknown on Windows (PR by @k0valik)
• #544 — Secure CLI tool detection via known installation paths (PR by @k0valik)
• #510 — Windows MSYS2/Git-Bash paths normalized automatically
• #492 — CLI detects mise/nvm-managed Node when app/server.js missing

Streaming & SSE

• PR #587 — Revert resolveDataDir import in responsesTransformer for Cloudflare Workers compat (@k0valik)
• PR #495 — Bottleneck 429 infinite wait: drop waiting jobs on rate limit (@xandr0s)
• #483 — Stop trailing data: null after [DONE] signal
• #473 — Zombie SSE streams: timeout reduced 300s → 120s for faster fallback

Media & Transcription

• Transcription — Deepgram video/mp4 → audio/mp4 MIME mapping, auto language detection, punctuation
• TTS — [object Object] error display fixed for ElevenLabs-style nested errors
• Upload limits — Media transcription increased to 2GB (nginx client_max_body_size 2g + maxDuration=300)

---

🔧 Infrastructure & Improvements

Sub2api Gap Analysis (T01–T15 + T23–T42)

• T01 — requested_model column in call logs (migration 009)
• T02 — Strip empty text blocks from nested tool_result.content
• T03 — Parse x-codex-5h-* / x-codex-7d-* quota headers
• T04 — X-Session-Id header for external sticky routing
• T05 — Rate-limit DB persistence with dedicated API
• T06 — Account deactivated → permanent block (1-year cooldown)
• T07 — X-Forwarded-For IP validation (extractClientIp())
• T08 — Per-API-key session limits with sliding-window enforcement
• T09 — Codex vs Spark rate-limit scopes (separate pools)
• T10 — Credits exhausted → distinct 1h cooldown fallback
• T11 — max reasoning effort → 131072 budget tokens
• T12 — MiniMax M2.7 pricing entries
• T13 — Stale quota display fix (reset window awareness)
• T14 — Proxy fast-fail TCP check (≤2s, cached 30s)
• T15 — Array content normalization for Anthropic
• T23 — Intelligent quota reset fallback (header extraction)
• T24 — 503 cooldown + 406 mapping
• T25 — Provider validation fallback
• T29 — Vertex AI Service Account JWT auth
• T33 — Thinking level to budget conversion
• T36 — 403 vs 429 error classification
• T38 — Centralized model specifications (modelSpecs.ts)
• T39 — Endpoint fallback for fetchAvailableModels
• T41 — Background task auto-redirect to flash models
• T42 — Image generation aspect ratio mapping

Other Improvements

• Per-model upstream custom headers — via configuration UI (PR #575 by @zhangqiang8vip)
• Model context length — configurable in model metadata (PR #578 by @hijak)
• Model prefix stripping — option to remove provider prefix from model names (PR #582 by @jay77721)
• Gemini CLI deprecation — marked deprecated with Google OAuth restriction warning
• YAML parser — replaced custom parser with js-yaml for correct OpenAPI spec parsing
• ZWS v5 — HMR leak fix (485 DB connections → 1, memory 2.4GB → 195MB)
• Log export — New JSON export button on dashboard with time range dropdown
• Update notification banner — dashboard homepage shows when new versions are available

---

🌐 i18n & Documentation

• 30 languages at 100% parity — 2,788 missing keys synced
• Czech — Full translation: 22 docs, 2,606 UI strings (PR by @zen0bit)
• Chinese (zh-CN) — Complete retranslation (PR by @only4copilot)
• VM Deployment Guide — Translated to English as source document
• API Reference — Added /v1/embeddings and /v1/audio/speech endpoints
• Provider count — Updated from 36+/40+/44+ to 67+ across README and all 30 i18n READMEs

---

🔀 Community PRs Merged (10)

PR	Author	Summary
#587	@k0valik	fix(sse): revert resolveDataDir import for Cloudflare Workers compat
#582	@jay77721	feat(proxy): model name prefix stripping option
#581	@jay77721	fix(npm): link electron-release to npm-publish workflow
#578	@hijak	feat: configurable context length in model metadata
#575	@zhangqiang8vip	feat: per-model upstream headers, compat PATCH, chat alignment
#562	@coobabm	fix: MCP session management, Claude passthrough, detectFormat
#561	@zen0bit	fix(i18n): Czech translation corrections
#555	@k0valik	fix(sse): centralized resolveDataDir() for path resolution
#546	@k0valik	fix(cli): --version returning unknown on Windows
#544	@k0valik	fix(cli): secure CLI tool detection via installation paths
#542	@rdself	fix(ui): light mode contrast CSS theme variables
#530	@kang-heewon	feat: OpenCode Zen + Go providers with OpencodeExecutor
#512	@zhangqiang8vip	feat: per-protocol model compatibility (compatByProtocol)
#497	@zhangqiang8vip	fix: dev-mode HMR resource leaks (ZWS v5)
#495	@xandr0s	fix: Bottleneck 429 infinite wait (drop waiting jobs)
#494	@zhangqiang8vip	feat: MiniMax developer→system role fix
#480	@prakersh	fix: stream flush usage extraction
#479	@prakersh	feat: Codex 5.3/5.4 and Anthropic pricing entries
#475	@only4copilot	feat(i18n): improved Chinese translation

Thank you to all contributors! 🙏

---

📋 Issues Resolved (50+)

#452 #458 #462 #464 #466 #473 #474 #481 #483 #487 #488 #489 #490 #491 #492 #493 #506 #508 #509 #510 #511 #513 #520 #521 #522 #524 #525 #527 #529 #531 #532 #535 #536 #537 #541 #546 #549 #563 #570 #574 #585

---

🧪 Tests

• 926 tests, 0 failures (up from 821 in v2.9.5)
• +105 new tests covering: model-combo mappings, registered keys, OpencodeExecutor, Bailian provider, route validation, error classification, aspect ratio mapping, and more

---

📦 Database Migrations

Migration	Description
008	registered_keys, provider_key_limits, account_key_limits tables
009	requested_model column in call_logs
010	model_combo_mappings table for per-model combo routing

---

⬆️ Upgrading from v2.9.5

# npm
npm install -g omniroute@3.0.0

# Docker
docker pull diegosouzapw/omniroute:3.0.0

# Migrations run automatically on first startup

Breaking changes: None. All existing configurations, combos, and API keys are preserved. Database migrations 008-010 run automatically on startup.

---

[3.0.0-rc.17] — 2026-03-24

🔒 Security & CI/CD

• CodeQL remediation — Fixed 10+ alerts:
  • 6 polynomial-redos in provider.ts / chatCore.ts (replaced (?:^|/) alternation patterns with segment-based matching)
  • 1 insecure-randomness in acp/manager.ts (Math.random() → crypto.randomUUID())
  • 1 shell-command-injection in prepublish.mjs (JSON.stringify() path escaping)
• Route validation — Added Zod schemas + validateBody() to 5 routes missing validation:
  • model-combo-mappings (POST, PUT), webhooks (POST, PUT), openapi/try (POST)
  • CI check:route-validation:t06 now passes: 176/176 routes validated

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• #585 — <omniModel> internal tags no longer leak to clients in SSE responses. Added outbound sanitization TransformStream in combo.ts

⚙️ Infrastructure

• Docker — Upgraded docker/setup-buildx-action from v3 → v4 (Node.js 20 deprecation fix)
• CI cleanup — Deleted 150+ failed/cancelled workflow runs

🧪 Tests

• Test suite: 926 tests, 0 failures (+3 new)

---

[3.0.0-rc.16] — 2026-03-24

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Increased media transcription limits

• Added Model Context Length to registry metadata

• Added per-model upstream custom headers via configuration UI

• Fixed multiple bugs, Zod valiadation for patches, and resolved various community issues.

[3.0.0-rc.15] — 2026-03-24

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• #563 — Per-model Combo Routing: map model name patterns (glob) to specific combos for automatic routing

  • New model_combo_mappings table (migration 010) with pattern, combo_id, priority, enabled
  • resolveComboForModel() DB function with glob-to-regex matching (case-insensitive, * and ? wildcards)
  • getComboForModel() in model.ts: augments getCombo() with model-pattern fallback
  • chat.ts: routing decision now checks model-combo mappings before single-model handling
  • API: GET/POST /api/model-combo-mappings, GET/PUT/DELETE /api/model-combo-mappings/:id
  • Dashboard: "Model Routing Rules" section added to Combos page with inline add/edit/toggle/delete
  • Examples: claude-sonnet* → code-combo, gpt-4o* → openai-combo, gemini-* → google-combo

🌐 i18n

• Full i18n Sync: 2,788 missing keys added across 30 language files — all languages now at 100% parity with en.json
• Agents page i18n: OpenCode Integration section fully internationalized (title, description, scanning, download labels)
• 6 new keys added to agents namespace for OpenCode section

🎨 UI/UX

• Provider Icons: 16 missing provider icons added (3 copied, 2 downloaded, 11 SVG created)
• SVG fallback: ProviderIcon component updated with 4-tier strategy: Lobehub → PNG → SVG → Generic icon
• Agents fingerprinting: Synced with CLI tools — added droid, openclaw, copilot, opencode to fingerprint list (14 total)

🔒 Security

• CVE fix: Resolved dompurify XSS vulnerability (GHSA-v2wj-7wpq-c8vv) via npm overrides forcing dompurify@^3.3.2
• npm audit now reports 0 vulnerabilities

🧪 Tests

• Test suite: 923 tests, 0 failures (+15 new model-combo mapping tests)

---

[3.0.0-rc.14] — 2026-03-23

🔀 Community PRs Merged

PR	Author	Summary
#562	@coobabm	fix(ux): MCP session management, Claude passthrough normalization, OAuth modal, detectFormat
#561	@zen0bit	fix(i18n): Czech translation corrections — HTTP method names and documentation updates

🧪 Tests

• Test suite: 908 tests, 0 failures

---

[3.0.0-rc.13] — 2026-03-23

🔧 Bug Fixes

• config: resolve real API key from keyId in CLI settings routes (codex-settings, droid-settings, kilo-settings) to prevent writing masked strings (#549)

---

[3.0.0-rc.12] — 2026-03-23

🔀 Community PRs Merged

PR	Author	Summary
#546	@k0valik	fix(cli): --version returning unknown on Windows — use JSON.parse(readFileSync) instead of ESM import
#555	@k0valik	fix(sse): centralized resolveDataDir() for path resolution in credentials, autoCombo, responses logger, and request logger
#544	@k0valik	fix(cli): secure CLI tool detection via known installation paths (8 tools) with symlink validation, file-type checks, size bounds, minimal env in healthcheck
#542	@rdself	fix(ui): improve light mode contrast — add missing CSS theme variables (bg-primary, bg-subtle, text-primary) and fix dark-only colors in log detail

🔧 Bug Fixes

• TDZ fix in cliRuntime.ts — validateEnvPath was used before initialization at module startup by getExpectedParentPaths(). Reordered declarations to fix ReferenceError.
• Build fixes — Added pino and pino-pretty to serverExternalPackages to prevent Turbopack from breaking Pino's internal worker loading.

🧪 Tests

• Test suite: 905 tests, 0 failures

---

[3.0.0-rc.10] — 2026-03-23

🔧 Bug Fixes

• #509 / #508 — Electron build regression: downgraded Next.js from 16.1.x to 16.0.10 to eliminate Turbopack module-hashing instability that caused blank screens in the Electron desktop bundle.
• Unit test fixes — Corrected two stale test assertions (nanobanana-image-handler aspect ratio/resolution, thinking-budget Gemini thinkingConfig field mapping) that had drifted after recent implementation changes.
• #541 — Responded to user feedback about installation complexity; no code changes required.

---

[3.0.0-rc.9] — 2026-03-23

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• T29 — Vertex AI SA JSON Executor: implemented using the jose library to handle JWT/Service Account auth, along with configurable regions in the UI and automatic partner model URL building.

• T42 — Image generation aspect ratio mapping: created sizeMapper logic for generic OpenAI formats (size), added native imagen3 handling, and updated NanoBanana endpoints to utilize mapped aspect ratios automatically.

• T38 — Centralized model specifications: modelSpecs.ts created for limits and parameters per model.

🔧 Improvements

• T40 — OpenCode CLI tools integration: native opencode-zen and opencode-go integration completed in earlier PR.

---

[3.0.0-rc.8] — 2026-03-23

🔧 Bug Fixes & Improvements (Fallback, Quota & Budget)

• T24 — 503 cooldown await fix + 406 mapping: mapped 406 Not Acceptable to 503 Service Unavailable with proper cooldown intervals.
• T25 — Provider validation fallback: graceful fallback to standard validation models when a specific validationModelId is not present.
• T36 — 403 vs 429 provider handling refinement: extracted into errorClassifier.ts to properly segregate hard permissions failures (403) from rate limits (429).
• T39 — Endpoint Fallback for fetchAvailableModels: implemented a tri-tier mechanism (/models -> /v1/models -> local generic catalog) + list_models_catalog MCP tool updates to reflect source and warning.
• T33 — Thinking level to budget conversion: translates qualitative thinking levels into precise budget allocations.
• T41 — Background task auto redirect: routes heavy background evaluation tasks to flash/efficient models automatically.
• T23 — Intelligent quota reset fallback: accurately extracts x-ratelimit-reset / retry-after header values or maps static cooldowns.

---

[3.0.0-rc.7] — 2026-03-23 (What's New vs v2.9.5 — will be released as v3.0.0)

Upgrade from v2.9.5: 16 issues resolved · 2 community PRs merged · 2 new providers · 7 new API endpoints · 3 new features · DB migration 008+009 · 832 tests passing · 15 sub2api gap improvements (T01–T15 complete).

🆕 New Providers

Provider	Alias	Tier	Notes
OpenCode Zen	opencode-zen	Free	3 models via opencode.ai/zen/v1 (PR #530 by @kang-heewon)
OpenCode Go	opencode-go	Paid	4 models via opencode.ai/zen/go/v1 (PR #530 by @kang-heewon)

Both providers use the new OpencodeExecutor with multi-format routing (/chat/completions, /messages, /responses, /models/{model}:generateContent).

---

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

🔑 Registered Keys Provisioning API (#464)

Auto-generate and issue OmniRoute API keys programmatically with per-provider and per-account quota enforcement.

Endpoint	Method	Description
/api/v1/registered-keys	POST	Issue a new key — raw key returned once only
/api/v1/registered-keys	GET	List registered keys (masked)
/api/v1/registered-keys/{id}	GET	Get key metadata
/api/v1/registered-keys/{id}	DELETE	Revoke a key
/api/v1/registered-keys/{id}/revoke	POST	Revoke (for clients without DELETE support)
/api/v1/quotas/check	GET	Pre-validate quota before issuing
/api/v1/providers/{id}/limits	GET/PUT	Configure per-provider issuance limits
/api/v1/accounts/{id}/limits	GET/PUT	Configure per-account issuance limits
/api/v1/issues/report	POST	Report quota events to GitHub Issues

DB — Migration 008: Three new tables: registered_keys, provider_key_limits, account_key_limits. Security: Keys stored as SHA-256 hashes. Raw key shown once on creation, never retrievable again. Quota types: maxActiveKeys, dailyIssueLimit, hourlyIssueLimit per provider and per account. Idempotency: idempotency_key field prevents duplicate issuance. Returns 409 IDEMPOTENCY_CONFLICT if key was already used. Budget per key: dailyBudget / hourlyBudget — limits how many requests a key can route per window. GitHub reporting: Optional. Set GITHUB_ISSUES_REPO + GITHUB_ISSUES_TOKEN to auto-create GitHub issues on quota exceeded or issuance failures.

🎨 Provider Icons — @lobehub/icons (#529)

All provider icons in the dashboard now use @lobehub/icons React components (130+ providers with SVG). Fallback chain: Lobehub SVG → existing /providers/{id}.png → generic icon. Uses a proper React ErrorBoundary pattern.

🔄 Model Auto-Sync Scheduler (#488)

OmniRoute now automatically refreshes model lists for connected providers every 24 hours.

• Runs on server startup via the existing /api/sync/initialize hook
• Configurable via MODEL_SYNC_INTERVAL_HOURS environment variable
• Covers 16 major providers
• Records last sync time in the settings database

---

🔧 Bug Fixes

OAuth & Auth

• #537 — Gemini CLI OAuth: Clear actionable error when GEMINI_OAUTH_CLIENT_SECRET is missing in Docker/self-hosted deployments. Previously showed cryptic client_secret is missing from Google. Now provides specific docker-compose.yml and ~/.omniroute/.env instructions.

Providers & Routing

• #536 — LongCat AI: Fixed baseUrl (api.longcat.chat/openai) and authHeader (Authorization: Bearer).
• #535 — Pinned model override: body.model is now correctly set to pinnedModel when context-cache protection is active.
• #532 — OpenCode Go key validation: Now uses the zen/v1 test endpoint (testKeyBaseUrl) — same key works for both tiers.

CLI & Tools

• #527 — Claude Code + Codex loop: tool_result blocks are now converted to text instead of dropped, stopping infinite tool-result loops.
• #524 — OpenCode config save: Added saveOpenCodeConfig() handler (XDG_CONFIG_HOME aware, writes TOML).
• #521 — Login stuck: Login no longer freezes after skipping password setup — redirects correctly to onboarding.
• #522 — API Manager: Removed misleading "Copy masked key" button (replaced with a lock icon tooltip).
• #532 — OpenCode Go config: Guide settings handler now handles opencode toolId.

Developer Experience

• #489 — Antigravity: Missing googleProjectId returns a structured 422 error with reconnect guidance instead of a cryptic crash.
• #510 — Windows paths: MSYS2/Git-Bash paths (/c/Program Files/...) are now normalized to C:\Program Files\... automatically.
• #492 — CLI startup: omniroute CLI now detects mise/nvm-managed Node when app/server.js is missing and shows targeted fix instructions.

---

📖 Documentation Updates

• #513 — Docker password reset: INITIAL_PASSWORD env var workaround documented
• #520 — pnpm: pnpm approve-builds better-sqlite3 step documented

---

✅ Issues Resolved in v3.0.0

#464 #488 #489 #492 #510 #513 #520 #521 #522 #524 #527 #529 #532 #535 #536 #537

---

🔀 Community PRs Merged

PR	Author	Summary
#530	@kang-heewon	OpenCode Zen + Go providers with OpencodeExecutor and improved tests

---

[3.0.0-rc.7] - 2026-03-23

🔧 Improvements (sub2api Gap Analysis — T05, T08, T09, T13, T14)

• T05 — Rate-limit DB persistence: setConnectionRateLimitUntil(), isConnectionRateLimited(), getRateLimitedConnections() in providers.ts. The existing rate_limited_until column is now exposed as a dedicated API — OAuth token refresh must NOT touch this field to prevent rate-limit loops.
• T08 — Per-API-key session limit: max_sessions INTEGER DEFAULT 0 added to api_keys via auto-migration. sessionManager.ts gains registerKeySession(), unregisterKeySession(), checkSessionLimit(), and getActiveSessionCountForKey(). Callers in chatCore.js can enforce the limit and decrement on req.close.
• T09 — Codex vs Spark rate-limit scopes: getCodexModelScope() and getCodexRateLimitKey() in codex.ts. Standard models (gpt-5.x-codex, codex-mini) get scope "codex"; spark models (codex-spark*) get scope "spark". Rate-limit keys should be ${accountId}:${scope} so exhausting one pool doesn't block the other.
• T13 — Stale quota display fix: getEffectiveQuotaUsage(used, resetAt) returns 0 when the reset window has passed; formatResetCountdown(resetAt) returns a human-readable countdown string (e.g. "2h 35m"). Both exported from providers.ts + localDb.ts for dashboard consumption.
• T14 — Proxy fast-fail: new src/lib/proxyHealth.ts with isProxyReachable(proxyUrl, timeoutMs=2000) (TCP check, ≤2s instead of 30s timeout), getCachedProxyHealth(), invalidateProxyHealth(), and getAllProxyHealthStatuses(). Results cached 30s by default; configurable via PROXY_FAST_FAIL_TIMEOUT_MS / PROXY_HEALTH_CACHE_TTL_MS.

🧪 Tests

• Test suite: 832 tests, 0 failures

---

[3.0.0-rc.6] - 2026-03-23

🔧 Bug Fixes & Improvements (sub2api Gap Analysis — T01–T15)

• T01 — requested_model column in call_logs (migration 009): track which model the client originally requested vs the actual routed model. Enables fallback rate analytics.
• T02 — Strip empty text blocks from nested tool_result.content: prevents Anthropic 400 errors (text content blocks must be non-empty) when Claude Code chains tool results.
• T03 — Parse x-codex-5h-* / x-codex-7d-* headers: parseCodexQuotaHeaders() + getCodexResetTime() extract Codex quota windows for precise cooldown scheduling instead of generic 5-min fallback.
• T04 — X-Session-Id header for external sticky routing: extractExternalSessionId() in sessionManager.ts reads x-session-id / x-omniroute-session headers with ext: prefix to avoid collision with internal SHA-256 session IDs. Nginx-compatible (hyphenated header).
• T06 — Account deactivated → permanent block: isAccountDeactivated() in accountFallback.ts detects 401 deactivation signals and applies a 1-year cooldown to prevent retrying permanently dead accounts.
• T07 — X-Forwarded-For IP validation: new src/lib/ipUtils.ts with extractClientIp() and getClientIpFromRequest() — skips unknown/non-IP entries in X-Forwarded-For chains (Nginx/proxy-forwarded requests).
• T10 — Credits exhausted → distinct fallback: isCreditsExhausted() in accountFallback.ts returns 1h cooldown with creditsExhausted flag, distinct from generic 429 rate limiting.
• T11 — max reasoning effort → 131072 budget tokens: EFFORT_BUDGETS and THINKING_LEVEL_MAP updated; reverse mapping now returns "max" for full-budget responses. Unit test updated.
• T12 — MiniMax M2.7 pricing entries added: minimax-m2.7, MiniMax-M2.7, minimax-m2.7-highspeed added to pricing table (sub2api PR #1120). M2.5/GLM-4.7/GLM-5/Kimi pricing already existed.
• T15 — Array content normalization: normalizeContentToString() helper in openai-to-claude.ts correctly collapses array-formatted system/tool messages to string before sending to Anthropic.

🧪 Tests

• Test suite: 832 tests, 0 failures (unchanged from rc.5)

---

[3.0.0-rc.5] - 2026-03-22

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• #464 — Registered Keys Provisioning API: auto-issue API keys with per-provider & per-account quota enforcement

  • POST /api/v1/registered-keys — issue keys with idempotency support
  • GET /api/v1/registered-keys — list (masked) registered keys
  • GET /api/v1/registered-keys/{id} — get key metadata
  • DELETE /api/v1/registered-keys/{id} / POST ../{id}/revoke — revoke keys
  • GET /api/v1/quotas/check — pre-validate before issuing
  • PUT /api/v1/providers/{id}/limits — set provider issuance limits
  • PUT /api/v1/accounts/{id}/limits — set account issuance limits
  • POST /api/v1/issues/report — optional GitHub issue reporting
  • DB migration 008: registered_keys, provider_key_limits, account_key_limits tables

---

[3.0.0-rc.4] - 2026-03-22

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• #530 (PR) — OpenCode Zen and OpenCode Go providers added (by @kang-heewon)

  • New OpencodeExecutor with multi-format routing (/chat/completions, /messages, /responses)
  • 7 models across both tiers

---

[3.0.0-rc.3] - 2026-03-22

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• #529 — Provider icons now use @lobehub/icons with graceful PNG fallback and a ProviderIcon component (130+ providers supported)

• #488 — Auto-update model lists every 24h via modelSyncScheduler (configurable via MODEL_SYNC_INTERVAL_HOURS)

🔧 Bug Fixes

• #537 — Gemini CLI OAuth: now shows clear actionable error when GEMINI_OAUTH_CLIENT_SECRET is missing in Docker/self-hosted deployments

---

[3.0.0-rc.2] - 2026-03-22

🔧 Bug Fixes

• #536 — LongCat AI key validation: fixed baseUrl (api.longcat.chat/openai) and authHeader (Authorization: Bearer)
• #535 — Pinned model override: body.model is now set to pinnedModel when context-cache protection detects a pinned model
• #524 — OpenCode config now saved correctly: added saveOpenCodeConfig() handler (XDG_CONFIG_HOME aware, writes TOML)

---

[3.0.0-rc.1] - 2026-03-22

🔧 Bug Fixes

• #521 — Login no longer gets stuck after skipping password setup (redirects to onboarding)
• #522 — API Manager: Removed misleading "Copy masked key" button (replaced with lock icon tooltip)
• #527 — Claude Code + Codex superpowers loop: tool_result blocks now converted to text instead of dropped
• #532 — OpenCode GO API key validation now uses the correct zen/v1 endpoint (testKeyBaseUrl)
• #489 — Antigravity: missing googleProjectId returns structured 422 error with reconnect guidance
• #510 — Windows: MSYS2/Git-Bash paths (/c/Program Files/...) are now normalized to C:\Program Files\...
• #492 — omniroute CLI now detects mise/nvm when app/server.js is missing and shows targeted fix

📖 Documentation

• #513 — Docker password reset: INITIAL_PASSWORD env var workaround documented
• #520 — pnpm: pnpm approve-builds better-sqlite3 documented

✅ Closed Issues

#489, #492, #510, #513, #520, #521, #522, #525, #527, #532

---

[2.9.5] — 2026-03-22

Sprint: New OpenCode providers, embedding credentials fix, CLI masked key bug, CACHE_TAG_PATTERN fix.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• CLI tools save masked API key to config files — claude-settings, cline-settings, and openclaw-settings POST routes now accept a keyId param and resolve the real API key from DB before writing to disk. ClaudeToolCard updated to send keyId instead of the masked display string. Fixes #523, #526.

• Custom embedding providers: No credentials error — /v1/embeddings now tracks credentialsProviderId separately from the routing prefix, so credentials are fetched from the matching provider node ID rather than the public prefix string. Fixes a regression where google/gemini-embedding-001 and similar custom-provider models would always fail with a credentials error. Fixes #532-related. (PR #528 by @jacob2826)

• Context cache protection regex misses prefix — CACHE_TAG_PATTERN in comboAgentMiddleware.ts updated to match both literal (backslash-n) and actual newline U+000A that combo.ts streaming injects around the <omniModel> tag after fix #515. Fixes #531.

✨ New Providers

• OpenCode Zen — Free tier gateway at opencode.ai/zen/v1 with 3 models: minimax-m2.5-free, big-pickle, gpt-5-nano
• OpenCode Go — Subscription service at opencode.ai/zen/go/v1 with 4 models: glm-5, kimi-k2.5, minimax-m2.7 (Claude format), minimax-m2.5 (Claude format)
• Both providers use the new OpencodeExecutor which routes dynamically to /chat/completions, /messages, /responses, or /models/{model}:generateContent based on the requested model. (PR #530 by @kang-heewon)

---

[2.9.4] — 2026-03-21

Sprint: Bug fixes — preserve Codex prompt cache key, fix tagContent JSON escaping, sync expired token status to DB.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(translator): Preserve prompt_cache_key in Responses API → Chat Completions translation (#517) — The field is a cache-affinity signal used by Codex; stripping it was preventing prompt cache hits. Fixed in openai-responses.ts and responsesApiHelper.ts.

• fix(combo): Escape in tagContent so injected JSON string is valid (#515) — Template literal newlines (U+000A) are not allowed unescaped inside JSON string values. Replaced with \n literal sequences in open-sse/services/combo.ts.

• fix(usage): Sync expired token status back to DB on live auth failure (#491) — When the Limits & Quotas live check returns 401/403, the connection testStatus is now updated to "expired" in the database so the Providers page reflects the same degraded state. Fixed in src/app/api/usage/[connectionId]/route.ts.

---

[2.9.3] — 2026-03-21

Sprint: Add 5 new free AI providers — LongCat, Pollinations, Cloudflare AI, Scaleway, AI/ML API.

✨ New Providers

• feat(providers/longcat): Add LongCat AI (lc/) — 50M tokens/day free (Flash-Lite) + 500K/day (Chat/Thinking) during public beta. OpenAI-compatible, standard Bearer auth.
• feat(providers/pollinations): Add Pollinations AI (pol/) — no API key required. Proxies GPT-5, Claude, Gemini, DeepSeek V3, Llama 4 (1 req/15s free). Custom executor handles optional auth.
• feat(providers/cloudflare-ai): Add Cloudflare Workers AI (cf/) — 10K Neurons/day free (~150 LLM responses or 500s Whisper audio). 50+ models on global edge. Custom executor builds dynamic URL with accountId from credentials.
• feat(providers/scaleway): Add Scaleway Generative APIs (scw/) — 1M free tokens for new accounts. EU/GDPR compliant (Paris). Qwen3 235B, Llama 3.1 70B, Mistral Small 3.2.
• feat(providers/aimlapi): Add AI/ML API (aiml/) — $0.025/day free credit, 200+ models (GPT-4o, Claude, Gemini, Llama) via single aggregator endpoint.

🔄 Provider Updates

• feat(providers/together): Add hasFree: true + 3 permanently free model IDs: Llama-3.3-70B-Instruct-Turbo-Free, Llama-Vision-Free, DeepSeek-R1-Distill-Llama-70B-Free
• feat(providers/gemini): Add hasFree: true + freeNote (1,500 req/day, no credit card needed, aistudio.google.com)
• chore(providers/gemini): Rename display name to Gemini (Google AI Studio) for clarity

⚙️ Infrastructure

• feat(executors/pollinations): New PollinationsExecutor — omits Authorization header when no API key provided
• feat(executors/cloudflare-ai): New CloudflareAIExecutor — dynamic URL construction requires accountId in provider credentials
• feat(executors): Register pollinations, pol, cloudflare-ai, cf executor mappings

📝 Documentation

• docs(readme): Expanded free combo stack to 11 providers ($0 forever)
• docs(readme): Added 4 new free provider sections (LongCat, Pollinations, Cloudflare AI, Scaleway) with model tables
• docs(readme): Updated pricing table with 4 new free tier rows
• docs(i18n/pt-BR): Updated pricing table + added LongCat/Pollinations/Cloudflare AI/Scaleway sections in Portuguese
• docs(new-features/ai): 10 task spec files + master implementation plan in docs/new-features/ai/

🧪 Tests

• Test suite: 821 tests, 0 failures (unchanged)

---

[2.9.2] — 2026-03-21

Sprint: Fix media transcription (Deepgram/HuggingFace Content-Type, language detection) and TTS error display.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(transcription): Deepgram and HuggingFace audio transcription now correctly map video/mp4 → audio/mp4 and other media MIME types via new resolveAudioContentType() helper. Previously, uploading .mp4 files consistently returned "No speech detected" because Deepgram was receiving Content-Type: video/mp4.

• fix(transcription): Added detect_language=true to Deepgram requests — auto-detects audio language (Portuguese, Spanish, etc.) instead of defaulting to English. Fixes non-English transcriptions returning empty or garbage results.

• fix(transcription): Added punctuate=true to Deepgram requests for higher-quality transcription output with correct punctuation.

• fix(tts): [object Object] error display in Text-to-Speech responses fixed in both audioSpeech.ts and audioTranscription.ts. The upstreamErrorResponse() function now correctly extracts nested string messages from providers like ElevenLabs that return { error: { message: "...", status_code: 401 } } instead of a flat error string.

🧪 Tests

• Test suite: 821 tests, 0 failures (unchanged)

Triaged Issues

• #508 — Tool call format regression: requested proxy logs and provider chain info (needs-info)
• #510 — Windows CLI healthcheck path: requested shell/Node version info (needs-info)
• #485 — Kiro MCP tool calls: closed as external Kiro issue (not OmniRoute)
• #442 — Baseten /models endpoint: closed (documented manual workaround)
• #464 — Key provisioning API: acknowledged as roadmap item

---

[2.9.1] — 2026-03-21

Sprint: Fix SSE omniModel data loss, merge per-protocol model compatibility.

Bug Fixes

• #511 — Critical: <omniModel> tag was sent after finish_reason:stop in SSE streams, causing data loss. Tag is now injected into the first non-empty content chunk, guaranteeing delivery before SDKs close the connection.

Merged PRs

• PR #512 (@zhangqiang8vip): Per-protocol model compatibility — normalizeToolCallId and preserveOpenAIDeveloperRole can now be configured per client protocol (OpenAI, Claude, Responses API). New compatByProtocol field in model config with Zod validation.

Triaged Issues

• #510 — Windows CLI healthcheck_failed: requested PATH/version info
• #509 — Turbopack Electron regression: upstream Next.js bug, documented workarounds
• #508 — macOS black screen: suggested --disable-gpu workaround

---

[2.9.0] — 2026-03-20

Sprint: Cross-platform machineId fix, per-API-key rate limits, streaming context cache, Alibaba DashScope, search analytics, ZWS v5, and 8 issues closed.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(search): Search Analytics tab in /dashboard/analytics — provider breakdown, cache hit rate, cost tracking. New API: GET /api/v1/search/analytics (#feat/search-provider-routing)

• feat(provider): Alibaba Cloud DashScope added with custom endpoint path validation — configurable chatPath and modelsPath per node (#feat/custom-endpoint-paths)

• feat(api): Per-API-key request-count limits — max_requests_per_day and max_requests_per_minute columns with in-memory sliding-window enforcement returning HTTP 429 (#452)

• feat(dev): ZWS v5 — HMR leak fix (485 DB connections → 1), memory 2.4GB → 195MB, globalThis singletons, Edge Runtime warning fix (@zhangqiang8vip)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(#506): Cross-platform machineId — getMachineIdRaw() rewritten with try/catch waterfall (Windows REG.exe → macOS ioreg → Linux file read → hostname → os.hostname()). Eliminates process.platform branching that Next.js bundler dead-code-eliminated, fixing 'head' is not recognized on Windows. Also fixes #466.

• fix(#493): Custom provider model naming — removed incorrect prefix stripping in DefaultExecutor.transformRequest() that mangled org-scoped model IDs like zai-org/GLM-5-FP8.

• fix(#490): Streaming + context cache protection — TransformStream intercepts SSE to inject <omniModel> tag before [DONE] marker, enabling context cache protection for streaming responses.

• fix(#458): Combo schema validation — system_message, tool_filter_regex, context_cache_protection fields now pass Zod validation on save.

• fix(#487): KIRO MITM card cleanup — removed ZWS_README, generified AntigravityToolCard to use dynamic tool metadata.

🧪 Tests

• Added Anthropic-format tools filter unit tests (PR #397) — 8 regression tests for tool.name without .function wrapper
• Test suite: 821 tests, 0 failures (up from 813)

📋 Issues Closed (8)

• #506 — Windows machineId head not recognized (fixed)
• #493 — Custom provider model naming (fixed)
• #490 — Streaming context cache (fixed)
• #452 — Per-API-key request limits (implemented)
• #466 — Windows login failure (same root cause as #506)
• #504 — MITM inactive (expected behavior)
• #462 — Gemini CLI PSA (resolved)
• #434 — Electron app crash (duplicate of #402)

[2.8.9] — 2026-03-20

Sprint: Merge community PRs, fix KIRO MITM card, dependency updates.

Merged PRs

• PR #498 (@Sajid11194): Fix Windows machine ID crash (undefined\REG.exe). Replaces node-machine-id with native OS registry queries. Closes #486.
• PR #497 (@zhangqiang8vip): Fix dev-mode HMR resource leaks — 485 leaked DB connections → 1, memory 2.4GB → 195MB. globalThis singletons, Edge Runtime warning fix, Windows test stability. (+1168/-338 across 22 files)
• PRs #499-503 (Dependabot): GitHub Actions updates — docker/build-push-action@7, actions/checkout@6, peter-evans/dockerhub-description@5, docker/setup-qemu-action@4, docker/login-action@4.

Bug Fixes

• #505 — KIRO MITM card now displays tool-specific instructions (api.anthropic.com) instead of Antigravity-specific text.
• #504 — Responded with UX clarification (MITM "Inactive" is expected behavior when proxy is not running).

---

[2.8.8] — 2026-03-20

Sprint: Fix OAuth batch test crash, add "Test All" button to individual provider pages.

Bug Fixes

• OAuth batch test crash (ERR_CONNECTION_REFUSED): Replaced sequential for-loop with 5-connection concurrency limit + 30s per-connection timeout via Promise.race() + Promise.allSettled(). Prevents server crash when testing large OAuth provider groups (~30+ connections).

Features

• "Test All" button on provider pages: Individual provider pages (e.g., /providers/codex) now show a "Test All" button in the Connections header when there are 2+ connections. Uses POST /api/providers/test-batch with {mode: "provider", providerId}. Results displayed in a modal with pass/fail summary and per-connection diagnosis.

---

[2.8.7] — 2026-03-20

Sprint: Merge PR #495 (Bottleneck 429 drop), fix #496 (custom embedding providers), triage features.

Bug Fixes

• Bottleneck 429 infinite wait (PR #495 by @xandr0s): On 429, limiter.stop({ dropWaitingJobs: true }) immediately fails all queued requests so upstream callers can trigger fallback. Limiter is deleted from Map so next request creates a fresh instance.
• Custom embedding models unresolvable (#496): POST /v1/embeddings now resolves custom embedding models from ALL provider_nodes (not just localhost). Enables models like google/gemini-embedding-001 added via dashboard.

Issues Responded

• #452 — Per-API-key request-count limits (acknowledged, on roadmap)
• #464 — Auto-issue API keys with provider/account limits (needs more detail)
• #488 — Auto-update model lists (acknowledged, on roadmap)
• #496 — Custom embedding provider resolution (fixed)

---

[2.8.6] — 2026-03-20

Sprint: Merge PR #494 (MiniMax role fix), fix KIRO MITM dashboard, triage 8 issues.

Features

• MiniMax developer→system role fix (PR #494 by @zhangqiang8vip): Per-model preserveDeveloperRole toggle. Adds "Compatibility" UI in providers page. Fixes 422 "role param error" for MiniMax and similar gateways.
• roleNormalizer: normalizeDeveloperRole() now accepts preserveDeveloperRole parameter with tri-state behavior (undefined=keep, true=keep, false=convert).
• DB: New getModelPreserveOpenAIDeveloperRole() and mergeModelCompatOverride() in models.ts.

Bug Fixes

• KIRO MITM dashboard (#481/#487): CLIToolsPageClient now routes any configType: "mitm" tool to AntigravityToolCard (MITM Start/Stop controls). Previously only Antigravity was hardcoded.
• AntigravityToolCard generic: Uses tool.image, tool.description, tool.id instead of hardcoded Antigravity values. Guards against missing defaultModels.

Cleanup

• Removed ZWS_README_V2.md (development-only docs from PR #494).

Issues Triaged (8)

• #487 — Closed (KIRO MITM fixed in this release)
• #486 — needs-info (Windows REG.exe PATH issue)
• #489 — needs-info (Antigravity projectId missing, OAuth reconnect needed)
• #492 — needs-info (missing app/server.js on mise-managed Node)
• #490 — Acknowledged (streaming + context cache blocking, fix planned)
• #491 — Acknowledged (Codex auth state inconsistency)
• #493 — Acknowledged (Modal provider model name prefix, workaround provided)
• #488 — Feature request backlog (auto-update model lists)

---

[2.8.5] — 2026-03-19

Sprint: Fix zombie SSE streams, context cache first-turn, KIRO MITM, and triage 5 external issues.

Bug Fixes

• Zombie SSE Streams (#473): Reduce STREAM_IDLE_TIMEOUT_MS from 300s → 120s for faster combo fallback when providers hang mid-stream. Configurable via env var.
• Context Cache Tag (#474): Fix injectModelTag() to handle first-turn requests (no assistant messages) — context cache protection now works from the very first response.
• KIRO MITM (#481): Change KIRO configType from guide → mitm so the dashboard renders MITM Start/Stop controls.
• E2E Test (CI): Fix providers-bailian-coding-plan.spec.ts — dismiss pre-existing modal overlay before clicking Add API Key button.

Closed Issues

• #473 — Zombie SSE streams bypass combo fallback
• #474 — Context cache <omniModel> tag missing on first turn
• #481 — MITM for KIRO not activatable from dashboard
• #468 — Gemini CLI remote server (superseded by #462 deprecation)
• #438 — Claude unable to write files (external CLI issue)
• #439 — AppImage doesn't work (documented libfuse2 workaround)
• #402 — ARM64 DMG "damaged" (documented xattr -cr workaround)
• #460 — CLI not runnable on Windows (documented PATH fix)

---

[2.8.4] — 2026-03-19

Sprint: Gemini CLI deprecation, VM guide i18n fix, dependabot security fix, provider schema expansion.

Features

• Gemini CLI Deprecation (#462): Mark gemini-cli provider as deprecated with warning — Google restricts third-party OAuth usage from March 2026
• Provider Schema (#462): Expand Zod validation with deprecated, deprecationReason, hasFree, freeNote, authHint, apiHint optional fields

Bug Fixes

• VM Guide i18n (#471): Add VM_DEPLOYMENT_GUIDE.md to i18n translation pipeline, regenerate all 30 locale translations from English source (were stuck in Portuguese)

Security

• deps: Bump flatted 3.3.3 → 3.4.2 — fixes CWE-1321 prototype pollution (#484, @dependabot)

Closed Issues

• #472 — Model Aliases regression (fixed in v2.8.2)
• #471 — VM guide translations broken
• #483 — Trailing data: null after [DONE] (fixed in v2.8.3)

Merged PRs

• #484 — deps: bump flatted from 3.3.3 to 3.4.2 (@dependabot)

---

[2.8.3] — 2026-03-19

Sprint: Czech i18n, SSE protocol fix, VM guide translation.

Features

• Czech Language (#482): Full Czech (cs) i18n — 22 docs, 2606 UI strings, language switcher updates (@zen0bit)
• VM Deployment Guide: Translated from Portuguese to English as the source document (@zen0bit)

Bug Fixes

• SSE Protocol (#483): Stop sending trailing data: null after [DONE] signal — fixes AI_TypeValidationError in strict AI SDK clients (Zod-based validators)

Merged PRs

• #482 — Add Czech language + Fix VM_DEPLOYMENT_GUIDE.md English source (@zen0bit)

---

[2.8.2] — 2026-03-19

Sprint: 2 merged PRs, model aliases routing fix, log export, and issue triage.

Features

• Log Export: New Export button on /dashboard/logs with time range dropdown (1h, 6h, 12h, 24h). Downloads JSON of request/proxy/call logs via /api/logs/export API (#user-request)

Bug Fixes

• Model Aliases Routing (#472): Settings → Model Aliases now correctly affect provider routing, not just format detection. Previously resolveModelAlias() output was only used for getModelTargetFormat() but the original model ID was sent to the provider
• Stream Flush Usage (#480): Usage data from the last SSE event in the buffer is now correctly extracted during stream flush (merged from @prakersh)

Merged PRs

• #480 — Extract usage from remaining buffer in flush handler (@prakersh)
• #479 — Add missing Codex 5.3/5.4 and Anthropic model ID pricing entries (@prakersh)

---

[2.8.1] — 2026-03-19

Sprint: Five community PRs — streaming call log fixes, Kiro compatibility, cache token analytics, Chinese translation, and configurable tool call IDs.

✨ Features

• feat(logs): Call log response content now correctly accumulated from raw provider chunks (OpenAI/Claude/Gemini) before translation, fixing empty response payloads in streaming mode (#470, @zhangqiang8vip)
• feat(providers): Per-model configurable 9-char tool call ID normalization (Mistral-style) — only models with the option enabled get truncated IDs (#470)
• feat(api): Key PATCH API expanded to support allowedConnections, name, autoResolve, isActive, and accessSchedule fields (#470)
• feat(dashboard): Response-first layout in request log detail UI (#470)
• feat(i18n): Improved Chinese (zh-CN) translation — complete retranslation (#475, @only4copilot)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(kiro): Strip injected model field from request body — Kiro API rejects unknown top-level fields (#478, @prakersh)

• fix(usage): Include cache read + cache creation tokens in usage history input totals for accurate analytics (#477, @prakersh)

• fix(callLogs): Support Claude format usage fields (input_tokens/output_tokens) alongside OpenAI format, include all cache token variants (#476, @prakersh)

---

[2.8.0] — 2026-03-19

Sprint: Bailian Coding Plan provider with editable base URLs, plus community contributions for Alibaba Cloud and Kimi Coding.

✨ Features

• feat(providers): Added Bailian Coding Plan (bailian-coding-plan) — Alibaba Model Studio with Anthropic-compatible API. Static catalog of 8 models including Qwen3.5 Plus, Qwen3 Coder, MiniMax M2.5, GLM 5, and Kimi K2.5. Includes custom auth validation (400=valid, 401/403=invalid) (#467, @Mind-Dragon)
• feat(admin): Editable default URL in Provider Admin create/edit flows — users can configure custom base URLs per connection. Persisted in providerSpecificData.baseUrl with Zod schema validation rejecting non-http(s) schemes (#467)

🧪 Tests

• Added 30+ unit tests and 2 e2e scenarios for Bailian Coding Plan provider covering auth validation, schema hardening, route-level behavior, and cross-layer integration

---

[2.7.10] — 2026-03-19

Sprint: Two new community-contributed providers (Alibaba Cloud Coding, Kimi Coding API-key) and Docker pino fix.

✨ Features

• feat(providers): Added Alibaba Cloud Coding Plan support with two OpenAI-compatible endpoints — alicode (China) and alicode-intl (International), each with 8 models (#465, @dtk1985)
• feat(providers): Added dedicated kimi-coding-apikey provider path — API-key-based Kimi Coding access is no longer forced through OAuth-only kimi-coding route. Includes registry, constants, models API, config, and validation test (#463, @Mind-Dragon)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(docker): Added missing split2 dependency to Docker image — pino-abstract-transport requires it at runtime but it was not being copied into the standalone container, causing Cannot find module 'split2' crashes (#459)

---

[2.7.9] — 2026-03-18

Sprint: Codex responses subpath passthrough natively supported, Windows MITM crash fixed, and Combos agent schemas adjusted.

✨ Features

• feat(codex): Native responses subpath passthrough for Codex — natively routes POST /v1/responses/compact to Codex upstream, maintaining Claude Code compatibility without stripping the /compact suffix (#457)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(combos): Zod schemas (updateComboSchema and createComboSchema) now include system_message, tool_filter_regex, and context_cache_protection. Fixes bug where agent-specific settings created via the dashboard were silently discarded by the backend validation layer (#458)

• fix(mitm): Kiro MITM profile crash on Windows fixed — node-machine-id failed due to missing REG.exe env, and the fallback threw a fatal crypto is not defined error. Fallback now safely and correctly imports crypto (#456)

---

[2.7.8] — 2026-03-18

Sprint: Budget save bug + combo agent features UI + omniModel tag security fix.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(budget): "Save Limits" no longer returns 422 — warningThreshold is now correctly sent as fraction (0–1) instead of percentage (0–100) (#451)

• fix(combos): <omniModel> internal cache tag is now stripped before forwarding requests to providers, preventing cache session breaks (#454)

✨ Features

• feat(combos): Agent Features section added to combo create/edit modal — expose system_message override, tool_filter_regex, and context_cache_protection directly from the dashboard (#454)

---

[2.7.7] — 2026-03-18

Sprint: Docker pino crash, Codex CLI responses worker fix, package-lock sync.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(docker): pino-abstract-transport and pino-pretty now explicitly copied in Docker runner stage — Next.js standalone trace misses these peer deps, causing Cannot find module pino-abstract-transport crash on startup (#449)

• fix(responses): Remove initTranslators() from /v1/responses route — was crashing Next.js worker with the worker has exited uncaughtException on Codex CLI requests (#450)

🔧 Maintenance

• chore(deps): package-lock.json now committed on every version bump to ensure Docker npm ci uses exact dependency versions

---

[2.7.5] — 2026-03-18

Sprint: UX improvements and Windows CLI healthcheck fix.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(ux): Show default password hint on login page — new users now see "Default password: 123456" below the password input (#437)

• fix(cli): Claude CLI and other npm-installed tools now correctly detected as runnable on Windows — spawn uses shell:true to resolve .cmd wrappers via PATHEXT (#447)

---

[2.7.4] — 2026-03-18

Sprint: Search Tools dashboard, i18n fixes, Copilot limits, Serper validation fix.

🚀 Features

• feat(search): Add Search Playground (10th endpoint), Search Tools page with Compare Providers/Rerank Pipeline/Search History, local rerank routing, auth guards on search API (#443 by @Regis-RCR)
  • New route: /dashboard/search-tools
  • Sidebar entry under Debug section
  • GET /api/search/providers and GET /api/search/stats with auth guards
  • Local provider_nodes routing for /v1/rerank
  • 30+ i18n keys in search namespace

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(search): Fix Brave news normalizer (was returning 0 results), enforce max_results truncation post-normalization, fix Endpoints page fetch URL (#443 by @Regis-RCR)

• fix(analytics): Localize analytics day/date labels — replace hardcoded Portuguese strings with Intl.DateTimeFormat(locale) (#444 by @hijak)

• fix(copilot): Correct GitHub Copilot account type display, filter misleading unlimited quota rows from limits dashboard (#445 by @hijak)

• fix(providers): Stop rejecting valid Serper API keys — treat non-4xx responses as valid authentication (#446 by @hijak)

---

[2.7.3] — 2026-03-18

Sprint: Codex direct API quota fallback fix.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(codex): Block weekly-exhausted accounts in direct API fallback (#440)

  • resolveQuotaWindow() prefix matching: "weekly" now matches "weekly (7d)" cache keys
  • applyCodexWindowPolicy() enforces useWeekly/use5h toggles correctly
  • 4 new regression tests (766 total)

---

[2.7.2] — 2026-03-18

Sprint: Light mode UI contrast fixes.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(logs): Fix light mode contrast in request logs filter buttons and combo badge (#378)

  • Error/Success/Combo filter buttons now readable in light mode
  • Combo row badge uses stronger violet in light mode

---

[2.7.1] — 2026-03-17

Sprint: Unified web search routing (POST /v1/search) with 5 providers + Next.js 16.1.7 security fixes (6 CVEs).

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(search): Unified web search routing — POST /v1/search with 5 providers (Serper, Brave, Perplexity, Exa, Tavily)

  • Auto-failover across providers, 6,500+ free searches/month
  • In-memory cache with request coalescing (configurable TTL)
  • Dashboard: Search Analytics tab in /dashboard/analytics with provider breakdown, cache hit rate, cost tracking
  • New API: GET /api/v1/search/analytics for search request statistics
  • DB migration: request_type column on call_logs for non-chat request tracking
  • Zod validation (v1SearchSchema), auth-gated, cost recorded via recordCost()

🔒 Security

• deps: Next.js 16.1.6 → 16.1.7 — fixes 6 CVEs:
  • Critical: CVE-2026-29057 (HTTP request smuggling via http-proxy)
  • High: CVE-2026-27977, CVE-2026-27978 (WebSocket + Server Actions)
  • Medium: CVE-2026-27979, CVE-2026-27980, CVE-2026-jcc7

📁 New Files

File	Purpose
open-sse/handlers/search.ts	Search handler with 5-provider routing
open-sse/config/searchRegistry.ts	Provider registry (auth, cost, quota, TTL)
open-sse/services/searchCache.ts	In-memory cache with request coalescing
src/app/api/v1/search/route.ts	Next.js route (POST + GET)
src/app/api/v1/search/analytics/route.ts	Search stats API
src/app/(dashboard)/dashboard/analytics/SearchAnalyticsTab.tsx	Analytics dashboard tab
src/lib/db/migrations/007_search_request_type.sql	DB migration
tests/unit/search-registry.test.mjs	277 lines of unit tests

---

[2.7.0] — 2026-03-17

Sprint: ClawRouter-inspired features — toolCalling flag, multilingual intent detection, benchmark-driven fallback, request deduplication, pluggable RouterStrategy, Grok-4 Fast + GLM-5 + MiniMax M2.5 + Kimi K2.5 pricing.

✨ New Models & Pricing

• feat(pricing): xAI Grok-4 Fast — $0.20/$0.50 per 1M tokens, 1143ms p50 latency, tool calling supported
• feat(pricing): xAI Grok-4 (standard) — $0.20/$1.50 per 1M tokens, reasoning flagship
• feat(pricing): GLM-5 via Z.AI — $0.5/1M, 128K output context
• feat(pricing): MiniMax M2.5 — $0.30/1M input, reasoning + agentic tasks
• feat(pricing): DeepSeek V3.2 — updated pricing $0.27/$1.10 per 1M
• feat(pricing): Kimi K2.5 via Moonshot API — direct Moonshot API access
• feat(providers): Z.AI provider added (zai alias) — GLM-5 family with 128K output

🧠 Routing Intelligence

• feat(registry): toolCalling flag per model in provider registry — combos can now prefer/require tool-calling capable models
• feat(scoring): Multilingual intent detection for AutoCombo scoring — PT/ZH/ES/AR script/language patterns influence model selection per request context
• feat(fallback): Benchmark-driven fallback chains — real latency data (p50 from comboMetrics) used to re-order fallback priority dynamically
• feat(dedup): Request deduplication via content-hash — 5-second idempotency window prevents duplicate provider calls from retrying clients
• feat(router): Pluggable RouterStrategy interface in autoCombo/routerStrategy.ts — custom routing logic can be injected without modifying core

🔧 MCP Server Improvements

• feat(mcp): 2 new advanced tool schemas: omniroute_get_provider_metrics (p50/p95/p99 per provider) and omniroute_explain_route (routing decision explanation)
• feat(mcp): MCP tool auth scopes updated — metrics:read scope added for provider metrics tools
• feat(mcp): omniroute_best_combo_for_task now accepts languageHint parameter for multilingual routing

📊 Observability

• feat(metrics): comboMetrics.ts extended with real-time latency percentile tracking per provider/account
• feat(health): Health API (/api/monitoring/health) now returns per-provider p50Latency and errorRate fields
• feat(usage): Usage history migration for per-model latency tracking

🗄️ DB Migrations

• feat(migrations): New column latency_p50 in combo_metrics table — zero-breaking, safe for existing users

🐛 Bug Fixes / Closures

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• close(#411): better-sqlite3 hashed module resolution on Windows — fixed in v2.6.10 (f02c5b5)

• close(#409): GitHub Copilot chat completions fail with Claude models when files attached — fixed in v2.6.9 (838f1d6)

• close(#405): Duplicate of #411 — resolved

[2.6.10] — 2026-03-17

Windows fix: better-sqlite3 prebuilt download without node-gyp/Python/MSVC (#426).

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(install/#426): On Windows, npm install -g omniroute used to fail with better_sqlite3.node is not a valid Win32 application because the bundled native binary was compiled for Linux. Adds Strategy 1.5 to scripts/postinstall.mjs: uses @mapbox/node-pre-gyp install --fallback-to-build=false (bundled within better-sqlite3) to download the correct prebuilt binary for the current OS/arch without requiring any build tools (no node-gyp, no Python, no MSVC). Falls back to npm rebuild only if the download fails. Adds platform-specific error messages with clear manual fix instructions.

---

[2.6.9] — 2026-03-17

CI fixes (t11 any-budget), bug fix #409 (file attachments via Copilot+Claude), release workflow correction.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(ci): Remove word "any" from comments in openai-responses.ts and chatCore.ts that were failing the t11 any budget check (false positive from regex counting comments)

• fix(chatCore): Normalize unsupported content part types before forwarding to providers (#409 — Cursor sends {type:"file"} when .md files are attached; Copilot and other OpenAI-compat providers reject with "type has to be either 'image_url' or 'text'"; fix converts file/document blocks to text and drops unknown types)

🔧 Workflow

• chore(generate-release): Add ATOMIC COMMIT RULE — version bump (npm version patch) MUST happen before committing feature files to ensure tag always points to a commit containing all version changes together

---

[2.6.8] — 2026-03-17

Sprint: Combo as Agent (system prompt + tool filter), Context Caching Protection, Auto-Update, Detailed Logs, MITM Kiro IDE.

🗄️ DB Migrations (zero-breaking — safe for existing users)

• 005_combo_agent_fields.sql: ALTER TABLE combos ADD COLUMN system_message TEXT DEFAULT NULL, tool_filter_regex TEXT DEFAULT NULL, context_cache_protection INTEGER DEFAULT 0
• 006_detailed_request_logs.sql: New request_detail_logs table with 500-entry ring-buffer trigger, opt-in via settings toggle

✨ Features

• feat(combo): System Message Override per Combo (#399 — system_message field replaces or injects system prompt before forwarding to provider)
• feat(combo): Tool Filter Regex per Combo (#399 — tool_filter_regex keeps only tools matching pattern; supports OpenAI + Anthropic formats)
• feat(combo): Context Caching Protection (#401 — context_cache_protection tags responses with <omniModel>provider/model</omniModel> and pins model for session continuity)
• feat(settings): Auto-Update via Settings (#320 — GET /api/system/version + POST /api/system/update — checks npm registry and updates in background with pm2 restart)
• feat(logs): Detailed Request Logs (#378 — captures full pipeline bodies at 4 stages: client request, translated request, provider response, client response — opt-in toggle, 64KB trim, 500-entry ring-buffer)
• feat(mitm): MITM Kiro IDE profile (#336 — src/mitm/targets/kiro.ts targets api.anthropic.com, reuses existing MITM infrastructure)

---

[2.6.7] — 2026-03-17

Sprint: SSE improvements, local provider_nodes extensions, proxy registry, Claude passthrough fixes.

✨ Features

• feat(health): Background health check for local provider_nodes with exponential backoff (30s→300s) and Promise.allSettled to avoid blocking (#423, @Regis-RCR)
• feat(embeddings): Route /v1/embeddings to local provider_nodes — buildDynamicEmbeddingProvider() with hostname validation (#422, @Regis-RCR)
• feat(audio): Route TTS/STT to local provider_nodes — buildDynamicAudioProvider() with SSRF protection (#416, @Regis-RCR)
• feat(proxy): Proxy registry, management APIs, and quota-limit generalization (#429, @Regis-RCR)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(sse): Strip Claude-specific fields (metadata, anthropic_version) when target is OpenAI-compat (#421, @prakersh)

• fix(sse): Extract Claude SSE usage (input_tokens, output_tokens, cache tokens) in passthrough stream mode (#420, @prakersh)

• fix(sse): Generate fallback call_id for tool calls with missing/empty IDs (#419, @prakersh)

• fix(sse): Claude-to-Claude passthrough — forward body completely untouched, no re-translation (#418, @prakersh)

• fix(sse): Filter orphaned tool_result items after Claude Code context compaction to avoid 400 errors (#417, @prakersh)

• fix(sse): Skip empty-name tool calls in Responses API translator to prevent placeholder_tool infinite loops (#415, @prakersh)

• fix(sse): Strip empty text content blocks before translation (#427, @prakersh)

• fix(api): Add refreshable: true to Claude OAuth test config (#428, @prakersh)

📦 Dependencies

• Bump vitest, @vitest/* and related devDependencies (#414, @dependabot)

---

[2.6.6] — 2026-03-17

Hotfix: Turbopack/Docker compatibility — remove node: protocol from all src/ imports.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(build): Removed node: protocol prefix from import statements in 17 files under src/. The node:fs, node:path, node:url, node:os etc. imports caused Ecmascript file had an error on Turbopack builds (Next.js 15 Docker) and on upgrades from older npm global installs. Affected files: migrationRunner.ts, core.ts, backup.ts, prompts.ts, dataPaths.ts, and 12 others in src/app/api/ and src/lib/.

• chore(workflow): Updated generate-release.md to make Docker Hub sync and dual-VPS deploy mandatory steps in every release.

---

[2.6.5] — 2026-03-17

Sprint: reasoning model param filtering, local provider 404 fix, Kilo Gateway provider, dependency bumps.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(api): Added Kilo Gateway (api.kilo.ai) as a new API Key provider (alias kg) — 335+ models, 6 free models, 3 auto-routing models (kilo-auto/frontier, kilo-auto/balanced, kilo-auto/free). Passthrough models supported via /api/gateway/models endpoint. (PR #408 by @Regis-RCR)

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(sse): Strip unsupported parameters for reasoning models (o1, o1-mini, o1-pro, o3, o3-mini). Models in the o1/o3 family reject temperature, top_p, frequency_penalty, presence_penalty, logprobs, top_logprobs, and n with HTTP 400. Parameters are now stripped at the chatCore layer before forwarding. Uses a declarative unsupportedParams field per model and a precomputed O(1) Map for lookup. (PR #412 by @Regis-RCR)

• fix(sse): Local provider 404 now results in a model-only lockout (5 seconds) instead of a connection-level lockout (2 minutes). When a local inference backend (Ollama, LM Studio, oMLX) returns 404 for an unknown model, the connection remains active and other models continue working immediately. Also fixes a pre-existing bug where model was not passed to markAccountUnavailable(). Local providers detected via hostname (localhost, 127.0.0.1, ::1, extensible via LOCAL_HOSTNAMES env var). (PR #410 by @Regis-RCR)

📦 Dependencies

• better-sqlite3 12.6.2 → 12.8.0
• undici 7.24.2 → 7.24.4
• https-proxy-agent 7 → 8
• agent-base 7 → 8

---

[2.6.4] — 2026-03-17

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(providers): Removed non-existent model names across 5 providers:

  • gemini / gemini-cli: removed gemini-3.1-pro/flash and gemini-3-*-preview (don't exist in Google API v1beta); replaced with gemini-2.5-pro, gemini-2.5-flash, gemini-2.0-flash, gemini-1.5-pro/flash
  • antigravity: removed gemini-3.1-pro-high/low and gemini-3-flash (invalid internal aliases); replaced with real 2.x models
  • github (Copilot): removed gemini-3-flash-preview and gemini-3-pro-preview; replaced with gemini-2.5-flash
  • nvidia: corrected nvidia/llama-3.3-70b-instruct → meta/llama-3.3-70b-instruct (NVIDIA NIM uses meta/ namespace for Meta models); added nvidia/llama-3.1-70b-instruct and nvidia/llama-3.1-405b-instruct

• fix(db/combo): Updated free-stack combo on remote DB: removed qw/qwen3-coder-plus (expired refresh token), corrected nvidia/llama-3.3-70b-instruct → nvidia/meta/llama-3.3-70b-instruct, corrected gemini/gemini-3.1-flash → gemini/gemini-2.5-flash, added if/deepseek-v3.2

---

[2.6.3] — 2026-03-16

Sprint: zod/pino hash-strip baked into build pipeline, Synthetic provider added, VPS PM2 path corrected.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(build): Turbopack hash-strip now runs at compile time for ALL packages — not just better-sqlite3. Step 5.6 in prepublish.mjs walks every .js in app/.next/server/ and strips the 16-char hex suffix from any hashed require(). Fixes zod-dcb22c..., pino-..., etc. MODULE_NOT_FOUND on global npm installs. Closes #398

• fix(deploy): PM2 on both VPS was pointing to stale git-clone directories. Reconfigured to app/server.js in the npm global package. Updated /deploy-vps workflow to use npm pack + scp (npm registry rejects 299MB packages).

✨ Features

• feat(provider): Synthetic (synthetic.new) — privacy-focused OpenAI-compatible inference. passthroughModels: true for dynamic HuggingFace model catalog. Initial models: Kimi K2.5, MiniMax M2.5, GLM 4.7, DeepSeek V3.2. (PR #404 by @Regis-RCR)

📋 Issues Closed

• close #398: npm hash regression — fixed by compile-time hash-strip in prepublish
• triage #324: Bug screenshot without steps — requested reproduction details

---

[2.6.2] — 2026-03-16

Sprint: module hashing fully fixed, 2 PRs merged (Anthropic tools filter + custom endpoint paths), Alibaba Cloud DashScope provider added, 3 stale issues closed.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(build): Extended webpack externals hash-strip to cover ALL serverExternalPackages, not just better-sqlite3. Next.js 16 Turbopack hashes zod, pino, and every other server-external package into names like zod-dcb22c6336e0bc69 that don't exist in node_modules at runtime. A HASH_PATTERN regex catch-all now strips the 16-char suffix and falls back to the base package name. Also added NEXT_PRIVATE_BUILD_WORKER=0 in prepublish.mjs to reinforce webpack mode, plus a post-build scan that reports any remaining hashed refs. (#396, #398, PR #403)

• fix(chat): Anthropic-format tool names (tool.name without .function wrapper) were silently dropped by the empty-name filter introduced in #346. LiteLLM proxies requests with anthropic/ prefix in Anthropic Messages API format, causing all tools to be filtered and Anthropic to return 400: tool_choice.any may only be specified while providing tools. Fixed by falling back to tool.name when tool.function.name is absent. Added 8 regression unit tests. (PR #397)

✨ Features

• feat(api): Custom endpoint paths for OpenAI-compatible provider nodes — configure chatPath and modelsPath per node (e.g. /v4/chat/completions) in the provider connection UI. Includes a DB migration (003_provider_node_custom_paths.sql) and URL path sanitization (no .. traversal, must start with /). (PR #400)
• feat(provider): Alibaba Cloud DashScope added as OpenAI-compatible provider. International endpoint: dashscope-intl.aliyuncs.com/compatible-mode/v1. 12 models: qwen-max, qwen-plus, qwen-turbo, qwen3-coder-plus/flash, qwq-plus, qwq-32b, qwen3-32b, qwen3-235b-a22b. Auth: Bearer API key.

📋 Issues Closed

• close #323: Cline connection error [object Object] — fixed in v2.3.7; instructed user to upgrade from v2.2.9
• close #337: Kiro credit tracking — implemented in v2.5.5 (#381); pointed user to Dashboard → Usage
• triage #402: ARM64 macOS DMG damaged — requested macOS version, exact error, and advised xattr -d com.apple.quarantine workaround

---

[2.6.1] — 2026-03-15

Critical startup fix: v2.6.0 global npm installs crashed with a 500 error due to a Turbopack/webpack module-name hashing bug in the Next.js 16 instrumentation hook.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(build): Force better-sqlite3 to always be required by its exact package name in the webpack server bundle. Next.js 16 compiled the instrumentation hook into a separate chunk and emitted require('better-sqlite3-<hash>') — a hashed module name that doesn't exist in node_modules — even though the package was listed in serverExternalPackages. Added an explicit externals function to the server webpack config so the bundler always emits require('better-sqlite3'), resolving the startup 500 Internal Server Error on clean global installs. (#394, PR #395)

🔧 CI

• ci: Added workflow_dispatch to npm-publish.yml with version sync safeguard for manual triggers (#392)
• ci: Added workflow_dispatch to docker-publish.yml, updated GitHub Actions to latest versions (#392)

---

[2.6.0] - 2026-03-15

Issue resolution sprint: 4 bugs fixed, logs UX improved, Kiro credit tracking added.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(media): ComfyUI and SD WebUI no longer appear in the Media page provider list when unconfigured — fetches /api/providers on mount and hides local providers with no connections (#390)

• fix(auth): Round-robin no longer re-selects rate-limited accounts immediately after cooldown — backoffLevel is now used as primary sort key in the LRU rotation (#340)

• fix(oauth): Qoder (and other providers that redirect to their own UI) no longer leave the OAuth modal stuck at "Waiting for Authorization" — popup-closed detector auto-transitions to manual URL input mode (#344)

• fix(logs): Request log table is now readable in light mode — status badges, token counts, and combo tags use adaptive dark: color classes (#378)

✨ Features

• feat(kiro): Kiro credit tracking added to usage fetcher — queries getUserCredits from AWS CodeWhisperer endpoint (#337)

🛠 Chores

• chore(tests): Aligned test:plan3, test:fixes, test:security to use same tsx/esm loader as npm test — eliminates module resolution false negatives in targeted runs (PR #386)

---

[2.5.9] - 2026-03-15

Codex native passthrough fix + route body validation hardening.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(codex): Preserve native Responses API passthrough for Codex clients — avoids unnecessary translation mutations (PR #387)

• fix(api): Validate request bodies on pricing/sync and task-routing routes — prevents crashes from malformed inputs (PR #388)

• fix(auth): JWT secrets persist across restarts via src/lib/db/secrets.ts — eliminates 401 errors after pm2 restart (PR #388)

---

[2.5.8] - 2026-03-15

Build fix: restore VPS connectivity broken by v2.5.7 incomplete publish.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(build): scripts/prepublish.mjs still used deprecated --webpack flag causing Next.js standalone build to fail silently — npm publish completed without app/server.js, breaking VPS deployment

---

[2.5.7] - 2026-03-15

Media playground error handling fixes.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(media): Transcription "API Key Required" false positive when audio contains no speech (music, silence) — now shows "No speech detected" instead

• fix(media): upstreamErrorResponse in audioTranscription.ts and audioSpeech.ts now returns proper JSON ({error:{message}}), enabling correct 401/403 credential error detection in the MediaPageClient

• fix(media): parseApiError now handles Deepgram's err_msg field and detects "api key" in error messages for accurate credential error classification

---

[2.5.6] - 2026-03-15

Critical security/auth fixes: Antigravity OAuth broken + JWT sessions lost after restart.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(oauth) #384: Antigravity Google OAuth now correctly sends client_secret to the token endpoint. The fallback for ANTIGRAVITY_OAUTH_CLIENT_SECRET was an empty string, which is falsy — so client_secret was never included in the request, causing "client_secret is missing" errors for all users without a custom env var. Closes #383.

• fix(auth) #385: JWT_SECRET is now persisted to SQLite (namespace='secrets') on first generation and reloaded on subsequent starts. Previously, a new random secret was generated each process startup, invalidating all existing cookies/sessions after any restart or upgrade. Affects both JWT_SECRET and API_KEY_SECRET. Closes #382.

---

[2.5.5] - 2026-03-15

Model list dedup fix, Electron standalone build hardening, and Kiro credit tracking.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(models) #380: GET /api/models now includes provider aliases when building the active-provider filter — models for claude (alias cc) and github (alias gh) were always shown regardless of whether a connection was configured, because PROVIDER_MODELS keys are aliases but DB connections are stored under provider IDs. Fixed by expanding each active provider ID to also include its alias via PROVIDER_ID_TO_ALIAS. Closes #353.

• fix(electron) #379: New scripts/prepare-electron-standalone.mjs stages a dedicated /.next/electron-standalone bundle before Electron packaging. Aborts with a clear error if node_modules is a symlink (electron-builder would ship a runtime dependency on the build machine). Cross-platform path sanitization via path.basename. By @kfiramar.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• feat(kiro) #381: Kiro credit balance tracking — usage endpoint now returns credit data for Kiro accounts by calling codewhisperer.us-east-1.amazonaws.com/getUserCredits (same endpoint Kiro IDE uses internally). Returns remaining credits, total allowance, renewal date, and subscription tier. Closes #337.

[2.5.4] - 2026-03-15

Logger startup fix, login bootstrap security fix, and dev HMR reliability improvement. CI infrastructure hardened.

🐛 Bug Fixes (PRs #374, #375, #376 by @kfiramar)

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(logger) #376: Restore pino transport logger path — formatters.level combined with transport.targets is rejected by pino. Transport-backed configs now strip the level formatter via getTransportCompatibleConfig(). Also corrects numeric level mapping in /api/logs/console: 30→info, 40→warn, 50→error (was shifted by one).

• fix(login) #375: Login page now bootstraps from the public /api/settings/require-login endpoint instead of the protected /api/settings. In password-protected setups, the pre-auth page was receiving a 401 and falling back to safe defaults unnecessarily. The public route now returns all bootstrap metadata (requireLogin, hasPassword, setupComplete) with a conservative 200 fallback on error.

• fix(dev) #374: Add localhost and 127.0.0.1 to allowedDevOrigins in next.config.mjs — HMR websocket was blocked when accessing the app via loopback address, producing repeated cross-origin warnings.

🔧 CI & Infrastructure

• ESLint OOM fix: eslint.config.mjs now ignores vscode-extension/**, electron/**, docs/**, app/.next/**, and clipr/** — ESLint was crashing with a JS heap OOM by scanning VS Code binary blobs and compiled chunks.
• Unit test fix: Removed stale ALTER TABLE provider_connections ADD COLUMN "group" from 2 test files — column is now part of the base schema (added in #373), causing SQLITE_ERROR: duplicate column name on every CI run.
• Pre-commit hook: Added npm run test:unit to .husky/pre-commit — unit tests now block broken commits before they reach CI.

[2.5.3] - 2026-03-14

Critical bugfixes: DB schema migration, startup env loading, provider error state clearing, and i18n tooltip fix. Code quality improvements on top of each PR.

🐛 Bug Fixes (PRs #369, #371, #372, #373 by @kfiramar)

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix(db) #373: Add provider_connections.group column to base schema + backfill migration for existing databases — column was used in all queries but missing from schema definition

• fix(i18n) #371: Replace non-existent t("deleteConnection") key with existing providers.delete key — fixes MISSING_MESSAGE: providers.deleteConnection runtime error on provider detail page

• fix(auth) #372: Clear stale error metadata (errorCode, lastErrorType, lastErrorSource) from provider accounts after genuine recovery — previously, recovered accounts kept appearing as failed

• fix(startup) #369: Unify env loading across npm run start, run-standalone.mjs, and Electron to respect DATA_DIR/.env → ~/.omniroute/.env → ./.env priority — prevents generating a new STORAGE_ENCRYPTION_KEY over an existing encrypted database

🔧 Code Quality

• Documented result.success vs response?.ok patterns in auth.ts (both intentional, now explained)
• Normalized overridePath?.trim() in electron/main.js to match bootstrap-env.mjs
• Added preferredEnv merge order comment in Electron startup

Codex account quota policy with auto-rotation, fast tier toggle, gpt-5.4 model, and analytics label fix.

✨ New Features (PRs #366, #367, #368)

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Codex Quota Policy (PR #366): Per-account 5h/weekly quota window toggles in Provider dashboard. Accounts are automatically skipped when enabled windows reach 90% threshold and re-admitted after resetAt. Includes quotaCache.ts with side-effect free status getter.

• Codex Fast Tier Toggle (PR #367): Dashboard → Settings → Codex Service Tier. Default-off toggle injects service_tier: "flex" only for Codex requests, reducing cost ~80%. Full stack: UI tab + API endpoint + executor + translator + startup restore.

• gpt-5.4 Model (PR #368): Adds cx/gpt-5.4 and codex/gpt-5.4 to the Codex model registry. Regression test included.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix #356: Analytics charts (Top Provider, By Account, Provider Breakdown) now display human-readable provider names/labels instead of raw internal IDs for OpenAI-compatible providers.

Major release: strict-random routing strategy, API key access controls, connection groups, external pricing sync, and critical bug fixes for thinking models, combo testing, and tool name validation.

✨ New Features (PRs #363 & #365)

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Strict-Random Routing Strategy: Fisher-Yates shuffle deck with anti-repeat guarantee and mutex serialization for concurrent requests. Independent decks per combo and per provider.

• API Key Access Controls: allowedConnections (restrict which connections a key can use), is_active (enable/disable key with 403), accessSchedule (time-based access control), autoResolve toggle, rename keys via PATCH.

• Connection Groups: Group provider connections by environment. Accordion view in Limits page with localStorage persistence and smart auto-switch.

• External Pricing Sync (LiteLLM): 3-tier pricing resolution (user overrides → synced → defaults). Opt-in via PRICING_SYNC_ENABLED=true. MCP tool omniroute_sync_pricing. 23 new tests.

• i18n: 30 languages updated with strict-random strategy, API key management strings. pt-BR fully translated.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• fix #355: Stream idle timeout increased from 60s to 300s — prevents aborting extended-thinking models (claude-opus-4-6, o3, etc.) during long reasoning phases. Configurable via STREAM_IDLE_TIMEOUT_MS.

• fix #350: Combo test now bypasses REQUIRE_API_KEY=true using internal header, and uses OpenAI-compatible format universally. Timeout extended from 15s to 20s.

• fix #346: Tools with empty function.name (forwarded by Claude Code) are now filtered before upstream providers receive them, preventing "Invalid input[N].name: empty string" errors.

🗑️ Closed Issues

• #341: Debug section removed — replacement is /dashboard/logs and /dashboard/health.

API Key Round-Robin support for multi-key provider setups, and confirmation of wildcard routing and quota window rolling already in place.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• API Key Round-Robin (T07): Provider connections can now hold multiple API keys (Edit Connection → Extra API Keys). Requests rotate round-robin between primary + extra keys via providerSpecificData.extraApiKeys[]. Keys are held in-memory indexed per connection — no DB schema changes required.

📝 Already Implemented (confirmed in audit)

• Wildcard Model Routing (T13): wildcardRouter.ts with glob-style wildcard matching (gpt*, claude-?-sonnet, etc.) is already integrated into model.ts with specificity ranking.
• Quota Window Rolling (T08): accountFallback.ts:isModelLocked() already auto-advances the window — if Date.now() > entry.until, lock is deleted immediately (no stale blocking).

UI polish, routing strategy additions, and graceful error handling for usage limits.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Fill-First & P2C Routing Strategies: Added fill-first (drain quota before moving on) and p2c (Power-of-Two-Choices low-latency selection) to combo strategy picker, with full guidance panels and color-coded badges.

• Free Stack Preset Models: Creating a combo with the Free Stack template now auto-fills 7 best-in-class free provider models (Gemini CLI, Kiro, Qoder×2, Qwen, NVIDIA NIM, Groq). Users just activate the providers and get a $0/month combo out-of-the-box.

• Wider Combo Modal: Create/Edit combo modal now uses max-w-4xl for comfortable editing of large combos.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Limits page HTTP 500 for Codex & GitHub: getCodexUsage() and getGitHubUsage() now return a user-friendly message when the provider returns 401/403 (expired token), instead of throwing and causing a 500 error on the Limits page.

• MaintenanceBanner false-positive: Banner no longer shows "Server is unreachable" spuriously on page load. Fixed by calling checkHealth() immediately on mount and removing stale show-state closure.

• Provider icon tooltips: Edit (pencil) and delete icon buttons in the provider connection row now have native HTML tooltips — all 6 action icons are now self-documented.

Multiple improvements from community issue analysis, new provider support, bug fixes for token tracking, model routing, and streaming reliability.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Task-Aware Smart Routing (T05): Automatic model selection based on request content type — coding → deepseek-chat, analysis → gemini-2.5-pro, vision → gpt-4o, summarization → gemini-2.5-flash. Configurable via Settings. New GET/PUT/POST /api/settings/task-routing API.

• HuggingFace Provider: Added HuggingFace Router as an OpenAI-compatible provider with Llama 3.1 70B/8B, Qwen 2.5 72B, Mistral 7B, Phi-3.5 Mini.

• Vertex AI Provider: Added Vertex AI (Google Cloud) provider with Gemini 2.5 Pro/Flash, Gemma 2 27B, Claude via Vertex.

• Playground File Uploads: Audio upload for transcription, image upload for vision models (auto-detect by model name), inline image rendering for image generation results.

• Model Select Visual Feedback: Already-added models in combo picker now show ✓ green badge — prevents duplicate confusion.

• Qwen Compatibility (PR #352): Updated User-Agent and CLI fingerprint settings for Qwen provider compatibility.

• Round-Robin State Management (PR #349): Enhanced round-robin logic to handle excluded accounts and maintain rotation state correctly.

• Clipboard UX (PR #360): Hardened clipboard operations with fallback for non-secure contexts; Claude tool normalization improvements.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Fix #302 — OpenAI SDK stream=False drops tool_calls: T01 Accept header negotiation no longer forces streaming when body.stream is explicitly false. Was causing tool_calls to be silently dropped when using the OpenAI Python SDK in non-streaming mode.

• Fix #73 — Claude Haiku routed to OpenAI without provider prefix: claude-* models sent without a provider prefix now correctly route to the antigravity (Anthropic) provider. Added gemini-*/gemma-* → gemini heuristic as well.

• Fix #74 — Token counts always 0 for Antigravity/Claude streaming: The message_start SSE event which carries input_tokens was not being parsed by extractUsage(), causing all input token counts to drop. Input/output token tracking now works correctly for streaming responses.

• Fix #180 — Model import duplicates with no feedback: ModelSelectModal now shows ✓ green highlight for models already in the combo, making it obvious they're already added.

• Media page generation errors: Image results now render as <img> tags instead of raw JSON. Transcription results shown as readable text. Credential errors show an amber banner instead of silent failure.

• Token refresh button on provider page: Manual token refresh UI added for OAuth providers.

🔧 Improvements

• Provider Registry: HuggingFace and Vertex AI added to providerRegistry.ts and providers.ts (frontend).
• Read Cache: New src/lib/db/readCache.ts for efficient DB read caching.
• Quota Cache: Improved quota cache with TTL-based eviction.

📦 Dependencies

• dompurify → 3.3.3 (PR #347)
• undici → 7.24.2 (PR #348, #361)
• docker/setup-qemu-action → v4 (PR #342)
• docker/setup-buildx-action → v4 (PR #343)

📁 New Files

File	Purpose
open-sse/services/taskAwareRouter.ts	Task-aware routing logic (7 task types)
src/app/api/settings/task-routing/route.ts	Task routing config API
src/app/api/providers/[id]/refresh/route.ts	Manual OAuth token refresh
src/lib/db/readCache.ts	Efficient DB read cache
src/shared/utils/clipboard.ts	Hardened clipboard with fallback

[2.4.1] - 2026-03-13

🐛 Fix

• Combos modal: Free Stack visible and prominent — Free Stack template was hidden (4th in 3-column grid). Fixed: moved to position 1, switched to 2x2 grid so all 4 templates are visible, green border + FREE badge highlight.

[2.4.0] - 2026-03-13

Major release — Free Stack ecosystem, transcription playground overhaul, 44+ providers, comprehensive free tier documentation, and UI improvements across the board.

✨ Features

• Combos: Free Stack template — New 4th template "Free Stack ($0)" using round-robin across Kiro + Qoder + Qwen + Gemini CLI. Suggests the pre-built zero-cost combo on first use.
• Media/Transcription: Deepgram as default — Deepgram (Nova 3, $200 free) is now the default transcription provider. AssemblyAI ($50 free) and Groq Whisper (free forever) shown with free credit badges.
• README: "Start Free" section — New early-README 5-step table showing how to set up zero-cost AI in minutes.
• README: Free Transcription Combo — New section with Deepgram/AssemblyAI/Groq combo suggestion and per-provider free credit details.
• providers.ts: hasFree flag — NVIDIA NIM, Cerebras, and Groq marked with hasFree badge and freeNote for the providers UI.
• i18n: templateFreeStack keys — Free Stack combo template translated and synced to all 30 languages.

[2.3.16] - 2026-03-13

📖 Documentation

• README: 44+ Providers — Updated all 3 occurrences of "36+ providers" to "44+" reflecting the actual codebase count (44 providers in providers.ts)
• README: New Section "🆓 Free Models — What You Actually Get" — Added 7-provider table with per-model rate limits for: Kiro (Claude unlimited via AWS Builder ID), Qoder (5 models unlimited), Qwen (4 models unlimited), Gemini CLI (180K/mo), NVIDIA NIM (~40 RPM dev-forever), Cerebras (1M tok/day / 60K TPM), Groq (30 RPM / 14.4K RPD). Includes the /usr/bin/bash Ultimate Free Stack combo recommendation.
• README: Pricing Table Updated — Added Cerebras to API KEY tier, fixed NVIDIA from "1000 credits" to "dev-forever free", updated Qoder/Qwen model counts and names
• README: Qoder 8→5 models (named: kimi-k2-thinking, qwen3-coder-plus, deepseek-r1, minimax-m2, kimi-k2)
• README: Qwen 3→4 models (named: qwen3-coder-plus, qwen3-coder-flash, qwen3-coder-next, vision-model)

[2.3.15] - 2026-03-13

✨ Features

• Auto-Combo Dashboard (Tier Priority): Added 🏷️ Tier as the 7th scoring factor label in the /dashboard/auto-combo factor breakdown display — all 7 Auto-Combo scoring factors are now visible.
• i18n — autoCombo section: Added 20 new translation keys for the Auto-Combo dashboard (title, status, modePack, providerScores, factorTierPriority, etc.) to all 30 language files.

[2.3.14] - 2026-03-13

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Qoder OAuth (#339): Restored the valid default clientSecret — was previously an empty string, causing "Bad client credentials" on every connect attempt. The public credential is now the default fallback (overridable via QODER_OAUTH_CLIENT_SECRET env var).

• MITM server not found (#335): prepublish.mjs now compiles src/mitm/*.ts to JavaScript using tsc before copying to the npm bundle. Previously only raw .ts files were copied — meaning server.js never existed in npm/Volta global installs.

• GeminiCLI missing projectId (#338): Instead of throwing a hard 500 error when projectId is missing from stored credentials (e.g. after Docker restart), OmniRoute now logs a warning and attempts the request — returning a meaningful provider-side error instead of an OmniRoute crash.

• Electron version mismatch (#323): Synced electron/package.json version to 2.3.13 (was 2.0.13) so the desktop binary version matches the npm package.

✨ New Models (#334)

• Kiro: claude-sonnet-4, claude-opus-4.6, deepseek-v3.2, minimax-m2.1, qwen3-coder-next, auto
• Codex: gpt5.4

🔧 Improvements

• Tier Scoring (API + Validation): Added tierPriority (weight 0.05) to the ScoringWeights Zod schema and the combos/auto API route — the 7th scoring factor is now fully accepted by the REST API and validated on input. stability weight adjusted from 0.10 to 0.05 to keep total sum = 1.0.

✨ New Features

• feat(docs): integrate multi-page documentation into OmniRoute dashboard (#1969)

• feat(settings): add request body limit setting (#1968)

• feat(auth): add Gemini CLI OAuth client secret default (#1974)

• feat(models): expose models.dev context windows in /v1/models (#1972)

• fix(db): resolve legacy encryption fallback causing re-encryption loops (#1941)

• fix(auth): fix Codex assistant final_answer response sanitization (#1965)

• feat(providers): Implement Image Generation and Editing capabilities for ChatGPT Web, including in-band chat image generation and caching (#1606).

• feat(ui): Integrate OpenCode Zen/Go API tool logo SVG and polish API key copy-to-clipboard interactions (#1607).

• feat(providers): Integrate AgentRouter as a new OpenAI-compatible passthrough provider with $200 free credits via sign-up (Issue #1572).

• feat(ui): Implement on-demand per-model testing in the provider dashboard, allowing single-token diagnostic checks without triggering rate-limits (Issue #1532).

• Tiered Quota Scoring (Auto-Combo): Added tierPriority as a 7th scoring factor — accounts with Ultra/Pro tiers are now preferred over Free tiers when other factors are equal. New optional fields accountTier and quotaResetIntervalSecs on ProviderCandidate. All 4 mode packs updated (ship-fast, cost-saver, quality-first, offline-friendly).

• Intra-Family Model Fallback (T5): When a model is unavailable (404/400/403), OmniRoute now automatically falls back to sibling models from the same family before returning an error (modelFamilyFallback.ts).

• Configurable API Bridge Timeout: API_BRIDGE_PROXY_TIMEOUT_MS env var lets operators tune the proxy timeout (default 30s). Fixes 504 errors on slow upstream responses. (#332)

• Star History: Replaced star-history.com widget with starchart.cc (?variant=adaptive) in all 30 READMEs — adapts to light/dark theme, real-time updates.

🐛 Bug Fixes

• fix(mitm): Compile MITM utilities as NodeNext ESM during prepublish, copy the CommonJS MITM server into the standalone artifact, and resolve MITM data paths without relying on Next.js aliases in packaged runtime.

• fix(build): Move the local .tmp/wine32 Wine prefix out of the isolated Next.js build path so Windows Electron packaging artifacts cannot trigger EACCES scans during Node 24 builds.

• fix(build): Copy the wreq-js native runtime directory into the isolated Next.js standalone output so packaged Playwright/E2E starts can load the instrumentation hook on Linux.

• fix(api): Validate the Codex Responses websocket bridge and /v1/batches JSON payloads with Zod before use, keeping request.json() route validation green and returning explicit 400 responses for invalid bodies.

• fix(providers): Add explicit typing to provider alias and category helpers so the strict typecheck:noimplicit:core CI gate passes.

• fix(ui): Keep the upstream proxy provider detail page labeled with a fallback "Managed via Upstream Proxy Settings" management surface when translations are unavailable.

• fix(electron): Harden the production desktop CSP by removing unsafe-eval outside development and adding object, base URI, form action, frame ancestor, and worker restrictions.

• fix(cli): Replace shell-interpolated setup and privileged command execution paths with argument-based spawn/execFile helpers for database setup, Tailscale sudo commands, MITM DNS edits, and certificate install/uninstall flows.

• fix(ui): Keep provider icons resilient by using direct @lobehub/icons components first, then local PNG/SVG fallbacks, avoiding the @lobehub/ui peer runtime in the dashboard.

• Auth — First-time password: INITIAL_PASSWORD env var is now accepted when setting the first dashboard password. Uses timingSafeEqual for constant-time comparison, preventing timing attacks. (#333)

• README Truncation: Fixed a missing </details> closing tag in the Troubleshooting section that caused GitHub to stop rendering everything below it (Tech Stack, Docs, Roadmap, Contributors).

• pnpm install: Removed redundant @swc/helpers override from package.json that conflicted with the direct dependency, causing EOVERRIDE errors on pnpm. Added pnpm.onlyBuiltDependencies config.

• CLI Path Injection (T12): Added isSafePath() validator in cliRuntime.ts to block path traversal and shell metacharacters in CLI_*_BIN env vars.

• CI: Regenerated package-lock.json after override removal to fix npm ci failures on GitHub Actions.

🔧 Improvements

• Response Format (T1): response_format (json_schema/json_object) now injected as a system prompt for Claude, enabling structured output compatibility.
• 429 Retry (T2): Intra-URL retry for 429 responses (2× attempts with 2s delay) before falling back to next URL.
• Gemini CLI Headers (T3): Added User-Agent and X-Goog-Api-Client fingerprint headers for Gemini CLI compatibility.
• Pricing Catalog (T9): Added deepseek-3.1, deepseek-3.2, and qwen3-coder-next pricing entries.

📁 New Files

File	Purpose
open-sse/services/modelFamilyFallback.ts	Model family definitions and intra-family fallback logic

Fixed

• KiloCode: kilocode healthcheck timeout already fixed in v2.3.11
• OpenCode: Add opencode to cliRuntime registry with 15s healthcheck timeout
• OpenClaw / Cursor: Increase healthcheck timeout to 15s for slow-start variants
• VPS: Install droid and openclaw npm packages; activate CLI_EXTRA_PATHS for kiro-cli
• cliRuntime: Add opencode tool registration and increase timeout for continue

[2.3.11] - 2026-03-12

Fixed

• KiloCode healthcheck: Increase healthcheckTimeoutMs from 4000ms to 15000ms — kilocode renders an ASCII logo banner on startup causing false healthcheck_failed on slow/cold-start environments

[2.3.10] - 2026-03-12

Fixed

• Lint: Fix check:any-budget:t11 failure — replace as any with as Record<string, unknown> in OAuthModal.tsx (3 occurrences)

Docs

• CLI-TOOLS.md: Complete guide for all 11 CLI tools (claude, codex, gemini, opencode, cline, kilocode, continue, kiro-cli, cursor, droid, openclaw)
• i18n: CLI-TOOLS.md synced to 30 languages with translated title + intro

[2.3.8] - 2026-03-12

[2.3.9] - 2026-03-12

Added

• /v1/completions: New legacy OpenAI completions endpoint — accepts both prompt string and messages array, normalizes to chat format automatically
• EndpointPage: Now shows all 3 OpenAI-compatible endpoint types: Chat Completions, Responses API, and Legacy Completions
• i18n: Added completionsLegacy/completionsLegacyDesc to 30 language files

Fixed

• OAuthModal: Fix [object Object] displayed on all OAuth connection errors — properly extract .message from error response objects in all 3 throw new Error(data.error) calls (exchange, device-code, authorize)
• Affects Cline, Codex, GitHub, Qwen, Kiro, and all other OAuth providers

[2.3.7] - 2026-03-12

Fixed

• Cline OAuth: Add decodeURIComponent before base64 decode so URL-encoded auth codes from the callback URL are parsed correctly, fixing "invalid or expired authorization code" errors on remote (LAN IP) setups
• Cline OAuth: mapTokens now populates name = firstName + lastName || email so Cline accounts show real user names instead of "Account #ID"
• OAuth account names: All OAuth exchange flows (exchange, poll, poll-callback) now normalize name = email when name is missing, so every OAuth account shows its email as the display label in the Providers dashboard
• OAuth account names: Removed sequential "Account N" fallback in db/providers.ts — accounts with no email/name now use a stable ID-based label via getAccountDisplayName() instead of a sequential number that changes when accounts are deleted

[2.3.6] - 2026-03-12

Fixed

• Provider test batch: Fixed Zod schema to accept providerId: null (frontend sends null for non-provider modes); was incorrectly returning "Invalid request" for all batch tests
• Provider test modal: Fixed [object Object] display by normalizing API error objects to strings before rendering in setTestResults and ProviderTestResultsView
• i18n: Added missing keys cliTools.toolDescriptions.opencode, cliTools.toolDescriptions.kiro, cliTools.guides.opencode, cliTools.guides.kiro to en.json
• i18n: Synchronized 1111 missing keys across all 29 non-English language files using English values as fallbacks

[2.3.5] - 2026-03-11

Fixed

• @swc/helpers: Added permanent postinstall fix to copy @swc/helpers into the standalone app's node_modules — prevents MODULE_NOT_FOUND crash on global npm installs

[2.3.4] - 2026-03-10

Added

• Multiple provider integrations and dashboard improvements