docs(quality): document how the CodeQL ratchet refreshes and how to tighten it

2026-09-03 21:42:41 +03:00 · 2026-09-03 11:45:34 -03:00
230 changed files with 873 additions and 6940 deletions
--- a/.env.example
+++ b/.env.example
@@ -1290,7 +1290,7 @@ GITHUB_OAUTH_CLIENT_ID=Iv1.b507a08c87ecfe98
 # Used by: open-sse/executors/base.ts — buildHeaders() dynamic lookup.
 # Update these when providers release new CLI versions to avoid blocks.

-CLAUDE_USER_AGENT="claude-cli/2.1.258 (external, cli)"
+CLAUDE_USER_AGENT="claude-cli/2.1.219 (external, cli)"

 # Disable the deterministic tool-name cloak applied on both Anthropic-bound paths
 # (executors/base.ts native OAuth + executors/cliproxyapi.ts CLIProxyAPI) —
--- a/AGENTS.md
+++ b/AGENTS.md
@@ -46,7 +46,7 @@ Repository map and Reference Documentation sections below.

 ## Project at a Glance

-**OmniRoute** — unified AI proxy/router. One endpoint, 356 LLM providers, auto-fallback.
+**OmniRoute** — unified AI proxy/router. One endpoint, 355 LLM providers, auto-fallback.

 | Layer         | Location                | Purpose                                                                                                                                                                   |
 | ------------- | ----------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
@@ -56,7 +56,7 @@ Repository map and Reference Documentation sections below.
 | Translators   | `open-sse/translator/`  | Format conversion (OpenAI↔Claude↔Gemini)                                                                                                                                  |
 | Transformer   | `open-sse/transformer/` | Responses API ↔ Chat Completions                                                                                                                                          |
 | Services      | `open-sse/services/`    | Combo routing, rate limits, caching, etc                                                                                                                                  |
-| Database      | `src/lib/db/`           | SQLite domain modules (169 migrations)                                                                                                                                    |
+| Database      | `src/lib/db/`           | SQLite domain modules (168 migrations)                                                                                                                                    |
 | Domain/Policy | `src/domain/`           | Policy engine, cost rules, fallback logic                                                                                                                                 |
 | MCP Server    | `open-sse/mcp-server/`  | 110 tools (45 canonical + memory/skill/GitHub/pool/gamification/plugin/Notion/Obsidian/local-corpus/RTK modules), 3 transports (stdio / SSE / Streamable HTTP), 33 scopes |
 | A2A Server    | `src/lib/a2a/`          | JSON-RPC 2.0 agent protocol                                                                                                                                               |
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -4,7 +4,6 @@

 ### ✨ New Features

- **feat(dashboard):** adaptive context-budget dial on the compression settings panel — mode (`off` / `floor` / `replace-autotrigger`) and policy (`reserve-output` / `percentage` / `absolute`) persist via `PUT /api/settings/compression` `contextBudget`. Completes the dashboard half of #7005 (API + DB already shipped in #7183).
 - **feat(sse): STRICT_ZERO_COST** — opt-in, off-by-default `freeAccessPolicy: "strict"` setting
  that hard-verifies every auto-combo candidate against live quota state and per-connection
  economic safety before it can be dispatched, going beyond `hidePaidModels`'s static catalog
--- a/README.md
+++ b/README.md
@@ -7,7 +7,7 @@

 # 🚀 OmniRoute — The Free AI Gateway

-<img src="./docs/diagrams/readme-hero.svg" width="100%" alt="OmniRoute — Never stop coding. Every AI tool → 356 providers — 150+ free — through one endpoint. Claude Code, Codex, Cursor, Cline, Copilot & Antigravity into FREE Claude / GPT / Gemini with auto-fallback. RTK + Caveman stacked compression saves 15–95% tokens (~89% avg) — never hit limits. 356 AI providers · 150+ free tiers · ~1.51B free tokens/mo · 19 routing strategies · $0 to start."/>
+<img src="./docs/diagrams/readme-hero.svg" width="100%" alt="OmniRoute — Never stop coding. Every AI tool → 355 providers — 150+ free — through one endpoint. Claude Code, Codex, Cursor, Cline, Copilot & Antigravity into FREE Claude / GPT / Gemini with auto-fallback. RTK + Caveman stacked compression saves 15–95% tokens (~89% avg) — never hit limits. 355 AI providers · 150+ free tiers · ~1.51B free tokens/mo · 19 routing strategies · $0 to start."/>

 </div>

@@ -209,7 +209,7 @@ curl http://localhost:20128/v1/chat/completions \

 </div>

-<img src="./docs/diagrams/promise-pillars.svg" width="100%" alt="The Promise — One endpoint and 356 providers. Automatic fallback keeps routing while another healthy target is available. Six pillars: resilient fallback across 356 providers · up to 95% token savings on eligible workloads · $0 to start with 150+ free tiers and 53 recurring/keyless free-forever providers · 36 CLI/agent integrations through one config · OpenAI, Claude, Gemini and Responses API compatibility at /v1 · production controls including circuit breakers, TLS stealth, MCP 110 tools, A2A, memory, guardrails, evals and 39,000+ static test declarations across 5,100+ tracked test files."/>
+<img src="./docs/diagrams/promise-pillars.svg" width="100%" alt="The Promise — One endpoint and 355 providers. Automatic fallback keeps routing while another healthy target is available. Six pillars: resilient fallback across 355 providers · up to 95% token savings on eligible workloads · $0 to start with 150+ free tiers and 53 recurring/keyless free-forever providers · 36 CLI/agent integrations through one config · OpenAI, Claude, Gemini and Responses API compatibility at /v1 · production controls including circuit breakers, TLS stealth, MCP 110 tools, A2A, memory, guardrails, evals and 39,000+ static test declarations across 5,100+ tracked test files."/>

 <br/>
 <br/>
@@ -462,7 +462,7 @@ All **19** strategies — mix & match per combo step:

 </div>

-<img src="./docs/diagrams/comparison-table.svg" width="100%" alt="What sets OmniRoute apart — a dated feature snapshot vs 9router, OpenRouter, CLIProxyAPI and LiteLLM across 13 capabilities. OmniRoute: 356 providers, 150+ free tiers built in, 19 routing strategies, 12-engine token compression, built-in MCP server with 110 tools, A2A agent protocol, persistent memory, guardrails, cloud agents, TLS fingerprint stealth, Desktop/Termux/PWA and 42 i18n UI locales. OmniRoute is MIT-licensed and self-hostable. Competitor capabilities and counts may change; see the linked methodology."/>
+<img src="./docs/diagrams/comparison-table.svg" width="100%" alt="What sets OmniRoute apart — a dated feature snapshot vs 9router, OpenRouter, CLIProxyAPI and LiteLLM across 13 capabilities. OmniRoute: 355 providers, 150+ free tiers built in, 19 routing strategies, 12-engine token compression, built-in MCP server with 110 tools, A2A agent protocol, persistent memory, guardrails, cloud agents, TLS fingerprint stealth, Desktop/Termux/PWA and 42 i18n UI locales. OmniRoute is MIT-licensed and self-hostable. Competitor capabilities and counts may change; see the linked methodology."/>

 <sub>📊 Full methodology &amp; per-feature detail vs 9router, OpenRouter, CLIProxyAPI &amp; LiteLLM → [`docs/comparison/OMNIROUTE_VS_ALTERNATIVES.md`](docs/comparison/OMNIROUTE_VS_ALTERNATIVES.md)</sub>

@@ -1244,7 +1244,7 @@ Métricas canônicas em 2026-08-24: **1.029 vídeos únicos** · **11.132.922 vi
  <tr><td nowrap><b>Runtime</b></td><td>Node.js 22.x / 24.x LTS — <code>&gt;=22.22.2 &lt;23 || &gt;=24.0.0 &lt;27</code></td></tr>
  <tr><td nowrap><b>Language</b></td><td>TypeScript 6.0 — <b>100% TypeScript</b> across <code>src/</code> and <code>open-sse/</code> (zero <code>any</code> in core since v2.0)</td></tr>
  <tr><td nowrap><b>Framework</b></td><td>Next.js 16 + React 19 + Tailwind CSS 4</td></tr>
-  <tr><td nowrap><b>Database</b></td><td>better-sqlite3 (SQLite, WAL journaling) + LowDB (JSON legacy) — 122 domain modules, 169 migrations</td></tr>
+  <tr><td nowrap><b>Database</b></td><td>better-sqlite3 (SQLite, WAL journaling) + LowDB (JSON legacy) — 122 domain modules, 168 migrations</td></tr>
  <tr><td nowrap><b>Memory</b></td><td>SQLite FTS5 full-text + int8-quantized vector embeddings, typed decay</td></tr>
  <tr><td nowrap><b>Schemas</b></td><td>Zod 4 — MCP tool I/O validation + API contracts</td></tr>
  <tr><td nowrap><b>Protocols</b></td><td>MCP (stdio / HTTP / SSE) + A2A v0.3 (JSON-RPC 2.0 + SSE)</td></tr>
--- a/changelog.d/features/11786-seekai-provider.md
+++ b/changelog.d/features/11786-seekai-provider.md
@@ -1 +0,0 @@
- **feat(providers):** add SeekAi (`seekai.cc`) as an OpenAI-compatible New-API gateway — catalog id `seekai` (alias `ska`), `https://seekai.cc/v1`, live `/v1/models` via `passthroughModels`, aggregator-list membership so New-API balance detection can opt in. No referral/aff codes. ([#11786](https://github.com/diegosouzapw/OmniRoute/issues/11786))
--- a/changelog.d/features/12071-csv-import-errors.md
+++ b/changelog.d/features/12071-csv-import-errors.md
@@ -1 +0,0 @@
- **feat(providers):** import-from-file modal shows per-row API errors and ships a downloadable CSV template ([#12071](https://github.com/diegosouzapw/OmniRoute/issues/12071))
--- a/changelog.d/features/12108-provider-search-name-baseurl.md
+++ b/changelog.d/features/12108-provider-search-name-baseurl.md
@@ -1 +0,0 @@
- **feat(providers):** dashboard search matches connection name and `baseUrl` so imported OpenAI-compat nodes surface on the provider card ([#12108](https://github.com/diegosouzapw/OmniRoute/issues/12108))
--- a/changelog.d/features/12306-headroom-url.md
+++ b/changelog.d/features/12306-headroom-url.md
@@ -1 +0,0 @@
- **feat(settings):** persist `headroomUrl` through Settings so status/start use the operator URL instead of only `HEADROOM_URL` ([#12306](https://github.com/diegosouzapw/OmniRoute/issues/12306))
--- a/changelog.d/fixes/12312-grok-cli-supergrok-quota.md
+++ b/changelog.d/fixes/12312-grok-cli-supergrok-quota.md
@@ -1 +0,0 @@
- **fix(grok-cli):** treat omitted SuperGrokPro `creditUsagePercent` as 0% used so Provider Limits still renders a weekly bar (proto3 zero-elision) ([#12312](https://github.com/diegosouzapw/OmniRoute/pull/12312)) — thanks @HouMinXi
--- a/changelog.d/fixes/12325-generic-quota-429-invalidate.md
+++ b/changelog.d/fixes/12325-generic-quota-429-invalidate.md
@@ -1 +0,0 @@
- **fix(quota):** drop the generic quota cache (agy / Antigravity / Claude OAuth) on an upstream 429 so reset-aware scoring does not keep a 60s stale snapshot, and force-refresh the next usage fetch so inner provider caches cannot recache the same window ([#12325](https://github.com/diegosouzapw/OmniRoute/pull/12325)) — thanks @HouMinXi
--- a/changelog.d/maintenance/12402-claude-code-2-1-258.md
+++ b/changelog.d/maintenance/12402-claude-code-2-1-258.md
@@ -1 +0,0 @@
- **chore(providers):** bump the Claude Code wire identity and the Devin bridge image pin from `2.1.220` to `2.1.258` ([#12402](https://github.com/diegosouzapw/OmniRoute/pull/12402)) — thanks @ggiak
--- a/changelog.d/maintenance/hartmark-batch-filesize.md
+++ b/changelog.d/maintenance/hartmark-batch-filesize.md
@@ -1 +0,0 @@
- **chore(quality):** rebaseline the file-size caps the hartmark batch grew past (`combos/page.tsx` via [#12355](https://github.com/diegosouzapw/OmniRoute/pull/12355), `open-sse/services/combo.ts` via [#12338](https://github.com/diegosouzapw/OmniRoute/pull/12338))
--- a/changelog.d/maintenance/houminxi-batch-filesize.md
+++ b/changelog.d/maintenance/houminxi-batch-filesize.md
@@ -1 +0,0 @@
- **chore(quality):** rebaseline the file-size caps the HouMinXi batch grew past when its PRs stacked (`providers/page.tsx`, `chatCore.ts`, `accountFallback.ts`) — each PR measured correctly in isolation, none saw the stacking
--- a/config/quality/file-size-baseline.json
+++ b/config/quality/file-size-baseline.json
@@ -1,7 +1,4 @@
 {
-  "_rebaseline_2026_09_03_moonshot_native_quota": "PR feat/moonshot-native-quota own growth on release/v3.8.51: src/lib/db/migrationRunner.ts 1201->1206 (+5, case 172 retroactive guard for daily_quota_reset_* columns); src/sse/handlers/chat.ts 2434->2450 (+16, registerMoonshotQuotaFetcher + startup node scan at the existing quota-fetcher registration chokepoint); src/sse/services/auth.ts 3427->3450 (+23, resolveDailyResetForProvider + dailyReset arg on checkFallbackError); open-sse/services/accountFallback.ts 2422->2461 (+39, compatible-node credits_exhausted carve-out + TPD node-clock lock); tests/unit/account-fallback-service.test.ts 2008->2056 (+48, TPD/empty-wallet cases). Wiring at existing chokepoints; Moonshot host predicates, daily reset clock, and the balance fetcher live in new leaves under cap. Covered by tests/unit/moonshot-*.test.ts + account-fallback-service.test.ts (135/135 focused).",
-  "_rebaseline_2026_09_02_11786_seekai_provider": "PR #11786 (feat/11786-seekai-provider, closes #11786) own growth: src/shared/constants/providers/apikey/gateways.ts 1438->1458 (check-file-size split-newline=1459; the seekai APIKEY_PROVIDERS_GATEWAYS catalog entry plus authHint, additive data at the existing registry chokepoint, same god-file no-split rationale as prior gateways.ts rebaselines: #10987 logfare, #10531 freebuff). Covered by tests/unit/seekai-provider.test.ts.",
-  "_rebaseline_2026_09_02_12325_generic_429_invalidate": "PR #12325 own growth: open-sse/handlers/chatCore.ts 5946->5955 (+9 = the non-Codex 429 else-if that drops the generic quota wrapper and stamps force-refresh, plus a source-regex breadcrumb). Irreducible call-site wiring next to the existing Codex 429 invalidateCodexQuotaCache branch; not extractable without splitting handleChatCore mid-response. Covered by tests/unit/generic-quota-fetcher.test.ts (31/31) and tests/unit/antigravity-429-quota-cooldown.test.ts.",
  "_rebaseline_2026_09_02_12429_wreq_migration_suite": "PR #12429 (wreq-js web-cookie transport): new test file tests/unit/tls-client-wreq-migration.test.ts at 1374 lines, above the 1200 new-file testCap. Frozen rather than split: it is the single cohesive regression suite for the transport migration (31 cases covering streaming, fragmented EOF sentinels, proxy isolation, first-byte and hard deadlines, binary responses and cancellation), and the cases share the native-transport harness the file sets up once. Splitting it during a merge would duplicate that harness across files for no coverage gain. Entered at the exact LOC, so it can only ratchet down from here.",
  "_rebaseline_2026_09_02_12239_chatgpt_web_cleanroom": "PR #12239 (backryun, codex/restore-chatgpt-web-cleanroom) own growth at the two existing chat chokepoints for the clean-room ChatGPT Web transport: src/sse/handlers/chat.ts 2384->2424 (+40); open-sse/handlers/chatCore.ts 5946->5976 (+30). Additive dispatch wiring; the retirement guard is narrowed to the GPL-derived cgpt-web alias rather than removed, so #11754's provenance decision still holds for the old implementation. Same own-growth rationale as _rebaseline_2026_08_20_10531_freebuff_provider.",
  "_rebaseline_2026_09_02_12412_grok_web_prettier": "PR #12412 (repository Prettier style applied to tests/unit/grok-web.test.ts): the reformat expands the file +277 lines (2436 -> 2713) with an identical parsed AST — no production code, no assertion changes. Cap set to 2985 rather than the exact 2713 on the operator's instruction (2026-09-02): ~10% headroom so routine additions to this suite do not re-trip the gate on formatting alone. Previous cap 2437. This is a deliberate exception to the down-only ratchet for one reformatted test file; every other entry keeps the #12411 tightening.",
@@ -207,7 +204,7 @@
    "_rebaseline_basered_codebuddy_cn": "Base-red fix (#4664 CodeBuddy CN): oauth-providers-config.test.ts 867->870 (+3) to align the EXPECTED provider list/config with the codebuddy-cn provider that #4664 added to the registry without updating this test (it asserts 'exactly once').",
    "_rebaseline_pr4613_compatible_provider_groups": "Reconcile #4613 already-merged growth: providers-page-utils.test.ts 1004->1052 (+48, buildCompatibleProviderGroups partition unit test). Fast-gate PR->release does not run check:file-size, so this surfaced post-merge.",
    "tests/integration/chat-pipeline.test.ts": 1644,
-    "tests/unit/account-fallback-service.test.ts": 2056,
+    "tests/unit/account-fallback-service.test.ts": 2008,
    "tests/unit/batch_api.test.ts": 1345,
    "tests/unit/cc-compatible-provider.test.ts": 1225,
    "tests/unit/chatcore-translation-paths.test.ts": 3447,
@@ -415,14 +412,14 @@
    "open-sse/executors/codex.ts": 1499,
    "open-sse/executors/cursor.ts": 1759,
    "open-sse/executors/muse-spark-web.ts": 1405,
-    "open-sse/handlers/chatCore.ts": 5984,
+    "open-sse/handlers/chatCore.ts": 5976,
    "open-sse/handlers/imageGeneration.ts": 3259,
    "open-sse/handlers/search.ts": 1789,
    "open-sse/mcp-server/schemas/tools.ts": 1621,
    "open-sse/mcp-server/server.ts": 1572,
-    "open-sse/services/accountFallback.ts": 2467,
+    "open-sse/services/accountFallback.ts": 2422,
    "open-sse/services/adobeFireflyBrowserLogin.ts": 1401,
-    "open-sse/services/combo.ts": 4036,
+    "open-sse/services/combo.ts": 4023,
    "open-sse/translator/response/openai-responses.ts": 1466,
    "open-sse/utils/cursorAgentProtobuf.ts": 1547,
    "open-sse/utils/proxyFetch.ts": 1271,
@@ -431,15 +428,15 @@
    "open-sse/vendor/codex-chatgpt-web/bridge.ts": 1322,
    "src/app/(dashboard)/dashboard/HomePageClient.tsx": 1344,
    "src/app/(dashboard)/dashboard/api-manager/ApiManagerPageClient.tsx": 3186,
-    "src/app/(dashboard)/dashboard/combos/page.tsx": 5018,
+    "src/app/(dashboard)/dashboard/combos/page.tsx": 5012,
    "src/app/(dashboard)/dashboard/costs/CostOverviewTab.tsx": 1319,
    "src/app/(dashboard)/dashboard/endpoint/EndpointPageClient.tsx": 2491,
    "src/app/(dashboard)/dashboard/providers/[id]/components/modals/EditConnectionModal.tsx": 1631,
-    "src/app/(dashboard)/dashboard/providers/page.tsx": 2025,
+    "src/app/(dashboard)/dashboard/providers/page.tsx": 2007,
    "src/app/(dashboard)/dashboard/runtime/RuntimePageClient.tsx": 1201,
    "src/app/(dashboard)/dashboard/settings/components/ProxyRegistryManager.tsx": 1475,
    "src/app/(dashboard)/dashboard/settings/components/ResilienceTab.tsx": 1271,
-    "src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx": 1607,
+    "src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx": 1606,
    "src/app/(dashboard)/dashboard/settings/components/SystemStorageTab.tsx": 1597,
    "src/app/(dashboard)/dashboard/usage/components/EvalsTab.tsx": 2152,
    "src/app/api/providers/[id]/models/route.ts": 2432,
@@ -448,14 +445,14 @@
    "src/app/docs/lib/openapi.generated.ts": 1347,
    "src/lib/db/apiKeys.ts": 1610,
    "src/lib/db/core.ts": 1745,
-    "src/lib/db/migrationRunner.ts": 1206,
+    "src/lib/db/migrationRunner.ts": 1201,
    "src/lib/tailscaleTunnel.ts": 1208,
    "src/lib/tokenHealthCheck.ts": 1218,
    "src/shared/components/RequestLoggerV2.tsx": 1718,
-    "src/shared/constants/providers/apikey/gateways.ts": 1459,
+    "src/shared/constants/providers/apikey/gateways.ts": 1439,
    "src/shared/services/cliRuntime.ts": 1296,
-    "src/sse/handlers/chat.ts": 2450,
-    "src/sse/services/auth.ts": 3450,
+    "src/sse/handlers/chat.ts": 2424,
+    "src/sse/services/auth.ts": 3427,
    "tests/unit/account-fallback-service.test.ts": 2453,
    "tests/unit/provider-validation-specialty.test.ts": 4656
  },
@@ -632,9 +629,5 @@
  "_rebaseline_2026_06_30_v3842_release_chatgptweb_compression": "v3.8.42 cycle-close file-size reconciliation (DRIFT measured OK on each PR's base, stacked above frozen at the merge tip; fast-path PR->release/** does not run check:file-size). (1) open-sse/executors/chatgpt-web.ts 2870->3206 (+336 = #5531 portable SHA3-512 sentinel-PoW wiring with the native-vs-fallback digest path + #5536 GPT-5.5 Pro handoff branch; the pure Keccak-f[1600] fallback itself already lives in the separate leaf open-sse/utils/sha3-512.ts — the executor growth is the cohesive call-site/handoff logic, not extractable without hiding the sentinel chokepoint). (2) tests/unit/chatgpt-web.test.ts 2855->3159 (+304 = #5536 GPT-5.5 Pro handoff coverage; pair-file with its executor). (3) open-sse/services/compression/strategySelector.ts 997->1022 (+25 = #5527 T02 honest default-on pipeline inflation guard wiring at the existing finalizeStackedResult choke). All cohesive at existing chokepoints; covered by tests/unit/chatgpt-web-sha3-boringssl-5531.test.ts, chatgpt-web.test.ts (GPT-5.5 Pro), compression-pipeline-inflation-guard.test.ts.",
  "open-sse/executors/chatgpt-web.ts": "3241",
  "_rebaseline_2026_08_30_11771_vercel_gateway_passthrough": "PR #11771 adds passthroughModels: true (1 line) to the Vercel AI Gateway registry entry — no split available, single-line provider-config addition.",
-  "_relax_velocity_2026_08_30": "127 frozen line caps and cap/testCap raised by 20% (velocity phase; see quality-baseline.json _policy).",
-  "_rebaseline_2026_09_02_12325_merge_v3851": "Merge of release/v3.8.51 into #12325. Both sides grew chatCore.ts at the same chokepoint: #12239 took it 5946->5976 upstream, and this PR adds its +9 non-Codex 429 branch on top. check-file-size.mjs counts split(\"\\\\n\").length (trailing-newline empty element), so the merged file is 5981. The cap is the merged LOC, not either side alone; no other entry moves.",
-  "_rebaseline_2026_09_03_houminxi_batch_stacked": "Crescimento medido DEPOIS que os 9 PRs da leva HouMinXi entraram, quando cada um empilhou sobre o rebaseline do anterior: providers/page.tsx 2007->2025 (+18 = feedback de erro por linha do import CSV do #12504 somado a busca por nome/baseUrl do #12495, ambos no mesmo painel de conexoes); chatCore.ts 5981->5984 (+3 = o #12325 invalida o cache generico de quota no 429 upstream, ao lado do ramo Codex ja existente); accountFallback.ts 2461->2467 (+6 = o #12566 empilha a carve-out de familia Antigravity sobre o rebaseline 2422->2461 que o #12590 registrou para o carve-out credits_exhausted da Moonshot; os dois tocam checkFallbackError). Cada PR mediu certo isoladamente, mas nenhum enxergava o empilhamento. Fiacao em chokepoints existentes. NAO cobre codex.ts nem stream.ts, que ja violavam no tip antes desta leva (drift da base).",
-  "_rebaseline_2026_09_03_12604_claude_code_2_1_258": "PR #12604 (bump da wire identity do Claude Code 2.1.220->2.1.258, commits do @ggiak vindos do #12402) crescimento proprio: src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx 1606->1607 (+1, a linha do seletor que acompanha a nova versao de identidade). Uma linha num painel de settings ja existente; nao ha o que extrair. Coberto por client-identity-profiles e claude-codex-identity-version-sync (138/138 focados).",
-  "_rebaseline_2026_09_03_hartmark_batch": "Leva hartmark (#12293 #12355 #12447 #12445 #12446 #12460 #12461 #12338 #12448) crescimento proprio, medido no tip com os nove mergeados: src/app/(dashboard)/dashboard/combos/page.tsx 5012->5018 (+6, #12355 impede que a falha de bundling do tiktoken de um provider sem relacao derrube /api/providers, e o painel passa a lidar com o estado degradado); open-sse/services/combo.ts 4023->4036 (+13, #12338 nos fixes do universal-handoff: nota de bare-fallback, escopo por mesma requisicao e log da falha silenciosa). Fiacao em chokepoints existentes do roteamento de combo. NAO cobre codex.ts nem stream.ts, ja violando no tip antes desta leva (drift da base)."
+  "_relax_velocity_2026_08_30": "127 frozen line caps and cap/testCap raised by 20% (velocity phase; see quality-baseline.json _policy)."
 }
--- a/docker/devin-bridge/Dockerfile
+++ b/docker/devin-bridge/Dockerfile
@@ -1,6 +1,6 @@
 FROM node:26.0.0-bookworm-slim

-ARG CLAUDE_CODE_VERSION=2.1.258
+ARG CLAUDE_CODE_VERSION=2.1.220
 ARG DEVIN_CLI_VERSION=3000.2.17
 ARG TARGETARCH

--- a/docker/devin-bridge/compose.yml
+++ b/docker/devin-bridge/compose.yml
@@ -28,7 +28,7 @@ x-runtime: &runtime
    context: ../..
    dockerfile: docker/devin-bridge/Dockerfile
    args:
-      CLAUDE_CODE_VERSION: 2.1.258
+      CLAUDE_CODE_VERSION: 2.1.220
      DEVIN_CLI_VERSION: 3000.2.17
  user: "10001:10001"
  read_only: true
--- a/docs/DEVIN_CLAUDE_BRIDGE.md
+++ b/docs/DEVIN_CLAUDE_BRIDGE.md
@@ -4,22 +4,16 @@
 Messages endpoint while the official Devin CLI supplies model responses over ACP stdio. It
 does not modify the existing Anthropic, Claude OAuth, Claude Web, or `devin-cli` providers.

-> **Current status: pinned Claude Code `2.1.258`; offline and live validation last recorded
-> on `2.1.220`.** The `2.1.220` pin completed three isolated scenarios through Devin CLI
-> `3000.2.17` and model `swe-1-7-lightning`; that final live run proved client-owned `Read`,
-> `Edit`, and `Bash` turns, successful `npm test` results, project command and skill
-> discovery, Devin-only routing, and zero Claude egress. The pin was then raised to `2.1.258`
-> (the CLI generation OmniRoute's Claude identity impersonates, and the first line that ships
-> the Fable 5.1 tier natively). On the new pin the install layer and `claude --version` were
-> verified on the pinned base image, and the bridge unit suite, `compose config` and the
-> static isolation proof pass — but the offline mock scenario and the live three-scenario
-> suite have not been re-run yet. Re-run them (see "Updating pinned tools") before relying on
-> the bridge with this pin.
+> **Current status: offline and live validated.** The pinned Claude Code `2.1.220` completed
+> three isolated scenarios through Devin CLI `3000.2.17` and model
+> `swe-1-7-lightning`. The final live run proved client-owned `Read`, `Edit`, and `Bash`
+> turns, successful `npm test` results, project command and skill discovery, Devin-only
+> routing, and zero Claude egress.

 ## Architecture

 ```text
-Claude Code 2.1.258 (isolated non-root Linux container)
+Claude Code 2.1.220 (isolated non-root Linux container)
  -> http://omniroute:20128/v1/messages
  -> devin-cli-agentic (Claude-format, no-auth provider)
  -> devin acp --agent-type summarizer (official ACP stdio, no Devin tools)
--- a/docs/architecture/QUALITY_GATES.md
+++ b/docs/architecture/QUALITY_GATES.md
@@ -90,17 +90,17 @@ Runs on every PR to `main`. Blocks merge on failure.

 Runs after `test-coverage`. Blocks merge on failure.

-| Script                       | Validates                                                                                                                                         | Blocking                  |
-| ---------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------- |
-| `quality:collect`            | Emits `quality-metrics.json` (ESLint warning count, coverage from merged shard report)                                                            | Yes (upstream of ratchet) |
-| `quality:ratchet`            | Each metric in `quality-baseline.json` has not regressed (ESLint warnings ≤ baseline; coverage ≥ baseline)                                        | Yes                       |
-| `check:duplication`          | Code duplication (jscpd@4) does not exceed baseline in `quality-baseline.json`                                                                    | Yes                       |
-| `check:complexity`           | File-level cyclomatic complexity does not exceed the cap (core ESLint `complexity` + `max-lines-per-function`)                                    | Yes                       |
-| `check:cognitive-complexity` | Cognitive complexity ratchet (`eslint-plugin-sonarjs`) — separate ESLint pass; CI runs both merged as the single `check:complexity-ratchets` step | Yes                       |
-| `check:dead-code`            | Unused exports / files ratchet (knip) does not regress vs baseline                                                                                | Yes                       |
-| `check:compression-budget`   | Compression benchmark budget — per-engine token-savings floors must not regress                                                                   | Yes                       |
-| `check:type-coverage`        | Percent-typed ratchet (`type-coverage`) does not regress; largely subsumes `typecheck:noimplicit:core`                                            | Yes                       |
-| `check:codeql-ratchet`       | Open CodeQL alert count does not regress (reads via `gh api`; graceful-skip without token)                                                        | Yes                       |
+| Script                       | Validates                                                                                                                                                   | Blocking                  |
+| ---------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------- |
+| `quality:collect`            | Emits `quality-metrics.json` (ESLint warning count, coverage from merged shard report)                                                                      | Yes (upstream of ratchet) |
+| `quality:ratchet`            | Each metric in `quality-baseline.json` has not regressed (ESLint warnings ≤ baseline; coverage ≥ baseline)                                                  | Yes                       |
+| `check:duplication`          | Code duplication (jscpd@4) does not exceed baseline in `quality-baseline.json`                                                                              | Yes                       |
+| `check:complexity`           | File-level cyclomatic complexity does not exceed the cap (core ESLint `complexity` + `max-lines-per-function`)                                              | Yes                       |
+| `check:cognitive-complexity` | Cognitive complexity ratchet (`eslint-plugin-sonarjs`) — separate ESLint pass; CI runs both merged as the single `check:complexity-ratchets` step           | Yes                       |
+| `check:dead-code`            | Unused exports / files ratchet (knip) does not regress vs baseline                                                                                          | Yes                       |
+| `check:compression-budget`   | Compression benchmark budget — per-engine token-savings floors must not regress                                                                             | Yes                       |
+| `check:type-coverage`        | Percent-typed ratchet (`type-coverage`) does not regress; largely subsumes `typecheck:noimplicit:core`                                                      | Yes                       |
+| `check:codeql-ratchet`       | Open CodeQL alert count does not regress (reads via `gh api`; graceful-skip without token) — refresh cadence and manual trigger: see "CodeQL ratchet" below | Yes                       |

 ### Job: `quality-extended`

@@ -324,6 +324,36 @@ Commit this file alongside the change that improved the metric. A PR that improv
 metric without updating the baseline will be caught by `--require-tighten` (Fase 6A.5,
 pending implementation).

+### CodeQL ratchet: refresh cadence and manual trigger
+
+`check:codeql-ratchet` reads **repo state, refreshed on a schedule — not per PR.**
+`gh api repos/diegosouzapw/OmniRoute/code-scanning/default-setup` reports
+`state: configured`, `schedule: weekly`: GitHub's default-setup scan, not a per-push
+analysis. Consequence: after a PR that FIXES alerts merges, the ratchet keeps reading
+the old, higher count until the next scheduled scan runs — so it reports a regression
+on every open PR, including the fixing PR's own follow-ups, until the scan catches up.
+
+**Manual refresh**: `gh workflow run codeql.yml --ref release/vX.Y.Z` re-runs the
+analysis and republishes alerts within minutes. Read `.github/workflows/codeql.yml`
+first — its header explains it is `workflow_dispatch`-only **because it conflicts with
+GitHub's "default setup"** (`CodeQL analyses from advanced configurations cannot be
+processed when the default setup is enabled`). Restoring `push`/`pull_request`/
+`schedule` triggers requires an **owner action first**: Settings → Code security →
+CodeQL: Default → Advanced. Do not add a `schedule:` trigger without that switch — it
+will only produce failing runs.
+
+**Tighten the baseline after the count drops** — `node scripts/check/check-codeql-ratchet.mjs
+--update` writes the new measured count into `quality-baseline.json` →
+`metrics.codeqlAlerts.value`, so the ratchet does not silently permit a regression back
+up to the old ceiling. Worked example (2026-09-02/03): PR #12502 fixed 7 real alerts
+(13 → 6 measured open); PR #12530 tightened the frozen baseline 11 → 6 to match; the
+remaining 6 were then dismissed with per-alert justification down to 0 open.
+
+**Dismissals are the operator's call (Hard Rule #14)** — never dismiss a CodeQL alert
+without recording the technical justification in the dismissal comment: `won't fix` for
+an upstream-protocol requirement, `used in tests` for a test fixture, `false positive`
+for a sanitizer CodeQL cannot see (precedent: `docs/security/ERROR_SANITIZATION.md`).
+
 ---

 ## Test Retry Policy (WS5.4, v3.8.49)
--- a/docs/compression/COMPRESSION_GUIDE.md
+++ b/docs/compression/COMPRESSION_GUIDE.md
@@ -309,7 +309,7 @@ Every compressed request includes stats in the server logs:
 | Phase 2  | Standard, Aggressive, Ultra                                                                                  | ✅ Shipped                                                             |
 | Phase 3  | RTK, Stacked, Compression Combos                                                                             | ✅ Shipped                                                             |
 | Phase 4  | Output Styles, SLM-tier Ultra, eval harness                                                                  | ✅ Shipped                                                             |
-| Phase 4C | Adaptive context-budget ("dial") — compute engine + API (`contextBudget` on `PUT /api/settings/compression`) + dashboard mode/policy controls | ✅ Shipped |
+| Phase 4C | Adaptive context-budget ("dial") — compute engine + API (`contextBudget` on `PUT /api/settings/compression`) | ✅ Shipped (API-configurable; dashboard controls not yet built, #7005) |

 ---

--- a/docs/diagrams/cli-terminal.svg
+++ b/docs/diagrams/cli-terminal.svg
@@ -1,4 +1,4 @@
-<svg viewBox="0 0 1200 350" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Animated terminal demoing the OmniRoute CLI: omniroute providers list (356 providers registered, anthropic, codex, glm, kimi shown active), omniroute combo list (always-on priority, cost-saver, fusion-panel, context-relay) and omniroute health (healthy, 18412 requests in 24h, p95 412ms, circuit breakers 24 closed, 1 half-open, 0 open), cycling over 86 top-level commands: providers, oauth, keys, combo, nodes, models, cache, compression, cost, usage, quota, health, resilience, telemetry, logs, audit, mcp, a2a, cloud, memory, skills, eval, doctor, repl, tunnel, backup, sync, webhooks, policy, pricing, translator, simulate and more.">
+<svg viewBox="0 0 1200 350" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Animated terminal demoing the OmniRoute CLI: omniroute providers list (355 providers registered, anthropic, codex, glm, kimi shown active), omniroute combo list (always-on priority, cost-saver, fusion-panel, context-relay) and omniroute health (healthy, 18412 requests in 24h, p95 412ms, circuit breakers 24 closed, 1 half-open, 0 open), cycling over 86 top-level commands: providers, oauth, keys, combo, nodes, models, cache, compression, cost, usage, quota, health, resilience, telemetry, logs, audit, mcp, a2a, cloud, memory, skills, eval, doctor, repl, tunnel, backup, sync, webhooks, policy, pricing, translator, simulate and more.">
 <desc>Compact animated terminal cycling three real OmniRoute CLI commands with a typewriter effect and a scrolling subcommand ticker; the first frame shows the completed providers-list screen.</desc>
 <defs><clipPath id="tickerClip"><rect x="12" y="304" width="1176" height="40"/></clipPath><clipPath id="tw0"><rect x="64" y="46" height="26" width="0"><animate attributeName="width" calcMode="discrete" values="0;31;61;92;122;153;184;214;245;245" keyTimes="0;0.012;0.018;0.024;0.030;0.036;0.042;0.048;0.054;1" dur="18s" repeatCount="indefinite"/></rect></clipPath><clipPath id="tw1"><rect x="64" y="46" height="26" width="0"><animate attributeName="width" calcMode="discrete" values="0;26;51;76;102;128;153;178;204;204" keyTimes="0;0.348;0.351;0.357;0.363;0.369;0.375;0.381;0.387;1" dur="18s" repeatCount="indefinite"/></rect></clipPath><clipPath id="tw2"><rect x="64" y="46" height="26" width="0"><animate attributeName="width" calcMode="discrete" values="0;20;41;61;82;102;122;143;163;163" keyTimes="0;0.678;0.684;0.690;0.696;0.702;0.708;0.714;0.720;1" dur="18s" repeatCount="indefinite"/></rect></clipPath></defs>
 <rect width="1200" height="350" fill="#0d1117"/>
--- a/docs/diagrams/comparison-table.svg
+++ b/docs/diagrams/comparison-table.svg
@@ -1,4 +1,4 @@
-<svg viewBox="0 0 1200 780" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Comparison table: OmniRoute versus 9router, OpenRouter, CLIProxyAPI and LiteLLM across 13 capabilities. OmniRoute is the only one with the full set: 356 providers, 150+ free providers built-in, 19 routing strategies, 12-engine token compression, a built-in MCP server with 110 tools, A2A protocol, persistent memory, guardrails, cloud agents, TLS fingerprint stealth, desktop/Termux/PWA, 42 UI locales and 100% MIT self-hosted. 9router has free providers, RTK compression and translation but no MCP, A2A, memory, guardrails, cloud agents or stealth. OpenRouter is a hosted SaaS with 400+ models, guardrails and a hosted MCP but is not self-hosted and lacks A2A, memory, cloud agents and stealth. CLIProxyAPI is a light OAuth proxy with two routing strategies. LiteLLM has 100+ providers, A2A and extensive guardrails but no memory, compression, free tier, stealth or cloud agents.">
+<svg viewBox="0 0 1200 780" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Comparison table: OmniRoute versus 9router, OpenRouter, CLIProxyAPI and LiteLLM across 13 capabilities. OmniRoute is the only one with the full set: 355 providers, 150+ free providers built-in, 19 routing strategies, 12-engine token compression, a built-in MCP server with 110 tools, A2A protocol, persistent memory, guardrails, cloud agents, TLS fingerprint stealth, desktop/Termux/PWA, 42 UI locales and 100% MIT self-hosted. 9router has free providers, RTK compression and translation but no MCP, A2A, memory, guardrails, cloud agents or stealth. OpenRouter is a hosted SaaS with 400+ models, guardrails and a hosted MCP but is not self-hosted and lacks A2A, memory, cloud agents and stealth. CLIProxyAPI is a light OAuth proxy with two routing strategies. LiteLLM has 100+ providers, A2A and extensive guardrails but no memory, compression, free tier, stealth or cloud agents.">
  <desc>Static-header comparison table where each capability row fades in top to bottom; the OmniRoute column is highlighted and shows a check or a leading value in every row, while competitors show a mix of checks, partials and crosses.</desc>
  <defs>
    <pattern id="gC" width="32" height="32" patternUnits="userSpaceOnUse"><path d="M 32 0 L 0 0 0 32" fill="none" stroke="#ffffff" stroke-opacity="0.05" stroke-width="1"/></pattern>
--- a/docs/diagrams/promise-pillars.svg
+++ b/docs/diagrams/promise-pillars.svg
@@ -1,4 +1,4 @@
-<svg viewBox="0 0 1200 540" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="The OmniRoute promise: one endpoint and 356 providers. Six pillars. Resilient fallback: automatic routing continues while another healthy target is available. Save up to 95 percent of eligible tokens: RTK plus Caveman stacked compression averages about 89 percent on tool-heavy sessions. Zero dollars to start: 150+ providers with a free tier and 53 recurring or keyless free-forever providers. Every tool works: 36 CLI and agent integration records, including Claude Code, Codex, Cursor, Cline, Copilot and Antigravity, through one config. One endpoint: OpenAI, Claude, Gemini and Responses API translation at /v1. Production controls: circuit breakers, TLS stealth, MCP with 110 tools, A2A, memory, guardrails, evals, and 39,000+ static test declarations across 5,100+ tracked test files.">
+<svg viewBox="0 0 1200 540" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="The OmniRoute promise: one endpoint and 355 providers. Six pillars. Resilient fallback: automatic routing continues while another healthy target is available. Save up to 95 percent of eligible tokens: RTK plus Caveman stacked compression averages about 89 percent on tool-heavy sessions. Zero dollars to start: 150+ providers with a free tier and 53 recurring or keyless free-forever providers. Every tool works: 36 CLI and agent integration records, including Claude Code, Codex, Cursor, Cline, Copilot and Antigravity, through one config. One endpoint: OpenAI, Claude, Gemini and Responses API translation at /v1. Production controls: circuit breakers, TLS stealth, MCP with 110 tools, A2A, memory, guardrails, evals, and 39,000+ static test declarations across 5,100+ tracked test files.">
  <desc>Animated promise card: six pillar tiles fade in in reading order, then a soft colored border highlight sweeps from tile to tile in a continuous cycle.</desc>
  <defs>
    <pattern id="gridPaperP" width="32" height="32" patternUnits="userSpaceOnUse">
@@ -21,7 +21,7 @@
    <line x1="150" y1="53" x2="1160" y2="53" stroke="#232b38" stroke-width="1.5"/>
  </g>
  <g>
-    <text x="40" y="100" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="23" font-weight="600" fill="#c9d1d9">One endpoint. <tspan fill="#a78bfa" font-weight="800">356 providers.</tspan> Never stop building — OmniRoute picks <tspan fill="#7ee787" font-weight="700">the cheapest one that works</tspan>.</text>
+    <text x="40" y="100" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="23" font-weight="600" fill="#c9d1d9">One endpoint. <tspan fill="#a78bfa" font-weight="800">355 providers.</tspan> Never stop building — OmniRoute picks <tspan fill="#7ee787" font-weight="700">the cheapest one that works</tspan>.</text>
  </g>

  <g font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif">
@@ -38,7 +38,7 @@
        <line x1="3.9" y1="3.9" x2="18.1" y2="18.1"/>
      </g>
      <text x="102" y="170" font-size="18" font-weight="800" fill="#74b9ff">Never hit limits</text>
-      <text x="66" y="204" font-size="13.5" fill="#a1a1aa">Auto-fallback across 356 providers in</text>
+      <text x="66" y="204" font-size="13.5" fill="#a1a1aa">Auto-fallback across 355 providers in</text>
      <text x="66" y="226" font-size="13.5" fill="#a1a1aa">milliseconds. Quota out? The next provider</text>
      <text x="66" y="248" font-size="13.5" fill="#a1a1aa">takes over while a healthy target remains.</text>
    </g>
--- a/docs/diagrams/readme-hero.svg
+++ b/docs/diagrams/readme-hero.svg
@@ -1,4 +1,4 @@
-<svg viewBox="0 0 1200 548" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="OmniRoute hero: Never stop coding. Every AI tool to 356 providers — 150+ free — through one endpoint. Claude Code, Codex, Cursor, Cline, Copilot and Antigravity into free Claude, GPT and Gemini with auto-fallback. RTK + Caveman stacked compression saves 15 to 95 percent of tokens — about 89 percent average on tool-heavy sessions — so you never hit limits. Stats: 356 AI providers, 150+ free tiers, about 1.51B free tokens per month, 15 to 95 percent token savings, 19 routing strategies, zero dollars to start.">
+<svg viewBox="0 0 1200 548" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="OmniRoute hero: Never stop coding. Every AI tool to 355 providers — 150+ free — through one endpoint. Claude Code, Codex, Cursor, Cline, Copilot and Antigravity into free Claude, GPT and Gemini with auto-fallback. RTK + Caveman stacked compression saves 15 to 95 percent of tokens — about 89 percent average on tool-heavy sessions — so you never hit limits. Stats: 355 AI providers, 150+ free tiers, about 1.51B free tokens per month, 15 to 95 percent token savings, 19 routing strategies, zero dollars to start.">
  <desc>Animated hero card: a pulse travels the divider line and a compression bar demo repeatedly shrinks a prompt by up to 95 percent; all headline content is static and readable on the first frame.</desc>
  <defs>
    <pattern id="gridPaperH" width="32" height="32" patternUnits="userSpaceOnUse">
@@ -28,7 +28,7 @@
  <text x="48" y="138" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="60" font-weight="800" fill="#e9edf3">Never stop coding<tspan fill="#a855f7">.</tspan></text>

  <!-- subheadline -->
-  <text x="48" y="184" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="25" font-weight="600" fill="#c9d1d9">Every AI tool → <tspan fill="#a78bfa" font-weight="800">356 providers</tspan> — <tspan fill="#7ee787" font-weight="800">150+ free</tspan> — through one endpoint.</text>
+  <text x="48" y="184" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="25" font-weight="600" fill="#c9d1d9">Every AI tool → <tspan fill="#a78bfa" font-weight="800">355 providers</tspan> — <tspan fill="#7ee787" font-weight="800">150+ free</tspan> — through one endpoint.</text>

  <!-- plug line -->
  <text x="48" y="222" font-family="Inter, 'Segoe UI', Arial, Helvetica, system-ui, sans-serif" font-size="16.5" fill="#a1a1aa">Claude Code · Codex · Cursor · Cline · Copilot · Antigravity&#160;&#160;→&#160;&#160;<tspan fill="#7ee787" font-weight="700">FREE</tspan> Claude / GPT / Gemini · auto-fallback</text>
--- a/docs/guides/USER_GUIDE.md
+++ b/docs/guides/USER_GUIDE.md
@@ -122,8 +122,6 @@ Access via: WhatsApp, Telegram, Slack, Discord, iMessage, Signal...

 ## 📖 Provider Setup

-To bulk-add API-key connections from a CSV or JSON file, use **Dashboard → Providers → Import from file**. Columns are positional (`provider,name,apiKey,baseUrl,priority`); `provider` must already exist as a managed provider or a compatible node. See [Import providers from a CSV or JSON file](../providers/CSV-IMPORT.md).
-
 ### 🔐 Subscription Providers

 #### Claude Code (Pro/Max)
--- a/docs/i18n/ar/llm.txt
+++ b/docs/i18n/ar/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/az/llm.txt
+++ b/docs/i18n/az/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/bg/llm.txt
+++ b/docs/i18n/bg/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/bn/llm.txt
+++ b/docs/i18n/bn/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/cs/llm.txt
+++ b/docs/i18n/cs/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/da/llm.txt
+++ b/docs/i18n/da/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/de/llm.txt
+++ b/docs/i18n/de/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/es/llm.txt
+++ b/docs/i18n/es/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/fa/llm.txt
+++ b/docs/i18n/fa/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/fi/llm.txt
+++ b/docs/i18n/fi/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/fr/llm.txt
+++ b/docs/i18n/fr/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/gu/llm.txt
+++ b/docs/i18n/gu/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/he/llm.txt
+++ b/docs/i18n/he/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/hi/llm.txt
+++ b/docs/i18n/hi/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/hu/llm.txt
+++ b/docs/i18n/hu/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/id/llm.txt
+++ b/docs/i18n/id/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/it/llm.txt
+++ b/docs/i18n/it/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ja/llm.txt
+++ b/docs/i18n/ja/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ko/llm.txt
+++ b/docs/i18n/ko/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/mr/llm.txt
+++ b/docs/i18n/mr/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ms/llm.txt
+++ b/docs/i18n/ms/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/nl/llm.txt
+++ b/docs/i18n/nl/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/no/llm.txt
+++ b/docs/i18n/no/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/phi/llm.txt
+++ b/docs/i18n/phi/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/pl/llm.txt
+++ b/docs/i18n/pl/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/pt-BR/llm.txt
+++ b/docs/i18n/pt-BR/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/pt/llm.txt
+++ b/docs/i18n/pt/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ro/llm.txt
+++ b/docs/i18n/ro/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ru/llm.txt
+++ b/docs/i18n/ru/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/sk/llm.txt
+++ b/docs/i18n/sk/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/sv/llm.txt
+++ b/docs/i18n/sv/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/sw/llm.txt
+++ b/docs/i18n/sw/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ta/llm.txt
+++ b/docs/i18n/ta/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/te/llm.txt
+++ b/docs/i18n/te/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/th/llm.txt
+++ b/docs/i18n/th/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/tr/llm.txt
+++ b/docs/i18n/tr/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/uk-UA/llm.txt
+++ b/docs/i18n/uk-UA/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/ur/llm.txt
+++ b/docs/i18n/ur/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/vi/llm.txt
+++ b/docs/i18n/vi/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/zh-CN/llm.txt
+++ b/docs/i18n/zh-CN/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/i18n/zh-TW/llm.txt
+++ b/docs/i18n/zh-TW/llm.txt
@@ -4,7 +4,7 @@

 ---

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -18,7 +18,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -128,7 +128,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -281,7 +281,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -393,7 +393,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -437,7 +437,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/docs/providers/AGENTROUTER.md
+++ b/docs/providers/AGENTROUTER.md
@@ -130,7 +130,7 @@ request (see `open-sse/services/claudeCodeCompatible.ts`):
 | Header                                      | Value                                                                                                   |
 | ------------------------------------------- | ------------------------------------------------------------------------------------------------------- |
 | `Authorization`                             | `Bearer <api-key>`                                                                                      |
-| `User-Agent`                                | `claude-cli/2.1.258 (external, sdk-cli)`                                                                |
+| `User-Agent`                                | `claude-cli/2.1.219 (external, sdk-cli)`                                                                |
 | `anthropic-version`                         | `2023-06-01`                                                                                            |
 | `anthropic-beta`                            | `claude-code-20250219,interleaved-thinking-2025-05-14,effort-2025-11-24`                                |
 | Per-connection redact-thinking beta toggle  | Adds `redact-thinking-2026-02-12` for upstreams that specifically require redacted thinking streams     |
--- a/docs/providers/CSV-IMPORT.md
+++ b/docs/providers/CSV-IMPORT.md
@@ -1,43 +0,0 @@
---
-title: "Import providers from a CSV or JSON file"
---
-
-# Import providers from a CSV or JSON file
-
-Dashboard → Providers → **Import from file** creates API-key connections from a CSV or JSON list. Each row can target a different provider. Partial failure is the contract: valid rows still import when others fail, and the modal lists why the failed rows were rejected.
-
-This import does **not** create new OpenAI/Anthropic-compatible endpoint nodes. Create those first (Dashboard → Providers → Add OpenAI-Compatible, or `omniroute nodes add`), then import rows whose `provider` column is that node's id. A per-row `baseUrl` can still override the node's URL.
-
-## CSV (positional)
-
-Column names are cosmetic. The parser splits each row and destructures by index:
-
-| Index | Field | Required | Notes |
-| ----- | ----- | -------- | ----- |
-| 0 | `provider` | yes | Existing managed provider id (`openai`, `anthropic`, …) **or** an already-registered OpenAI/Anthropic-compatible **node** id |
-| 1 | `name` | yes | Connection display name |
-| 2 | `apiKey` | yes | API key |
-| 3 | `baseUrl` | no | Per-row URL override |
-| 4 | `priority` | no | Integer 1–100 |
-
-A first line whose first column is the literal word `provider` (any case) is skipped as a header. Blank lines and `#` comments are skipped.
-
-Download a starter file from the import modal (**Download CSV template**). Example:
-
-```csv
-# OmniRoute provider import (positional columns)
-provider,name,apiKey,baseUrl,priority
-openai,Prod OpenAI,sk-your-openai-key,,1
-```
-
-A made-up id such as `openai-compatible-chat-001` is not a node. The API returns `Unknown or unsupported provider` for that row; the modal shows it next to the row name.
-
-## JSON
-
-A JSON array of objects with the same fields (`provider`, `name`, `apiKey`, `baseUrl?`, `priority?`). Unlike CSV, JSON keys are named.
-
-```json
-[
-  { "provider": "openai", "name": "Prod OpenAI", "apiKey": "sk-your-openai-key", "priority": 1 }
-]
-```
--- a/docs/providers/meta.json
+++ b/docs/providers/meta.json
@@ -8,7 +8,6 @@
    "AGENTROUTER",
    "ZED-DOCKER",
    "CURSOR-DOCKER",
-    "CURSOR-API-KEY-AND-CLI",
-    "CSV-IMPORT"
+    "CURSOR-API-KEY-AND-CLI"
  ]
 }
--- a/docs/reference/ENVIRONMENT.md
+++ b/docs/reference/ENVIRONMENT.md
@@ -622,7 +622,7 @@ process.env[`${PROVIDER_ID}_USER_AGENT`]

 | Variable                         | Default Value                                 | When to Update                                                                                   |
 | -------------------------------- | --------------------------------------------- | ------------------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
-| `CLAUDE_USER_AGENT`              | `claude-cli/2.1.258 (external, cli)`          | When Anthropic releases a new CLI version                                                        |
+| `CLAUDE_USER_AGENT`              | `claude-cli/2.1.219 (external, cli)`          | When Anthropic releases a new CLI version                                                        |
 | `CLAUDE_DISABLE_TOOL_NAME_CLOAK` | `false`                                       | `executors/base.ts` + `executors/cliproxyapi.ts`                                                 | Set to `1`/`true` to forward third-party harness tool names verbatim to Anthropic on both Anthropic-bound paths (native OAuth and CLIProxyAPI). By default the executor deterministically aliases non-Claude-Code tool names (Claude Code canonical mapping where one exists, otherwise PascalCase) and reverses them on the response via `_toolNameMap`, so harnesses with snake_case tools are not refused as fingerprinted third-party clients. Debugging only. |
 | `CODEX_USER_AGENT`               | `codex-cli/0.142.0 (Windows 10.0.26200; x64)` | When OpenAI updates the Codex CLI                                                                |
 | `CODEX_CLIENT_VERSION`           | `0.131.0`                                     | Override Codex client version independently of full UA string                                    |
--- a/docs/reference/PROVIDER_REFERENCE.md
+++ b/docs/reference/PROVIDER_REFERENCE.md
@@ -1,16 +1,16 @@
 ---
 title: "Provider Reference"
 version: 3.8.51
-lastUpdated: 2026-09-03
+lastUpdated: 2026-09-02
 ---

 # Provider Reference

 > **Auto-generated** from `src/shared/constants/providers.ts` — do not edit by hand.
 > Regenerate with: `npm run gen:provider-reference`
-> **Last generated:** 2026-09-03
+> **Last generated:** 2026-09-02

-Total providers: **356**. See category breakdown below.
+Total providers: **355**. See category breakdown below.

 ## Categories

@@ -118,7 +118,7 @@ Use the dashboard at `/dashboard/providers` to enable, configure, and test each
 | `zai-web` | `zw` | Z.ai Web | Web cookie | [link](https://chat.z.ai) | Copy the "token" value from chat.z.ai → DevTools → Application → Local Storage. Do not copy cookies; OmniRoute handles the per-request CAPTCHA through its browser transport. | — |
 | `zenmux-free` | `zmf` | ZenMux Free (Web) | Web cookie | [link](https://zenmux.ai) | Login at zenmux.ai, then export all cookies using EditThisCookie or Cookie-Editor and paste the full Cookie header string here. Refresh every ~30 days. | — |

-## API Key Providers (paid / paid-with-free-credits) (238)
+## API Key Providers (paid / paid-with-free-credits) (237)

 | ID | Alias | Name | Tags | Website | Notes |
 |----|-------|------|------|---------|-------|
@@ -310,7 +310,6 @@ Use the dashboard at `/dashboard/providers` to enable, configure, and test each
 | `sarvam` | `sarvam` | Sarvam AI | API key | [link](https://docs.sarvam.ai) | ₹1,000 in free signup credits — never expire |
 | `scaleway` | `scw` | Scaleway AI | API key | [link](https://www.scaleway.com/en/docs/ai-data/generative-apis/) | 1M free tokens for new accounts — EU/GDPR compliant (Paris), Qwen3 235B & Llama 70B |
 | `sealion` | `sealion` | SEA-LION | API key | [link](https://sea-lion.ai) | Sign in at sea-lion.ai with Google (no card, no region wall), create an API key, then paste it here. |
-| `seekai` | `ska` | SeekAi | API key, aggregator | [link](https://seekai.cc) | Create an API key at https://seekai.cc, then paste it here as a Bearer token. |
 | `segmind` | `segmind` | Segmind | API key, image, video | [link](https://segmind.com) | Use your Segmind API key in the x-api-key header. OmniRoute targets https://api.segmind.com/v1/<model> and returns the generated image/video bytes directly. |
 | `sensenova` | `sensenova` | SenseNova | API key | [link](https://platform.sensenova.cn) | Get API key at platform.sensenova.cn |
 | `siliconflow` | `siliconflow` | SiliconFlow | API key | [link](https://cloud.siliconflow.com) | $1 free credits plus currently listed $0 models after identity verification; availability and limits may change |
--- a/docs/security/STEALTH_GUIDE.md
+++ b/docs/security/STEALTH_GUIDE.md
@@ -117,8 +117,8 @@ Applied to: `system` blocks, all `messages[].content`, and `tools[].description`

 For third-party Anthropic relays that only accept "real Claude Code" traffic:

- `CLAUDE_CODE_COMPATIBLE_USER_AGENT = "claude-cli/2.1.258 (external, sdk-cli)"`
- `CLAUDE_CODE_COMPATIBLE_STAINLESS_PACKAGE_VERSION = "0.112.1"`
+- `CLAUDE_CODE_COMPATIBLE_USER_AGENT = "claude-cli/2.1.220 (external, sdk-cli)"`
+- `CLAUDE_CODE_COMPATIBLE_STAINLESS_PACKAGE_VERSION = "0.94.0"`
 - `CLAUDE_CODE_COMPATIBLE_STAINLESS_RUNTIME_VERSION = "v26.3.0"`
 - `anthropic-beta = "claude-code-20250219,interleaved-thinking-2025-05-14,effort-2025-11-24"` by default
 - The per-connection "Enable redact-thinking beta" toggle adds `redact-thinking-2026-02-12` when a CC Compatible upstream specifically requires redacted thinking streams
@@ -241,7 +241,7 @@ All MITM endpoints require management auth (`requireCliToolsAuth`). The sudo pas

 | Variable                 | Default                                                         |
 | ------------------------ | --------------------------------------------------------------- |
-| `CLAUDE_USER_AGENT`      | `claude-cli/2.1.258 (external, cli)`                            |
+| `CLAUDE_USER_AGENT`      | `claude-cli/2.1.220 (external, cli)`                            |
 | `CODEX_USER_AGENT`       | `codex-cli/0.149.0 (Windows 10.0.26200; x64)`                   |
 | `GITHUB_USER_AGENT`      | `GitHubCopilotChat/0.54.0`                                      |
 | `ANTIGRAVITY_USER_AGENT` | `antigravity/2.0.1 linux/arm64 google-api-nodejs-client/10.3.0` |
--- a/llm.txt
+++ b/llm.txt
@@ -1,6 +1,6 @@
 # OmniRoute

-> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 356 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.
+> OmniRoute is a free, open-source AI Gateway that acts as a universal API proxy for multi-provider LLMs. It provides smart routing, automatic fallback, load balancing, and format translation across 355 AI providers — all through a single OpenAI-compatible endpoint. Includes a built-in MCP Server (110 tools), A2A v0.3 protocol, Memory/Skills systems, Cloud Agents (codex, cursor, devin, jules), Guardrails framework, and an Electron desktop app.

 ## Overview

@@ -14,7 +14,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo

 - **Runtime:** Node.js `>=22.22.2 <23 || >=24.0.0 <27`, ES Modules (`"type": "module"`)
 - **Framework:** Next.js 16 (App Router) with TypeScript 6
- **Database:** SQLite via better-sqlite3 (local, zero-config, 169 migrations)
+- **Database:** SQLite via better-sqlite3 (local, zero-config, 168 migrations)
 - **State management:** Zustand (client), SQLite (server persistence)
 - **UI:** React 19, Tailwind CSS 4, Recharts for analytics, @lobehub/icons for 130+ provider SVG icons
 - **Auth:** OAuth 2.0 (PKCE) for providers, bcrypt for local user auth
@@ -124,7 +124,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 │   │   │   ├── secrets.ts        # Secrets management
 │   │   │   ├── stateReset.ts     # State reset utilities
 │   │   │   ├── migrationRunner.ts # Schema migration runner
-│   │   │   └── migrations/       # 169 versioned SQL migration files
+│   │   │   └── migrations/       # 168 versioned SQL migration files
 │   │   ├── evals/                # Eval runner and scheduler
 │   │   ├── memory/               # Persistent conversational memory
 │   │   │   ├── extraction.ts     # Memory extraction from conversations
@@ -277,7 +277,7 @@ OmniRoute solves the problem of managing multiple AI provider subscriptions, quo
 ## Key Features (v3.8.50)

 ### Core Proxy
- **356 AI providers** with automatic format translation
+- **355 AI providers** with automatic format translation
 - **Provider categories**: Free (90+ free tiers), OAuth, API Key, Self-Hosted, Custom (OpenAI/Anthropic-compatible)
 - **19 routing strategies**: priority, weighted, round-robin, fill-first, p2c, random, least-used, cost-optimized, reset-aware, reset-window, headroom, strict-random, auto, lkgp, context-optimized, cache-optimized, context-relay, fusion, pipeline
 - **4-tier fallback**: Subscription → API Key → Cheap → Free
@@ -389,7 +389,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 8. **ProviderIcon component:** Unified icon system using `@lobehub/icons` (130+ SVG) with PNG fallback and generic icon fallback chain. Used on providers, dashboard, and agents pages.

-9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 169 SQL migrations.
+9. **DB architecture:** `localDb.ts` is a re-export layer only — real logic lives in 122 `src/lib/db/` modules with 168 SQL migrations.

 10. **Upstream headers:** Custom headers merged in executors after default auth; same header name replaces executor value. Forbidden header names in `src/shared/constants/upstreamHeaders.ts`.

@@ -433,7 +433,7 @@ diagnostics) plus **memory**, **skill**, **agentSkill**, **githubSkill**, **pool

 4. **Environment variables:** All configuration is in `.env` (from `.env.example`). Key vars: `PORT`, `NEXT_PUBLIC_BASE_URL`, `API_KEY`, `ADMIN_PASSWORD`.

-5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 169 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.
+5. **Database layer:** Operations go through `src/lib/db/` modules (122 domain-specific files, 168 migrations). `localDb.ts` is re-exports only — add new functions to the proper `db/*.ts` module.

 6. **Tests** use Node.js built-in test runner + Vitest. Run `npm test`. Vitest for MCP/autoCombo (`npm run test:vitest`). Playwright for E2E (`npm run test:e2e`). Coverage gate: ratchet vs `quality-baseline.json`, absolute floor 60% statements/lines/functions/branches.

--- a/open-sse/config/providers/index.ts
+++ b/open-sse/config/providers/index.ts
@@ -270,7 +270,6 @@ import { voidAiProvider } from "./registry/void-ai/index.ts";
 import { helixmindProvider } from "./registry/helixmind/index.ts";
 import { tabitokenProvider } from "./registry/tabitoken/index.ts";
 import { logfareProvider } from "./registry/logfare/index.ts";
-import { seekaiProvider } from "./registry/seekai/index.ts";

 export const REGISTRY: Record<string, RegistryEntry> = {
  aimlapi: aimlapiProvider,
@@ -545,5 +544,4 @@ export const REGISTRY: Record<string, RegistryEntry> = {
  helixmind: helixmindProvider,
  tabitoken: tabitokenProvider,
  logfare: logfareProvider,
-  seekai: seekaiProvider,
 };
--- a/open-sse/config/providers/registry/seekai/index.ts
+++ b/open-sse/config/providers/registry/seekai/index.ts
@@ -1,18 +0,0 @@
-import { buildOpenAiCompatibleRegistryEntry } from "../../shared.ts";
-
-/**
- * SeekAi (https://seekai.cc) — QuantumNous New-API gateway.
- * Live-verified 2026-09-02: GET /api/status → system_name=SeekAi,
- * version=v1.0.0-rc.25, quota_display_type=USD. GET /v1/models is
- * API-key gated (401 Invalid token without a key). Catalog is dynamic;
- * no static seed. Referral/aff query params stay out of this entry
- * (no-hardcoded-referral-codes).
- */
-export const seekaiProvider = buildOpenAiCompatibleRegistryEntry({
-  id: "seekai",
-  alias: "ska",
-  baseUrl: "https://seekai.cc/v1/chat/completions",
-  modelsUrl: "https://seekai.cc/v1/models",
-  models: [],
-  passthroughModels: true,
-});
--- a/open-sse/executors/antigravity.ts
+++ b/open-sse/executors/antigravity.ts
@@ -26,7 +26,6 @@ import {
 } from "../services/antigravityCredits.ts";
 import { persistCreditBalance, getAllPersistedCreditBalances } from "@/lib/db/creditBalance";
 import { setConnectionRateLimitUntil } from "@/lib/db/providers";
-import { markAntigravityModelQuotaExhausted } from "../services/antigravityFamilyCooldown.ts";
 import { getMitmAlias } from "@/lib/db/models";
 import {
  MAX_ANTIGRAVITY_OUTPUT_TOKENS,
@@ -246,15 +245,17 @@ export function createCreditsExtractionTransform(
  );
 }

-export function markConnectionQuotaExhausted(
-  connectionId: string,
-  retryAfterMs: number,
-  model?: string | null
-): void {
+/**
+ * Persist a quota-exhausted cooldown to the DB for `connectionId` so that
+ * cross-request and post-restart routing skips this connection until the
+ * cooldown expires. Exported for unit testing. @internal
+ */
+export function markConnectionQuotaExhausted(connectionId: string, retryAfterMs: number): void {
  try {
-    if (markAntigravityModelQuotaExhausted(connectionId, retryAfterMs, model)) return;
    setConnectionRateLimitUntil(connectionId, Date.now() + retryAfterMs);
-  } catch {}
+  } catch {
+    // DB write failure must never crash the request path
+  }
 }

 /**
@@ -1619,7 +1620,7 @@ export class AntigravityExecutor extends BaseExecutor {
          updateAntigravityRemainingCredits
        );
        if (creditsResult) return { kind: "return", result: creditsResult };
-        if (retryMs) markConnectionQuotaExhausted(accountId, retryMs, ctx.model);
+        if (retryMs) markConnectionQuotaExhausted(accountId, retryMs);
      }

      return {
--- a/open-sse/handlers/chatCore.ts
+++ b/open-sse/handlers/chatCore.ts
@@ -384,7 +384,6 @@ import { sanitizeOpenAITool } from "../services/toolSchemaSanitizer.ts";
 import { isCompactResponsesEndpoint } from "../executors/codex.ts";
 import { persistCodexChildQuotaResponse } from "../services/codexAccount/index.ts";
 import { invalidateCodexQuotaCache } from "../services/codexQuotaFetcher.ts";
-import { invalidateGenericQuotaCacheOnStatus } from "../services/genericQuotaFetcher.ts";
 import { translateNonStreamingResponse } from "./responseTranslator.ts";
 import { extractToolSchemaMap } from "../translator/response/openai-responses/toolSchemas.ts";
 import { unwrapClineNonStreamingEnvelope } from "./chatCore/clineResponseEnvelope.ts";
@@ -3240,14 +3239,6 @@ export async function handleChatCore({
                  const errMessage = err instanceof Error ? err.message : String(err);
                  log?.debug?.("CODEX", `Failed to persist codex quota state: ${errMessage}`);
                }
-              } else if (attemptConnectionId && res.response.status === 429) {
-                // Dropped generic quota cache after 429
-                invalidateGenericQuotaCacheOnStatus({
-                  provider,
-                  connectionId: String(attemptConnectionId),
-                  status: res.response.status,
-                  isolateProbe: await shouldIsolateProbeFailures(),
-                });
              }

              // Track Gemini RPM + RPD request counts for 429 classification
--- a/open-sse/handlers/chatCore/attemptLogging.ts
+++ b/open-sse/handlers/chatCore/attemptLogging.ts
@@ -13,7 +13,6 @@
 import { extractProviderWarnings } from "@/lib/compliance/providerAudit";
 import { logAuditEvent } from "@/lib/compliance";
 import { emit } from "@/lib/events/eventBus";
-import { maybeLogToolCallSpecViolation } from "./toolCallSpecViolationAudit.ts";
 import type { RequestCompletedPayload, RequestFailedPayload } from "@/lib/events/types";
 import { saveCallLog } from "@/lib/usageDb";
 import type { VideoBridgeLogRedactionEntry } from "@/lib/guardrails/videoBridge";
@@ -393,15 +392,6 @@ export function persistAttemptLogs(args: PersistAttemptLogsArgs, ctx: PersistAtt
    });
  }

-  maybeLogToolCallSpecViolation({
-    responseBody,
-    provider,
-    model,
-    connectionId: finalConnectionId,
-    httpStatus: status,
-    requestId: skillRequestId,
-  });
-
  const capturedPipeline = reqLogger?.getPipelinePayloads?.() ?? null;
  const pipelinePayloads = detailedLoggingEnabled
    ? (capturedPipeline ?? {})
--- a/open-sse/handlers/chatCore/toolCallSpecViolationAudit.ts
+++ b/open-sse/handlers/chatCore/toolCallSpecViolationAudit.ts
@@ -1,46 +0,0 @@
-/**
- * Post-request on-spec audit for duplicated tool_calls.
- *
- * Extracted from persistAttemptLogs so attemptLogging.ts stays at the frozen
- * complexity count. validateResponseQuality's streaming peek only sees the
- * START of a stream, so a duplicate that arrives after real content has
- * already been relayed cannot fail the attempt over — this is the first
- * point the fully assembled body is available. Too late to retry; a durable
- * audit row still beats a clean HTTP 200 with no trace.
- *
- * Observed: minimax-m3:free via OpenRouter/GMICloud, 2026-09-02, duplicated
- * a heartbeat_respond call byte-for-byte.
- */
-
-import { logAuditEvent } from "@/lib/compliance";
-import { findToolCallSpecViolation } from "../../services/combo/validateQuality.ts";
-
-export function maybeLogToolCallSpecViolation(input: {
-  responseBody: unknown;
-  provider: string | null | undefined;
-  model: string | null | undefined;
-  connectionId: string | null;
-  httpStatus: number;
-  requestId: string;
-}): void {
-  const violation = findToolCallSpecViolation(input.responseBody);
-  if (!violation) return;
-  logAuditEvent({
-    action: "provider.spec_violation",
-    actor: "system",
-    target:
-      [input.provider, input.connectionId].filter(Boolean).join(":") ||
-      input.provider ||
-      input.model,
-    resourceType: "provider_spec_violation",
-    status: "warning",
-    requestId: input.requestId,
-    details: {
-      provider: input.provider,
-      model: input.model,
-      connectionId: input.connectionId,
-      httpStatus: input.httpStatus,
-      violation,
-    },
-  });
-}
--- a/open-sse/services/accountFallback.ts
+++ b/open-sse/services/accountFallback.ts
@@ -49,8 +49,7 @@ import {
 } from "../../src/shared/constants/providers";
 import { resolveUseUpstream429BreakerHints } from "../../src/shared/utils/providerHints";
 import { getCodexModelScope } from "../config/codexQuotaScopes.ts";
-import { getQuotaScopedModelForProvider, isAntigravityQuotaProvider } from "./antigravityQuotaFamily.ts";
-import { persistAntigravityFamilyCooldownIfQuota } from "./antigravityFamilyCooldown.ts";
+import { getQuotaScopedModelForProvider } from "./antigravityQuotaFamily.ts";
 import {
  classifyGeminiQuotaMetricFromText,
  isRpdExhausted,
@@ -64,8 +63,6 @@ import {
  parseDelayString,
  MAX_SHORT_RETRY_HINT_MS,
 } from "./retryAfterJson.ts";
-import { isMoonshotAccountBalanceExhausted } from "./usage/moonshotOpenPlatform.ts";
-import { isTpdRateLimit, resolveTpdCooldownMs } from "./dailyQuotaReset.ts";

 // Pre-compiled regex constants for hot-path retry parsing (avoid per-call compilation)
 const RETRY_AFTER_RE = /retry\s+after\s+(\d+)\s*s/i;
@@ -653,9 +650,6 @@ export async function recordCoreOwnedAntigravityQuotaState({
      exactCooldownIsUpstreamReset: retryHintBypassesMaxCooldownMs(fallback.retryHintSource),
    }
  );
-  if (lockout.cooldownMs > 0 && isProviderExhaustedReason(fallback)) {
-    persistAntigravityFamilyCooldownIfQuota({ provider, connectionId, model, cooldownMs: lockout.cooldownMs, reason: "quota_exhausted" });
-  }
  return { cooldownMs: lockout.cooldownMs, failureCount: lockout.failureCount };
 }

@@ -1607,8 +1601,7 @@ export function isDailyQuotaExhausted(errorText: string): boolean {
  return (
    lower.includes("today's quota") ||
    lower.includes("daily quota") ||
-    lower.includes("try again tomorrow") ||
-    lower.includes("tpd rate limit")
+    lower.includes("try again tomorrow")
  );
 }

@@ -1654,12 +1647,7 @@ export function checkFallbackError(
  headers: Headers | Record<string, string> | null = null,
  profileOverride: ProviderProfile | null = null,
  structuredError?: { code?: string | null; type?: string | null } | null,
-  rotation?: { account?: unknown } | null,
-  dailyReset?: {
-    timezone?: unknown;
-    hour?: unknown;
-    nowMs?: number;
-  } | null,
+  rotation?: { account?: unknown } | null
 ): {
  shouldFallback: boolean;
  cooldownMs: number;
@@ -1942,13 +1930,8 @@ export function checkFallbackError(
      }
    }

-    // T10 (sub2api #1169) + #8247: credits/quota exhausted; *-compatible-* nicknames stay model-scoped
-    // unless the body is an account-level Open Platform empty wallet.
-    if (
-      shouldUseQuotaSignal &&
-      isCreditsExhausted(errorStr) &&
-      (!isCompatibleProvider(provider) || isMoonshotAccountBalanceExhausted(errorStr))
-    ) {
+    // T10 (sub2api #1169) + #8247: credits/quota exhausted; *-compatible-* nicknames stay model-scoped.
+    if (shouldUseQuotaSignal && isCreditsExhausted(errorStr) && !isCompatibleProvider(provider)) {
      return {
        shouldFallback: true,
        cooldownMs: COOLDOWN_MS.paymentRequired ?? 3600 * 1000, // 1h cooldown
@@ -1957,43 +1940,17 @@ export function checkFallbackError(
      };
    }

-    // Daily quota exhausted. TPD uses the node clock / header; other daily
-    // quota text still uses getMsUntilTomorrow. TPD without either is not a
-    // host-midnight lock — fall through to short 429.
+    // Daily quota exhausted — lock model until tomorrow
    if (shouldUseQuotaSignal && isDailyQuotaExhausted(errorStr)) {
-      if (isTpdRateLimit(errorStr)) {
-        const headerResetAtMs = parseResetFromHeaders(headers);
-        const tpdMs = resolveTpdCooldownMs(errorStr, {
-          timezone: dailyReset?.timezone,
-          hour: dailyReset?.hour,
-          nowMs: dailyReset?.nowMs,
-          headerResetAtMs,
-        });
-        if (tpdMs == null) {
-          // no clock, no header — short 429, do not guess midnight
-          console.warn(
-            "[accountFallback] TPD 429 without node daily-reset clock or Reset header; using short cooldown",
-            { provider },
-          );
-        } else {
-          return {
-            shouldFallback: true,
-            cooldownMs: tpdMs,
-            reason: RateLimitReason.QUOTA_EXHAUSTED,
-            dailyQuotaExhausted: true,
-          };
-        }
-      } else {
-        const msUntilTomorrow = getMsUntilTomorrow();
-        // Cap at 24 hours to handle timezone edge cases
-        const cooldownMs = Math.min(msUntilTomorrow, 24 * 60 * 60 * 1000);
-        return {
-          shouldFallback: true,
-          cooldownMs,
-          reason: RateLimitReason.QUOTA_EXHAUSTED,
-          dailyQuotaExhausted: true,
-        };
-      }
+      const msUntilTomorrow = getMsUntilTomorrow();
+      // Cap at 24 hours to handle timezone edge cases
+      const cooldownMs = Math.min(msUntilTomorrow, 24 * 60 * 60 * 1000);
+      return {
+        shouldFallback: true,
+        cooldownMs,
+        reason: RateLimitReason.QUOTA_EXHAUSTED,
+        dailyQuotaExhausted: true,
+      };
    }

    // Issue #2321 (5h subscription quota) + Issue #3709 (ollama-cloud weekly
@@ -2432,7 +2389,12 @@ export function applyErrorState<T extends AccountState | null | undefined>(
  // (`markConnectionQuotaExhausted`) so a DB failure can never crash the
  // chat path. See issue #1 (per-account 429 cascade not persisting).
  const connId = (account as AccountState | null | undefined)?.id;
-  if (typeof connId === "string" && connId.length > 0 && effectiveCooldownMs > 0 && nextState.rateLimitedUntil && !isAntigravityQuotaProvider(prov)) {
+  if (
+    typeof connId === "string" &&
+    connId.length > 0 &&
+    effectiveCooldownMs > 0 &&
+    nextState.rateLimitedUntil
+  ) {
    try {
      const untilMs = cooldownUntilMs(nextState.rateLimitedUntil);
      if (Number.isFinite(untilMs) && untilMs > Date.now()) {
--- a/open-sse/services/antigravityFamilyCooldown.ts
+++ b/open-sse/services/antigravityFamilyCooldown.ts
@@ -1,158 +0,0 @@
-/**
- * Persist Antigravity/agy quota cooldowns per model family (gemini vs claude)
- * on the connection row, without cooling the whole account.
- */
-import { lockModel } from "./accountFallback.ts";
-import {
-  getAntigravityQuotaFamily,
-  isAntigravityQuotaProvider,
-} from "./antigravityQuotaFamily.ts";
-
-type JsonRecord = Record<string, unknown>;
-
-const FAMILY_PSD_KEY = "antigravityFamilyRateLimitedUntil";
-
-function asRecord(value: unknown): JsonRecord {
-  return value && typeof value === "object" && !Array.isArray(value)
-    ? (value as JsonRecord)
-    : {};
-}
-
-function parseUntilMs(value: unknown): number {
-  if (typeof value === "number" && Number.isFinite(value)) return value;
-  if (typeof value === "string" && value.trim()) {
-    const ms = /^\d+(\.\d+)?$/.test(value.trim()) ? Number(value) : Date.parse(value);
-    return Number.isFinite(ms) ? ms : NaN;
-  }
-  return NaN;
-}
-
-function dummyModelForFamily(family: "gemini" | "claude"): string {
-  return family === "gemini" ? "gemini-family-lock" : "claude-family-lock";
-}
-
-function lockAntigravityFamilyModel(
-  connectionId: string,
-  model: string,
-  reason: string,
-  cooldownMs: number
-): void {
-  lockModel("agy", connectionId, model, reason, cooldownMs);
-  lockModel("antigravity", connectionId, model, reason, cooldownMs);
-}
-
-export async function persistAntigravityFamilyCooldown(params: {
-  connectionId: string;
-  model: string;
-  rateLimitedUntil: string;
-}): Promise<JsonRecord | null> {
-  if (!params.model.trim()) return null;
-  const family = getAntigravityQuotaFamily(params.model);
-  if (family === "other") return null;
-
-  const { getProviderConnectionById, updateProviderConnection } = await import(
-    "@/lib/db/providers"
-  );
-  const conn = (await getProviderConnectionById(params.connectionId)) as
-    | { provider?: string; providerSpecificData?: JsonRecord | null }
-    | null;
-  if (!conn || !isAntigravityQuotaProvider(conn.provider ?? null)) return null;
-
-  const psd = asRecord(conn.providerSpecificData);
-  const untils = asRecord(psd[FAMILY_PSD_KEY]);
-  const existingMs = parseUntilMs(untils[family]);
-  const nextMs = parseUntilMs(params.rateLimitedUntil);
-  if (!Number.isFinite(nextMs)) return psd;
-  if (Number.isFinite(existingMs) && existingMs > Date.now() && existingMs >= nextMs) {
-    return psd;
-  }
-
-  const nextPsd: JsonRecord = {
-    ...psd,
-    [FAMILY_PSD_KEY]: { ...untils, [family]: params.rateLimitedUntil },
-  };
-  await updateProviderConnection(params.connectionId, { providerSpecificData: nextPsd });
-  return nextPsd;
-}
-
-/** Fire-and-forget family PSD write. RPM/burst 429s must pass reason !== quota_exhausted. */
-export function persistAntigravityFamilyCooldownIfQuota(params: {
-  provider?: string | null;
-  connectionId: string;
-  model?: string | null;
-  cooldownMs: number;
-  reason?: string | null;
-}): void {
-  if (!isAntigravityQuotaProvider(params.provider)) return;
-  if (!params.model?.trim() || params.cooldownMs <= 0) return;
-  if (params.reason != null && params.reason !== "quota_exhausted") return;
-  void persistAntigravityFamilyCooldown({
-    connectionId: params.connectionId,
-    model: params.model,
-    rateLimitedUntil: new Date(Date.now() + params.cooldownMs).toISOString(),
-  }).catch(() => {});
-}
-
-export async function persistAntigravityPreflightFamilyLock(params: {
-  provider: string;
-  connectionId: string;
-  model: string;
-  unavailableUntil: string;
-}): Promise<void> {
-  const cooldownMs = Math.max(0, Date.parse(params.unavailableUntil) - Date.now());
-  lockAntigravityFamilyModel(params.connectionId, params.model, "quota_exhausted", cooldownMs);
-  await persistAntigravityFamilyCooldown({
-    connectionId: params.connectionId,
-    model: params.model,
-    rateLimitedUntil: params.unavailableUntil,
-  });
-}
-
-export function rehydrateAntigravityFamilyLocks(
-  provider: string,
-  connectionId: string,
-  providerSpecificData: JsonRecord | null | undefined
-): void {
-  if (!isAntigravityQuotaProvider(provider)) return;
-  const untils = asRecord(asRecord(providerSpecificData)[FAMILY_PSD_KEY]);
-  const now = Date.now();
-  for (const family of ["gemini", "claude"] as const) {
-    const untilMs = parseUntilMs(untils[family]);
-    if (!Number.isFinite(untilMs) || untilMs <= now) continue;
-    const model = dummyModelForFamily(family);
-    const remainingMs = untilMs - now;
-    lockAntigravityFamilyModel(connectionId, model, "quota_exhausted", remainingMs);
-  }
-}
-
-export function rehydrateAntigravityFamilyLocksForConnections(
-  provider: string,
-  connections: Array<{ id: string; providerSpecificData?: unknown }>
-): void {
-  if (!isAntigravityQuotaProvider(provider)) return;
-  for (const conn of connections) {
-    rehydrateAntigravityFamilyLocks(
-      provider,
-      conn.id,
-      conn.providerSpecificData as JsonRecord | null | undefined
-    );
-  }
-}
-
-/** Family lock for executor quota exhaustion. Returns false when model is absent. */
-export function markAntigravityModelQuotaExhausted(
-  connectionId: string,
-  retryAfterMs: number,
-  model?: string | null
-): boolean {
-  if (!model) return false;
-  lockAntigravityFamilyModel(connectionId, model, "quota_exhausted", retryAfterMs);
-  persistAntigravityFamilyCooldownIfQuota({
-    provider: "agy",
-    connectionId,
-    model,
-    cooldownMs: retryAfterMs,
-    reason: "quota_exhausted",
-  });
-  return true;
-}
--- a/open-sse/services/antigravityQuotaFamily.ts
+++ b/open-sse/services/antigravityQuotaFamily.ts
@@ -54,77 +54,3 @@ export function getQuotaScopeLabelForProvider(
  if (provider !== "antigravity" && provider !== "agy") return "model";
  return getAntigravityQuotaFamily(model) === "other" ? "model" : "family";
 }
-
-export function isAntigravityQuotaProvider(provider: string | null | undefined): boolean {
-  return provider === "antigravity" || provider === "agy";
-}
-
-export function quotaWindowNamesForScope(
-  names: string[],
-  scope?: { provider?: string | null; requestedModel?: string | null }
-): string[] {
-  if (!scope?.requestedModel || !isAntigravityQuotaProvider(scope.provider)) return names;
-  const scoped = selectAntigravityQuotaWindowNames(names, scope.requestedModel);
-  return scoped.length > 0 ? scoped : names;
-}
-
-/** Min remaining % across scoped windows, or 100 when an Antigravity family scope matched none. */
-export function remainingPercentFromQuotaWindows(
-  rawWindows: Record<string, unknown>,
-  scope?: { provider?: string | null; requestedModel?: string | null }
-): number | null {
-  const names = Object.keys(rawWindows);
-  const namesToScan = quotaWindowNamesForScope(names, scope);
-  let minRemaining: number | null = null;
-  for (const name of namesToScan) {
-    const windowInfo = rawWindows[name];
-    if (!windowInfo || typeof windowInfo !== "object") continue;
-    const percentUsed = Number((windowInfo as Record<string, unknown>).percentUsed);
-    if (!Number.isFinite(percentUsed)) continue;
-    const remaining = Math.max(0, Math.min(100, (1 - percentUsed) * 100));
-    minRemaining = minRemaining === null ? remaining : Math.min(minRemaining, remaining);
-  }
-  if (minRemaining !== null) return minRemaining;
-  if (scope?.requestedModel && namesToScan !== names) return 100;
-  return null;
-}
-
-/**
- * Windows that belong to the requested Antigravity family. Claude weekly must
- * not ride along on a Gemini request (and the reverse).
- */
-export function selectAntigravityQuotaWindowNames(
-  quotaNames: string[],
-  requestedModel: string | null | undefined
-): string[] {
-  if (!requestedModel) return quotaNames;
-  const requestedFamily = getAntigravityQuotaFamily(requestedModel);
-  const cleanRequestedModel = requestedModel.replace(/^(antigravity|agy)\//, "");
-  const bareModel = cleanRequestedModel.includes("/")
-    ? cleanRequestedModel.slice(cleanRequestedModel.lastIndexOf("/") + 1)
-    : cleanRequestedModel;
-
-  if (requestedFamily === "other") {
-    return quotaNames.filter((windowName) => {
-      const bare = windowName.replace(/^(antigravity|agy)\//, "");
-      return bare === bareModel || bare === cleanRequestedModel;
-    });
-  }
-
-  const familyAggregates =
-    requestedFamily === "gemini"
-      ? ["gemini_weekly"]
-      : requestedFamily === "claude"
-        ? ["claude_gpt_weekly"]
-        : [];
-
-  const exactWindows = quotaNames.filter((windowName) => {
-    const bare = windowName.replace(/^(antigravity|agy)\//, "");
-    return bare === bareModel;
-  });
-  const aggregateWindows = familyAggregates.filter((key) => quotaNames.includes(key));
-  const scoped = [...exactWindows, ...aggregateWindows];
-  if (scoped.length > 0) return scoped;
-
-  return quotaNames.filter((windowName) => getAntigravityQuotaFamily(windowName) === requestedFamily);
-}
--- a/open-sse/services/combo.ts
+++ b/open-sse/services/combo.ts
@@ -613,13 +613,12 @@ export async function buildAutoCandidates(
        const quota = await quotaPromises.get(quotaKey)!;
        resetWindowAffinity = calculateResetWindowAffinity(quota, resetWindowConfig);
        if (!quotaCutoffBlocked) {
-          quotaRemaining = quotaRemainingPercentFromQuota(quota, { provider, requestedModel: modelStr });
+          quotaRemaining = quotaRemainingPercentFromQuota(quota);
        }
        if (!quotaCutoffBlocked && quotaCutoffEnabled) {
          const cutoffDecision = evaluateQuotaCutoff(
            quota as QuotaInfo | null,
-            buildAutoQuotaThresholds(provider, connection, resilienceSettings),
-            { provider, requestedModel: modelStr }
+            buildAutoQuotaThresholds(provider, connection, resilienceSettings)
          );
          if (!cutoffDecision.proceed) {
            quotaCutoffBlocked = true;
@@ -1380,7 +1379,7 @@ async function handleComboChatInner({
            resilienceSettings,
            quotaCutoffResetWindowConfig,
            combo.name,
-            log, modelStr
+            log
          );
          if (quotaCutoff.blocked) {
            log.info(
@@ -1661,16 +1660,8 @@ async function handleComboChatInner({
            }
          }

-          // Universal handoff: inject existing handoff if model changed. i === 0
-          // only: a fallback target (i > 0) serves the SAME client request the
-          // failed primary target would have served, with the original messages
-          // already intact -- there's nothing to hand off, since the client never
-          // saw the earlier target fail. Injecting a handoff note there replaces
-          // real context with a context-free note, which weaker fallback models
-          // have been observed treating as license to fabricate content instead
-          // of just answering the actual request (#12227 follow-up).
+          // Universal handoff: inject existing handoff if model changed
          if (
-            i === 0 &&
            universalHandoffConfig.enabled &&
            relayOptions?.sessionId &&
            !(body as Record<string, unknown>)?.[SKIP_UNIVERSAL_HANDOFF_FLAG]
@@ -1936,14 +1927,7 @@ async function handleComboChatInner({
                provider,
                target.connectionId ?? undefined
              );
-              // i === 0 only: a same-request fallback target (i > 0) never
-              // needs a summary generated for it -- see the injection-site
-              // comment above. recordSessionModelUsage above stays
-              // unconditional regardless of i: it must reflect whichever
-              // model actually served THIS response, since the next
-              // request's i === 0 comparison depends on that being
-              // accurate even when this response came from a fallback.
-              if (i === 0 && prevModel && prevModel !== modelStr) {
+              if (prevModel && prevModel !== modelStr) {
                const handoffSourceMessages =
                  Array.isArray(body?.messages) && body.messages.length > 0
                    ? body.messages
@@ -4031,5 +4015,8 @@ async function handleRoundRobinCombo({
  }

  log.warn("COMBO-RR", `All models failed | ${msg}`);
-  return new Response(JSON.stringify({ error: { message: msg } }), { status, headers: { "Content-Type": "application/json" } });
+  return new Response(JSON.stringify({ error: { message: msg } }), {
+    status,
+    headers: { "Content-Type": "application/json" },
+  });
 }
--- a/open-sse/services/combo/comboPredicates.ts
+++ b/open-sse/services/combo/comboPredicates.ts
@@ -7,7 +7,6 @@
 */

 import { EXECUTOR_CONTRACT_VIOLATION_CODE } from "../../config/constants.ts";
-import { remainingPercentFromQuotaWindows } from "../antigravityQuotaFamily.ts";
 import { errorResponse } from "../../utils/error.ts";
 import { parseModel } from "../model.ts";
 import { isSelfInflictedUpstreamTimeout } from "../../handlers/chatCore/cooldownClassification.ts";
@@ -432,24 +431,24 @@ export function clampPercent(value: number): number {
  return Math.max(0, Math.min(100, value));
 }

-export function quotaRemainingPercentFromQuota(
-  quota: unknown,
-  scope?: { provider?: string | null; requestedModel?: string | null }
-): number {
+export function quotaRemainingPercentFromQuota(quota: unknown): number {
  if (!quota || typeof quota !== "object") return 100;
  const record = quota as Record<string, unknown>;
+  if (record.limitReached === true) return 0;

  const windows = record.windows;
  if (windows && typeof windows === "object" && !Array.isArray(windows)) {
-    const fromWindows = remainingPercentFromQuotaWindows(
-      windows as Record<string, unknown>,
-      scope
-    );
-    if (fromWindows !== null) return fromWindows;
+    let minRemaining: number | null = null;
+    for (const windowInfo of Object.values(windows as Record<string, unknown>)) {
+      if (!windowInfo || typeof windowInfo !== "object") continue;
+      const percentUsed = Number((windowInfo as Record<string, unknown>).percentUsed);
+      if (!Number.isFinite(percentUsed)) continue;
+      const remaining = clampPercent((1 - percentUsed) * 100);
+      minRemaining = minRemaining === null ? remaining : Math.min(minRemaining, remaining);
+    }
+    if (minRemaining !== null) return minRemaining;
  }

-  if (record.limitReached === true) return 0;
-
  const percentUsed = Number(record.percentUsed);
  if (Number.isFinite(percentUsed)) return clampPercent((1 - percentUsed) * 100);
  return 100;
--- a/open-sse/services/combo/nativeCodexTurnPin.ts
+++ b/open-sse/services/combo/nativeCodexTurnPin.ts
@@ -245,8 +245,7 @@ export async function isPinnedTargetModelScopedUnusable(args: {
      resilienceSettings,
      quotaCutoffResetWindowConfig,
      comboName,
-      log ?? { info: () => {}, warn: () => {}, error: () => {}, debug: () => {} },
-      target.modelStr
+      log ?? { info: () => {}, warn: () => {}, error: () => {}, debug: () => {} }
    );
    if (cutoff.blocked) return true;
  }
--- a/open-sse/services/combo/quotaExhaustionCutoff.ts
+++ b/open-sse/services/combo/quotaExhaustionCutoff.ts
@@ -97,8 +97,7 @@ export async function resolveQuotaExhaustionCutoffForTarget(
  resilienceSettings: ResilienceSettings | null | undefined,
  resetWindowConfig: ResetWindowConfig,
  comboName: string,
-  log: { debug?: (...args: unknown[]) => void; warn?: (...args: unknown[]) => void },
-  requestedModel?: string | null
+  log: { debug?: (...args: unknown[]) => void; warn?: (...args: unknown[]) => void }
 ): Promise<{ blocked: boolean; reason?: string }> {
  const quotaCutoffEnabled =
    (resilienceSettings ?? resolveResilienceSettings(null))?.quotaPreflight?.enabled === true;
@@ -127,8 +126,7 @@ export async function resolveQuotaExhaustionCutoffForTarget(
    });
    const cutoffDecision = evaluateQuotaCutoff(
      quota as QuotaInfo | null,
-      buildAutoQuotaThresholds(provider, connection, resilienceSettings),
-      { provider, requestedModel: requestedModel ?? null }
+      buildAutoQuotaThresholds(provider, connection, resilienceSettings)
    );
    if (!cutoffDecision.proceed) {
      return { blocked: true, reason: cutoffDecision.reason || "quota_exhausted" };
--- a/open-sse/services/combo/validateQuality.ts
+++ b/open-sse/services/combo/validateQuality.ts
@@ -16,45 +16,6 @@ import { evaluateResponseValidation, type ResponseValidationConfig } from "./res
 import { getReasoningTokens } from "../../../src/lib/usage/tokenAccounting.ts";
 import type { ComboRetryAfter } from "./types.ts";

-/**
- * Detects tool_calls entries within one assistant message that repeat the
- * exact same function name + arguments verbatim -- always a bug (no
- * legitimate use calls one tool twice with identical arguments in the same
- * turn), and a real observed failure mode of at least one free-tier
- * streaming model (minimax-m3:free via OpenRouter/GMICloud, 2026-09-02:
- * duplicated a heartbeat_respond call byte-for-byte, confirmed at the raw
- * SSE wire level -- an upstream bug, not an OmniRoute reconstruction
- * artifact). Used two ways: to fail a non-streaming response over to a
- * sibling combo target (see validateResponseQuality below), and, post-
- * stream, to flag an already-relayed streaming response as an on-spec
- * violation despite its clean HTTP 200 (see attemptLogging.ts's
- * persistAttemptLogs) -- a streaming response can't be retried once real
- * content has started reaching the client (the quality-gate peek below only
- * ever validates the START of a stream, by design, to avoid buffering the
- * whole response and defeating streaming's latency purpose), so flagging it
- * after the fact is what's actually achievable for that path.
- */
-export function findToolCallSpecViolation(responseBody: unknown): string | null {
-  const json = isRecord(responseBody) ? responseBody : null;
-  const choices = json?.choices;
-  const firstChoice = Array.isArray(choices) ? choices[0] : null;
-  const message = isRecord(firstChoice) ? firstChoice.message : null;
-  const toolCalls = isRecord(message) ? message.tool_calls : null;
-  if (!Array.isArray(toolCalls) || toolCalls.length < 2) return null;
-
-  const seen = new Set<string>();
-  for (const call of toolCalls) {
-    const fn = isRecord(call) ? call.function : null;
-    if (!isRecord(fn) || typeof fn.name !== "string" || typeof fn.arguments !== "string") {
-      continue;
-    }
-    const signature = `${fn.name}\u0000${fn.arguments}`;
-    if (seen.has(signature)) return `duplicate tool_calls entry for "${fn.name}"`;
-    seen.add(signature);
-  }
-  return null;
-}
-
 export function toRetryAfterDisplayValue(value: ComboRetryAfter): string | Date {
  if (typeof value !== "number") return value;
  if (value > 0 && value < 1_000_000_000) {
@@ -366,9 +327,9 @@ export async function validateResponseQuality(
    function isTerminalUsageOnlyChunk(parsed: Record<string, unknown>, eventType: string): boolean {
      return Boolean(
        parsed.usage &&
-        typeof parsed.usage === "object" &&
-        !Array.isArray(parsed.choices) &&
-        !eventType.startsWith("response.")
+          typeof parsed.usage === "object" &&
+          !Array.isArray(parsed.choices) &&
+          !eventType.startsWith("response.")
      );
    }

@@ -773,11 +734,6 @@ export async function validateResponseQuality(
  }
  const hasToolCalls = Array.isArray(toolCalls) && toolCalls.length > 0;

-  const specViolation = findToolCallSpecViolation(json);
-  if (specViolation) {
-    return { valid: false, reason: specViolation };
-  }
-
  if (!hasContent && !hasToolCalls) {
    return { valid: false, reason: "empty content and no tool_calls in response" };
  }
--- a/open-sse/services/contextHandoff.ts
+++ b/open-sse/services/contextHandoff.ts
@@ -407,10 +407,7 @@ async function generateHandoffAsync(options: {
    relayConfig.relayMode
  );
  const historyText = formatMessagesForPrompt(selectedMessages);
-  if (!historyText) {
-    logUniversalHandoffOutcome("unavailable", options.comboName, "empty selected-message history");
-    return;
-  }
+  if (!historyText) return;

  const summaryPrompt = HANDOFF_PROMPT_TEMPLATE.replace("{HISTORY}", historyText);
  const summaryBody = {
@@ -424,14 +421,7 @@ async function generateHandoffAsync(options: {
  };

  const response = await options.handleSingleModel(summaryBody, summaryModel);
-  if (!response.ok) {
-    logUniversalHandoffOutcome(
-      "unavailable",
-      options.comboName,
-      `summary model call failed: status=${response.status} model=${summaryModel}`
-    );
-    return;
-  }
+  if (!response.ok) return;

  let content = "";
  try {
@@ -446,14 +436,7 @@ async function generateHandoffAsync(options: {
  }

  const parsed = parseHandoffJSON(content);
-  if (!parsed) {
-    logUniversalHandoffOutcome(
-      "unparseable",
-      options.comboName,
-      `model=${summaryModel} contentPreview=${JSON.stringify(content.slice(0, 200))}`
-    );
-    return;
-  }
+  if (!parsed) return;

  upsertHandoff({
    sessionId: options.sessionId,
@@ -589,7 +572,7 @@ export function buildUniversalHandoffSystemMessage(
 <transfer_reason>${escapedReason}</transfer_reason>
 <previous_model>${escapedPrev}</previous_model>
 <current_model>${escapedCurr}</current_model>
-<note>No prior-session summary is available for this handoff. The input below (e.g. a tool result) is the entire context you have -- do not assume or invent details about a broader conversation you cannot see.</note>
+<note>A continuación se resume toda la conversacion para continuar sin perder el hilo.</note>
 </context_handoff>`;
  }

@@ -704,23 +687,6 @@ export function resetUniversalHandoffCooldowns(): void {
  universalHandoffCooldowns.clear();
 }

-// Every non-"generated" outcome across both handoff generators (this one and
-// the older generateHandoffAsync above) used to be silent -- context_handoffs
-// staying empty gave no signal on WHY (upstream call failing vs. malformed
-// output vs. no history to summarize). Every live handoff then falls back to
-// the bare no-summary note (buildUniversalHandoffSystemMessage's `!payload`
-// branch / the context-relay equivalent), which is what actually reaches the
-// model/user; without this log that always reads as a mystery instead of a
-// traceable cause.
-function logUniversalHandoffOutcome(
-  outcome: "unavailable" | "unparseable",
-  comboName: string,
-  detail: string
-): void {
-  if (process.env.NODE_ENV === "test") return;
-  console.warn(`[universal-handoff] ${outcome} (combo=${comboName}): ${detail}`);
-}
-
 /**
 * Generate a universal handoff summary for any model/provider switch.
 */
@@ -743,10 +709,7 @@ async function generateUniversalHandoffAsync(options: {
    options.relayMode
  );
  const historyText = formatMessagesForPrompt(selectedMessages);
-  if (!historyText) {
-    logUniversalHandoffOutcome("unavailable", options.comboName, "empty selected-message history");
-    return "unavailable";
-  }
+  if (!historyText) return "unavailable";

  const summaryPrompt = HANDOFF_PROMPT_TEMPLATE.replace("{HISTORY}", historyText);
  const summaryModel = options.handoffModel || options.currModel;
@@ -772,25 +735,22 @@ async function generateUniversalHandoffAsync(options: {
  };

  const response = await options.handleSingleModel(summaryBody, summaryModel);
-  if (!response.ok) {
-    const detail = `summary model call failed: status=${response.status} model=${summaryModel}`;
-    logUniversalHandoffOutcome("unavailable", options.comboName, detail);
-    return "unavailable";
-  }
+  if (!response.ok) return "unavailable";

  let content = "";
  try {
-    content = getResponseText((await response.clone().json()) as Record<string, unknown>);
+    const json = (await response.clone().json()) as Record<string, unknown>;
+    content = getResponseText(json);
  } catch {
-    content = await response.clone().text().catch(() => "");
+    try {
+      content = await response.clone().text();
+    } catch {
+      content = "";
+    }
  }

  const parsed = parseHandoffJSON(content);
-  if (!parsed) {
-    const preview = JSON.stringify(content.slice(0, 200));
-    logUniversalHandoffOutcome("unparseable", options.comboName, `model=${summaryModel} contentPreview=${preview}`);
-    return "unparseable";
-  }
+  if (!parsed) return "unparseable";

  upsertHandoff({
    sessionId: options.sessionId,
--- a/open-sse/services/conversationTurnContent.ts
+++ b/open-sse/services/conversationTurnContent.ts
@@ -23,90 +23,6 @@ export type TurnDisplayContent = {
  toolName: string | null;
 };

-type CanonicalTurnLike = {
-  role: "system" | "user" | "assistant" | "tool";
-  text: string;
-  blockKind: "text" | "tool_use" | "tool_result";
-  toolName: string | null;
-};
-
-type JsonRecord = Record<string, unknown>;
-
-function asRecord(value: unknown): JsonRecord | null {
-  return value && typeof value === "object" ? (value as JsonRecord) : null;
-}
-
-function turnsFromBody(body: unknown): CanonicalTurnLike[] {
-  const rec = asRecord(body);
-  return rec ? extractCanonicalTurns(rec) : [];
-}
-
-/**
- * extractCanonicalTurns's Chat Completions branch only reads a message's
- * `content` -- a tool-calling assistant message carries its call in
- * `tool_calls` instead with `content: null`, so it silently produces no turn
- * at all and the matching conversation_turn_nodes row can never resolve.
- * Deliberately scoped to this read-only display path instead of extending
- * extractCanonicalTurns itself: that function also drives
- * conversationTracker.ts's write-path identity/hashing, and this codebase's
- * only caller of it there (chat.ts's resolveConversationId) always feeds the
- * client-facing Responses-API body -- never Chat Completions
- * `messages`/`tool_calls` -- so extending it there would be unreachable for
- * real traffic here but still carries real write-path identity-hash risk for
- * any other caller/format that function might ever serve. Mirrors
- * extractCanonicalTurns's own Responses-shape function_call handling: one
- * turn per call, role "tool" (matches how a Responses API function_call item,
- * which also carries no `role`, canonicalizes -- not "assistant"), toolName
- * from the call, text the raw arguments string untouched (already a JSON
- * string in both APIs, so passing it through unmodified is what a
- * byte-identical hash against the original Responses-shaped item needs).
- */
-function extractChatCompletionsToolUseTurns(messages: unknown): CanonicalTurnLike[] {
-  if (!Array.isArray(messages)) return [];
-  const turns: CanonicalTurnLike[] = [];
-  for (const item of messages) {
-    const rec = asRecord(item) ?? {};
-    if (rec.role !== "assistant" || !Array.isArray(rec.tool_calls)) continue;
-    for (const call of rec.tool_calls) {
-      const fn = asRecord(asRecord(call)?.function);
-      const args = fn?.arguments;
-      if (typeof args !== "string" || !args) continue;
-      turns.push({
-        role: "tool",
-        text: args,
-        blockKind: "tool_use",
-        toolName: typeof fn?.name === "string" ? fn.name : null,
-      });
-    }
-  }
-  return turns;
-}
-
-function turnsFromClientResponse(clientResponse: unknown): CanonicalTurnLike[] {
-  const rec = asRecord(clientResponse);
-  if (!rec) return [];
-  const summary = asRecord(rec.summary);
-  const output = Array.isArray(rec.output) ? rec.output : summary?.output;
-  return Array.isArray(output) ? extractCanonicalTurns({ input: output }) : [];
-}
-
-function turnsFromProviderRequest(body: unknown): CanonicalTurnLike[] {
-  const rec = asRecord(body);
-  return [...turnsFromBody(rec), ...extractChatCompletionsToolUseTurns(rec?.messages)];
-}
-
-function indexTurns(result: Map<string, TurnDisplayContent>, turns: CanonicalTurnLike[]): void {
-  for (const turn of turns) {
-    const hash = hashTurnContent(turn);
-    if (result.has(hash)) continue;
-    result.set(hash, {
-      textPreview: turn.text,
-      blockKind: turn.blockKind,
-      toolName: turn.toolName,
-    });
-  }
-}
-
 /**
 * Resolve display content for a batch of turn nodes, keyed by content_hash.
 * Content_hash is sha256(role+text) only — real traffic has plenty of
@@ -148,10 +64,19 @@ export function resolveTurnDisplayContent(
  for (const relPath of artifactPathByCorrelationId.values()) {
    const { artifact, state } = readCallArtifact(relPath);
    if (state !== "ready") continue;
-    const pipeline = asRecord(artifact?.pipeline);
-    indexTurns(result, turnsFromBody(asRecord(pipeline?.clientRawRequest)?.body));
-    indexTurns(result, turnsFromClientResponse(pipeline?.clientResponse));
-    indexTurns(result, turnsFromProviderRequest(asRecord(pipeline?.providerRequest)?.body));
+    const clientRawRequest = artifact?.pipeline?.clientRawRequest as { body?: unknown } | undefined;
+    const body = clientRawRequest?.body;
+    if (!body || typeof body !== "object") continue;
+
+    for (const turn of extractCanonicalTurns(body as Record<string, unknown>)) {
+      const hash = hashTurnContent(turn);
+      if (result.has(hash)) continue;
+      result.set(hash, {
+        textPreview: turn.text,
+        blockKind: turn.blockKind,
+        toolName: turn.toolName,
+      });
+    }
  }
  return result;
 }
--- a/open-sse/services/dailyQuotaReset.ts
+++ b/open-sse/services/dailyQuotaReset.ts
@@ -1,145 +0,0 @@
-/**
- * Node-level daily quota reset clock.
- *
- * TPD cooldown endpoint: operator-configured IANA timezone + local hour.
- * No default timezone. Do not call getMsUntilTomorrow() from here.
- */
-
-export function isValidIanaTimeZone(tz: string): boolean {
-  if (typeof tz !== "string" || tz.trim() === "") return false;
-  try {
-    new Intl.DateTimeFormat("en-US", { timeZone: tz.trim() }).format();
-    return true;
-  } catch {
-    return false;
-  }
-}
-
-export function isValidResetHour(hour: unknown): hour is number {
-  return typeof hour === "number" && Number.isInteger(hour) && hour >= 0 && hour <= 23;
-}
-
-export function nodeDailyResetConfigured(timezone: unknown, hour: unknown): boolean {
-  return typeof timezone === "string" && isValidIanaTimeZone(timezone) && isValidResetHour(hour);
-}
-
-type ZonedParts = {
-  year: number;
-  month: number;
-  day: number;
-  hour: number;
-  minute: number;
-  second: number;
-};
-
-function zonedParts(ms: number, timeZone: string): ZonedParts {
-  const fmt = new Intl.DateTimeFormat("en-US", {
-    timeZone,
-    hourCycle: "h23",
-    year: "numeric",
-    month: "2-digit",
-    day: "2-digit",
-    hour: "2-digit",
-    minute: "2-digit",
-    second: "2-digit",
-  });
-  const bag: Record<string, string> = {};
-  for (const part of fmt.formatToParts(new Date(ms))) {
-    if (part.type !== "literal") bag[part.type] = part.value;
-  }
-  return {
-    year: Number(bag.year),
-    month: Number(bag.month),
-    day: Number(bag.day),
-    hour: Number(bag.hour),
-    minute: Number(bag.minute),
-    second: Number(bag.second),
-  };
-}
-
-function addCalendarDay(year: number, month: number, day: number): {
-  year: number;
-  month: number;
-  day: number;
-} {
-  const utc = Date.UTC(year, month - 1, day + 1);
-  const dt = new Date(utc);
-  return { year: dt.getUTCFullYear(), month: dt.getUTCMonth() + 1, day: dt.getUTCDate() };
-}
-
-/** Convert wall-clock time in `timeZone` to epoch ms. */
-function zonedLocalToUtc(
-  year: number,
-  month: number,
-  day: number,
-  hour: number,
-  minute: number,
-  second: number,
-  timeZone: string,
-): number {
-  const wanted = Date.UTC(year, month - 1, day, hour, minute, second);
-  let guess = wanted;
-  for (let i = 0; i < 4; i++) {
-    const p = zonedParts(guess, timeZone);
-    const asIfUtc = Date.UTC(p.year, p.month - 1, p.day, p.hour, p.minute, p.second);
-    const delta = asIfUtc - wanted;
-    if (delta === 0) return guess;
-    guess -= delta;
-  }
-  return guess;
-}
-
-/**
- * Next local `hour:00:00` in `timezone` strictly after `nowMs`.
- * If now lands exactly on that instant, return the following cycle.
- */
-export function nextDailyResetAtMs(timezone: string, hour: number, nowMs: number): number {
-  const now = zonedParts(nowMs, timezone);
-  let date = { year: now.year, month: now.month, day: now.day };
-  let next = zonedLocalToUtc(date.year, date.month, date.day, hour, 0, 0, timezone);
-  if (next <= nowMs) {
-    date = addCalendarDay(date.year, date.month, date.day);
-    next = zonedLocalToUtc(date.year, date.month, date.day, hour, 0, 0, timezone);
-  }
-  return next;
-}
-
-export function parseTpdLimitFromText(text: string): number | null {
-  const m = /limit:\s*(\d+)/i.exec(text);
-  if (!m) return null;
-  const n = Number(m[1]);
-  return Number.isFinite(n) ? n : null;
-}
-
-export function isTpdRateLimit(errorText: string | null | undefined): boolean {
-  return String(errorText || "")
-    .toLowerCase()
-    .includes("tpd rate limit");
-}
-
-export type TpdCooldownOptions = {
-  timezone?: unknown;
-  hour?: unknown;
-  nowMs?: number;
-  headerResetAtMs?: number | null;
-};
-
-/**
- * Cooldown for a TPD 429. Header reset wins; else the node clock.
- * Both missing → null (caller uses short 429, does not guess midnight).
- */
-export function resolveTpdCooldownMs(
-  errorText: string | null | undefined,
-  options: TpdCooldownOptions = {},
-): number | null {
-  if (!isTpdRateLimit(errorText)) return null;
-  const now = options.nowMs ?? Date.now();
-  if (typeof options.headerResetAtMs === "number" && options.headerResetAtMs > now) {
-    return options.headerResetAtMs - now;
-  }
-  if (typeof options.timezone === "string" && isValidResetHour(options.hour)) {
-    if (!nodeDailyResetConfigured(options.timezone, options.hour)) return null;
-    return nextDailyResetAtMs(options.timezone, options.hour, now) - now;
-  }
-  return null;
-}
--- a/open-sse/services/genericQuotaFetcher.ts
+++ b/open-sse/services/genericQuotaFetcher.ts
@@ -25,53 +25,10 @@ import {
  type QuotaInfo,
 } from "./quotaPreflight.ts";

-type UsageFetcher = (
-  connection: Parameters<typeof getUsageForProvider>[0],
-  options?: { forceRefresh?: boolean }
-) => Promise<unknown>;
-
-let usageFetcherOverride: UsageFetcher | null = null;
-
 // 60s — matches Codex's TTL. Long enough to avoid hammering upstream usage
 // endpoints on every routing decision, short enough that a near-exhausted
 // account is skipped within one minute of crossing its threshold.
 const CACHE_TTL_MS = 60_000;
-/** Drop unused force-refresh flags once inner provider caches (60s–5min) have expired. */
-const PENDING_FORCE_REFRESH_TTL_MS = CACHE_TTL_MS * 5;
-/** key → Date.now() when invalidate asked the next fetch to force-refresh. */
-const pendingForceRefresh = new Map<string, number>();
-/** key → last convert-null / throw while force-refresh was pending. */
-const pendingForceRefreshMiss = new Map<string, number>();
-
-/** Test-only: inject the usage dispatcher; pass null to restore. */
-export function __setGenericUsageFetcherForTests(fetcher: UsageFetcher | null): void {
-  usageFetcherOverride = fetcher;
-}
-
-/** Test-only: backdate a pending force-refresh so TTL expiry is unit-testable. */
-export function __agePendingForceRefreshForTests(
-  provider: string,
-  connectionId: string,
-  ageMs: number
-): void {
-  pendingForceRefresh.set(cacheKey(provider, connectionId), Date.now() - ageMs);
-}
-
-/** Test-only: backdate a convert-null miss so the 60s hammer-guard is unit-testable. */
-export function __agePendingForceRefreshMissForTests(
-  provider: string,
-  connectionId: string,
-  ageMs: number
-): void {
-  pendingForceRefreshMiss.set(cacheKey(provider, connectionId), Date.now() - ageMs);
-}
-
-/** Test-only: drop all wrapper/flag maps so tests cannot leak across ids. */
-export function __resetGenericQuotaFetcherForTests(): void {
-  cache.clear();
-  pendingForceRefresh.clear();
-  pendingForceRefreshMiss.clear();
-}

 interface CacheEntry {
  quota: QuotaInfo;
@@ -81,72 +38,15 @@ interface CacheEntry {
 const cache = new Map<string, CacheEntry>();

 function cacheKey(provider: string, connectionId: string): string {
-  return `${provider.trim()}::${connectionId.trim()}`;
+  return `${provider}::${connectionId}`;
 }

-function dropExpiredPendingForceRefresh(key: string, now: number): boolean {
-  const stampedAt = pendingForceRefresh.get(key);
-  if (stampedAt === undefined) return true;
-  if (now - stampedAt > PENDING_FORCE_REFRESH_TTL_MS) {
-    pendingForceRefresh.delete(key);
-    pendingForceRefreshMiss.delete(key);
-    return true;
-  }
-  return false;
-}
-
-// Lazy expiry on read — same as the provider breaker. Name stays `is*` because
-// callers only need a boolean; the map is not a public API.
-function isPendingForceRefresh(key: string, now: number = Date.now()): boolean {
-  if (dropExpiredPendingForceRefresh(key, now)) return false;
-  return pendingForceRefresh.has(key);
-}
-
-function markPendingForceRefreshMiss(key: string): void {
-  if (isPendingForceRefresh(key)) pendingForceRefreshMiss.set(key, Date.now());
-}
-
-function cachedQuotaIfFresh(
-  key: string,
-  forceRefresh: boolean,
-  now: number
-): QuotaInfo | null {
-  if (forceRefresh) return null;
-  const cached = cache.get(key);
-  if (cached && now - cached.fetchedAt < CACHE_TTL_MS) return cached.quota;
-  return null;
-}
-
-function isForceRefreshMissCooling(
-  key: string,
-  forceRefresh: boolean,
-  now: number
-): boolean {
-  if (!forceRefresh) return false;
-  const missedAt = pendingForceRefreshMiss.get(key);
-  return missedAt !== undefined && now - missedAt < CACHE_TTL_MS;
-}
-
-/** True when a concurrent 429 re-stamped a still-live flag during fetchUsage. */
-function isConcurrentForceRefresh(key: string, refreshStamp: number | undefined): boolean {
-  const currentStamp = pendingForceRefresh.get(key);
-  if (currentStamp === refreshStamp) return false;
-  return (
-    currentStamp !== undefined &&
-    Date.now() - currentStamp <= PENDING_FORCE_REFRESH_TTL_MS
-  );
-}
-
-// 5min — same as Codex. Expiry is lazy on read (`isPendingForceRefresh`);
-// this timer only reaps keys nobody fetches after the 5min TTL.
+// Auto-cleanup stale entries — same shape as codexQuotaFetcher.
 const _cacheCleanup = setInterval(() => {
  const now = Date.now();
  for (const [key, entry] of cache) {
    if (now - entry.fetchedAt > CACHE_TTL_MS * 5) cache.delete(key);
  }
-  for (const key of pendingForceRefresh.keys()) {
-    dropExpiredPendingForceRefresh(key, now);
-  }
 }, 5 * 60_000);
 if (typeof _cacheCleanup === "object" && "unref" in _cacheCleanup) {
  (_cacheCleanup as { unref?: () => void }).unref?.();
@@ -317,47 +217,24 @@ function normalizeQuotaWindows(
 export const fetchGenericQuota: QuotaFetcher = async (connectionId, connection) => {
  if (!connection) return null;
  const conn = connection as ConnectionInputs;
-  const provider = typeof conn.provider === "string" ? conn.provider.trim() : "";
+  const provider = typeof conn.provider === "string" ? conn.provider : null;
  if (!provider) return null;

  const key = cacheKey(provider, connectionId);
-  const now = Date.now();
-  const forceRefresh = isPendingForceRefresh(key, now);
-  const hit = cachedQuotaIfFresh(key, forceRefresh, now);
-  if (hit) return hit;
-  // convert-null / throw keep the force-refresh flag (agy inner caches are
-  // still stale) but must not hammer those endpoints on every routing tick.
-  if (isForceRefreshMissCooling(key, forceRefresh, now)) return null;
-
-  // Capture before await: a 429 during fetchUsage re-stamps this; writing
-  // the pre-429 snapshot would wipe that flag and recache stale quota.
-  const refreshStamp = pendingForceRefresh.get(key);
+  const cached = cache.get(key);
+  if (cached && Date.now() - cached.fetchedAt < CACHE_TTL_MS) {
+    return cached.quota;
+  }

  let usage: unknown;
  try {
-    const fetchUsage = usageFetcherOverride ?? getUsageForProvider;
-    usage = await fetchUsage(conn as Parameters<typeof getUsageForProvider>[0], {
-      ...(forceRefresh ? { forceRefresh: true } : {}),
-    });
+    usage = await getUsageForProvider(conn as Parameters<typeof getUsageForProvider>[0]);
  } catch {
-    markPendingForceRefreshMiss(key);
    return null;
  }

  const quota = convertUsageToQuotaInfo(usage);
-  if (!quota) {
-    markPendingForceRefreshMiss(key);
-    return null;
-  }
-
-  // Concurrent 429 re-stamped a still-live flag — do not recache the
-  // pre-429 snapshot. A vanished or expired stamp is not a 429.
-  if (isConcurrentForceRefresh(key, refreshStamp)) {
-    return quota;
-  }
-
-  pendingForceRefresh.delete(key);
-  pendingForceRefreshMiss.delete(key);
+  if (!quota) return null;

  // Refresh the static window catalog so the dashboard can render the right
  // modal inputs without waiting for the user to open the page.
@@ -373,33 +250,7 @@ export const fetchGenericQuota: QuotaFetcher = async (connectionId, connection)
 * fresh data instead of a 60s stale window.
 */
 export function invalidateGenericQuotaCache(provider: string, connectionId: string): void {
-  const key = cacheKey(provider, connectionId);
-  cache.delete(key);
-  // Next fetch must bypass provider-inner usage caches (agy retrieveUserQuota /
-  // weekly are 60s–5min). Without this, dropping the 60s wrapper recaches stale.
-  // TTL matches those inner caches: after 5min the flag is a no-op.
-  pendingForceRefresh.set(key, Date.now());
-  pendingForceRefreshMiss.delete(key);
-}
-
-/**
- * Drop the generic quota cache after an upstream 429, matching Codex's
- * `invalidateCodexQuotaCache` on 429. Probe-origin failures must not mutate
- * routing caches (#9817).
- */
-export function invalidateGenericQuotaCacheOnStatus(args: {
-  provider: string | null | undefined;
-  connectionId: string | null | undefined;
-  status: number;
-  isolateProbe?: boolean;
-}): boolean {
-  if (args.isolateProbe === true) return false; // undefined from callers that omit isolateProbe must still invalidate
-  if (args.status !== 429) return false;
-  const provider = typeof args.provider === "string" ? args.provider.trim() : "";
-  const connectionId = typeof args.connectionId === "string" ? args.connectionId.trim() : "";
-  if (!provider || !connectionId) return false;
-  invalidateGenericQuotaCache(provider, connectionId);
-  return true;
+  cache.delete(cacheKey(provider, connectionId));
 }

 /**
--- a/open-sse/services/moonshotQuotaFetcher.ts
+++ b/open-sse/services/moonshotQuotaFetcher.ts
@@ -1,229 +0,0 @@
-/**
- * moonshotQuotaFetcher.ts — Moonshot Open Platform balance quota fetcher
- *
- * GET {origin}/v1/users/me/balance
- *   { code: 0, data: { available_balance, voucher_balance, cash_balance } }
- *
- * Origin comes from the connection baseUrl (api.moonshot.cn or api.moonshot.ai).
- * Do not hardcode .ai as a fallback for .cn keys.
- *
- * Cache: 60s in-memory. Registration: registerMoonshotQuotaFetcher() at startup.
- */
-
-import { toNumber } from "@/shared/utils/numeric";
-import { registerQuotaFetcher, type QuotaInfo } from "./quotaPreflight.ts";
-import { registerMonitorFetcher } from "./quotaMonitor.ts";
-import { throttleQuotaFetch } from "./quotaFetchThrottle.ts";
-import {
-  isMoonshotOpenPlatformConnection,
-  moonshotBalanceUrl,
-  resolveMoonshotOrigin,
-} from "./usage/moonshotOpenPlatform.ts";
-import type { UsageQuota } from "./usage/quota.ts";
-
-const CACHE_TTL_MS = 60_000;
-
-export interface MoonshotQuota extends QuotaInfo {
-  availableBalance: number;
-  voucherBalance: number;
-  cashBalance: number;
-  origin: string;
-  limitReached: boolean;
-}
-
-interface CacheEntry {
-  quota: MoonshotQuota;
-  fetchedAt: number;
-}
-
-const quotaCache = new Map<string, CacheEntry>();
-
-const _cacheCleanup = setInterval(() => {
-  const now = Date.now();
-  for (const [key, entry] of quotaCache) {
-    if (now - entry.fetchedAt > CACHE_TTL_MS * 5) {
-      quotaCache.delete(key);
-    }
-  }
-}, 5 * 60_000);
-
-if (typeof _cacheCleanup === "object" && "unref" in _cacheCleanup) {
-  (_cacheCleanup as { unref?: () => void }).unref?.();
-}
-
-function toRecord(value: unknown): Record<string, unknown> {
-  return value && typeof value === "object" && !Array.isArray(value)
-    ? (value as Record<string, unknown>)
-    : {};
-}
-
-function parseMoonshotQuotaResponse(data: unknown, origin: string): MoonshotQuota | null {
-  const obj = toRecord(data);
-  const code = obj.code;
-  if (code !== 0 && code !== undefined) return null;
-  const payload = toRecord(obj.data);
-  if (!("available_balance" in payload) && !("availableBalance" in payload)) return null;
-  const availableBalance = toNumber(payload.available_balance ?? payload.availableBalance, 0);
-  const voucherBalance = toNumber(payload.voucher_balance ?? payload.voucherBalance, 0);
-  const cashBalance = toNumber(payload.cash_balance ?? payload.cashBalance, 0);
-  const limitReached = availableBalance <= 0;
-  const percentUsed = limitReached ? 1 : 0;
-  return {
-    used: percentUsed * 100,
-    total: 100,
-    percentUsed,
-    resetAt: null,
-    availableBalance,
-    voucherBalance,
-    cashBalance,
-    origin,
-    limitReached,
-    windows: { balance: { percentUsed, resetAt: null } },
-  };
-}
-
-function connectionApiKey(connection?: Record<string, unknown>): string | null {
-  const apiKey = connection?.apiKey;
-  return typeof apiKey === "string" && apiKey.trim().length > 0 ? apiKey : null;
-}
-
-export async function fetchMoonshotQuota(
-  connectionId: string,
-  connection?: Record<string, unknown>
-): Promise<QuotaInfo | null> {
-  const cached = quotaCache.get(connectionId);
-  if (cached && Date.now() - cached.fetchedAt < CACHE_TTL_MS) {
-    return cached.quota;
-  }
-
-  const apiKey = connectionApiKey(connection);
-  if (!apiKey) return null;
-
-  const origin = resolveMoonshotOrigin({
-    provider: typeof connection?.provider === "string" ? connection.provider : undefined,
-    providerSpecificData: connection?.providerSpecificData,
-  });
-  if (!origin) return null;
-
-  const url = moonshotBalanceUrl(origin);
-  const authHeader = ["Bearer", apiKey].join(" ");
-
-  try {
-    await throttleQuotaFetch();
-    const response = await fetch(url, {
-      method: "GET",
-      headers: {
-        Authorization: authHeader,
-        "Content-Type": "application/json",
-        Accept: "application/json",
-      },
-      signal: AbortSignal.timeout(8_000),
-    });
-
-    if (response.status === 401 || response.status === 403) {
-      quotaCache.delete(connectionId);
-      return null;
-    }
-    if (!response.ok) return null;
-
-    const data = await response.json();
-    const quota = parseMoonshotQuotaResponse(data, origin);
-    if (!quota) return null;
-    quotaCache.set(connectionId, { quota, fetchedAt: Date.now() });
-    return quota;
-  } catch {
-    return null;
-  }
-}
-
-export function invalidateMoonshotQuotaCache(connectionId: string): void {
-  quotaCache.delete(connectionId);
-}
-
-export type MoonshotUsageConnection = {
-  id?: string;
-  provider?: string;
-  apiKey?: string;
-  providerSpecificData?: unknown;
-};
-
-export async function getMoonshotOpenPlatformUsage(
-  connection: MoonshotUsageConnection
-): Promise<{
-  plan?: string;
-  quotas?: Record<string, UsageQuota>;
-  message?: string;
-  limitReached?: boolean;
-}> {
-  const origin = resolveMoonshotOrigin(connection);
-  if (!origin) {
-    return { message: "Not a Moonshot Open Platform connection." };
-  }
-  const quota = (await fetchMoonshotQuota(connection.id || "moonshot", {
-    apiKey: connection.apiKey,
-    provider: connection.provider,
-    providerSpecificData: connection.providerSpecificData,
-  })) as MoonshotQuota | null;
-  if (!quota) {
-    return { message: "Moonshot API key not available. Add a key to view usage." };
-  }
-  const domestic = origin.includes("moonshot.cn");
-  return {
-    plan: domestic ? "Kimi 开放平台（国内）" : "Kimi Open Platform",
-    quotas: buildMoonshotBalanceQuotas(quota, domestic ? "CNY" : "USD"),
-    limitReached: quota.limitReached,
-  };
-}
-
-function balanceQuota(
-  remaining: number,
-  remainingPercentage: number,
-  currency: string
-): UsageQuota {
-  return {
-    used: 0,
-    total: 0,
-    remaining,
-    remainingPercentage,
-    resetAt: null,
-    unlimited: true,
-    currency,
-  };
-}
-
-function buildMoonshotBalanceQuotas(
-  quota: MoonshotQuota,
-  currency: string
-): Record<string, UsageQuota> {
-  return {
-    available: balanceQuota(quota.availableBalance, quota.limitReached ? 0 : 100, currency),
-    voucher: balanceQuota(quota.voucherBalance, 100, currency),
-    cash: balanceQuota(quota.cashBalance, 100, currency),
-  };
-}
-
-export function registerMoonshotQuotaFetcher(): void {
-  registerQuotaFetcher("moonshot", fetchMoonshotQuota);
-  registerQuotaFetcher("kimi", fetchMoonshotQuota);
-  registerMonitorFetcher("moonshot", fetchMoonshotQuota);
-  registerMonitorFetcher("kimi", fetchMoonshotQuota);
-}
-
-export function registerMoonshotFetchersForNodes(
-  nodes: Array<{ id?: string | null; prefix?: string | null; baseUrl?: string | null }>
-): void {
-  for (const node of nodes) {
-    const origin = resolveMoonshotOrigin({}, node.baseUrl);
-    if (!origin) continue;
-    if (typeof node.id === "string" && node.id) {
-      registerQuotaFetcher(node.id, fetchMoonshotQuota);
-      registerMonitorFetcher(node.id, fetchMoonshotQuota);
-    }
-    if (typeof node.prefix === "string" && node.prefix) {
-      registerQuotaFetcher(node.prefix, fetchMoonshotQuota);
-      registerMonitorFetcher(node.prefix, fetchMoonshotQuota);
-    }
-  }
-}
-
-export { isMoonshotOpenPlatformConnection };
--- a/open-sse/services/quotaPreflight.ts
+++ b/open-sse/services/quotaPreflight.ts
@@ -21,22 +21,12 @@
 import { isCompatibleProviderConnectionId } from "@/shared/utils/compatibleProviderId";
 import { isFeatureFlagEnabled } from "@/shared/utils/featureFlags";
 import { fetchNewApiAggregatorQuota } from "./newApiAggregatorQuotaFetcher.ts";
-import {
-  isAntigravityQuotaProvider,
-  selectAntigravityQuotaWindowNames,
-} from "./antigravityQuotaFamily.ts";

 export interface PreflightQuotaResult {
  proceed: boolean;
  reason?: string;
  quotaPercent?: number;
  resetAt?: string | null;
-  windowName?: string | null;
-}
-
-export interface QuotaCutoffScope {
-  provider?: string | null;
-  requestedModel?: string | null;
 }

 export interface QuotaWindowInfo {
@@ -166,36 +156,15 @@ function isRemainingAtOrBelowThreshold(
  return remainingPercent <= thresholdPercent + REMAINING_PERCENT_EPSILON;
 }

-function exhaustedResult(
-  quotaPercent: number,
-  resetAt: string | null,
-  windowName?: string | null
-): PreflightQuotaResult {
+function exhaustedResult(quotaPercent: number, resetAt: string | null): PreflightQuotaResult {
  return {
    proceed: false,
    reason: "quota_exhausted",
    quotaPercent,
    resetAt,
-    windowName: windowName ?? null,
  };
 }

-function windowsForScope(
-  windows: NonNullable<QuotaInfo["windows"]>,
-  scope?: QuotaCutoffScope
-): NonNullable<QuotaInfo["windows"]> {
-  if (!scope?.requestedModel || !isAntigravityQuotaProvider(scope.provider ?? null)) {
-    return windows;
-  }
-  const selected = selectAntigravityQuotaWindowNames(Object.keys(windows), scope.requestedModel);
-  if (selected.length === 0) return windows;
-  const scoped: NonNullable<QuotaInfo["windows"]> = {};
-  for (const name of selected) {
-    if (windows[name]) scoped[name] = windows[name];
-  }
-  return Object.keys(scoped).length > 0 ? scoped : windows;
-}
-
 function limitReachedResult(quota: QuotaInfo): PreflightQuotaResult {
  return exhaustedResult(
    Number.isFinite(quota.percentUsed) ? quota.percentUsed : 1,
@@ -232,9 +201,7 @@ function quotaWindowCutoffResult(
    worstResetAt = windowInfo.resetAt ?? null;
  }

-  return worstWindow === null
-    ? null
-    : exhaustedResult(worstUsedPercent, worstResetAt, worstWindow);
+  return worstWindow === null ? null : exhaustedResult(worstUsedPercent, worstResetAt);
 }

 function quotaPercentCutoffResult(
@@ -260,27 +227,21 @@ function quotaPercentCutoffResult(
 */
 export function evaluateQuotaCutoff(
  quota: QuotaInfo | null | undefined,
-  thresholds?: PreflightQuotaThresholds,
-  scope?: QuotaCutoffScope
+  thresholds?: PreflightQuotaThresholds
 ): PreflightQuotaResult {
  if (!quota) return { proceed: true };
+  if (quota.limitReached === true) return limitReachedResult(quota);

  const windows = quota.windows;
  if (windows && Object.keys(windows).length > 0) {
-    const scopedWindows = windowsForScope(windows, scope);
-    const cutoff = quotaWindowCutoffResult(scopedWindows, thresholds);
-    if (cutoff) return cutoff;
-    if (isAntigravityQuotaProvider(scope?.provider ?? null) && scope?.requestedModel) {
-      return { proceed: true, quotaPercent: quota.percentUsed };
-    }
-    if (quota.limitReached === true) return limitReachedResult(quota);
-    return {
-      proceed: true,
-      quotaPercent: quota.percentUsed,
-    };
+    return (
+      quotaWindowCutoffResult(windows, thresholds) ?? {
+        proceed: true,
+        quotaPercent: quota.percentUsed,
+      }
+    );
  }

-  if (quota.limitReached === true) return limitReachedResult(quota);
  return quotaPercentCutoffResult(quota, thresholds);
 }

@@ -336,40 +297,61 @@ export async function preflightQuota(
    return { proceed: true };
  }

-  const requestedModel =
-    typeof connection.requestedModel === "string" ? connection.requestedModel : null;
-  const scope: QuotaCutoffScope = { provider, requestedModel };
-  const windows = quota.windows;
-  if (windows && Object.keys(windows).length > 0) {
-    const scopedWindows = windowsForScope(windows, scope);
-    for (const [windowName, windowInfo] of Object.entries(scopedWindows)) {
+  if (quota.limitReached === true) {
+    return limitReachedResult(quota);
+  }
+
+  // Per-window evaluation — only when the fetcher surfaces a windows map.
+  // We block as soon as ANY single window's remaining quota drops to its
+  // configured cutoff or below; warnings are logged independently per window.
+  if (quota.windows && Object.keys(quota.windows).length > 0) {
+    let worstUsedPercent = 0;
+    let worstWindow: string | null = null;
+    let worstResetAt: string | null = null;
+    for (const [windowName, windowInfo] of Object.entries(quota.windows)) {
+      const minRemainingPercent = resolveOrDefault(
+        thresholds?.resolveMinRemainingPercent,
+        windowName,
+        DEFAULT_MIN_REMAINING_PERCENT
+      );
      const warnRemainingPercent = resolveOrDefault(
        thresholds?.resolveWarnRemainingPercent,
        windowName,
        DEFAULT_WARN_REMAINING_PERCENT
      );
      const remainingPercent = remainingPercentFrom(windowInfo.percentUsed);
-      if (isRemainingAtOrBelowThreshold(remainingPercent, warnRemainingPercent)) {
+
+      if (isRemainingAtOrBelowThreshold(remainingPercent, minRemainingPercent)) {
+        // Track the most-depleted blocking window so the response can name it.
+        if (windowInfo.percentUsed > worstUsedPercent) {
+          worstUsedPercent = windowInfo.percentUsed;
+          worstWindow = windowName;
+          worstResetAt = windowInfo.resetAt ?? null;
+        } else if (worstWindow === null) {
+          worstWindow = windowName;
+          worstResetAt = windowInfo.resetAt ?? null;
+        }
+      } else if (isRemainingAtOrBelowThreshold(remainingPercent, warnRemainingPercent)) {
        console.warn(
          `[QuotaPreflight] ${provider}/${connectionId} ${windowName}: ${remainingPercent.toFixed(1)}% remaining — approaching cutoff`
        );
      }
    }
-  }

-  const decision = evaluateQuotaCutoff(quota, thresholds, scope);
-  if (!decision.proceed) {
-    const windowLabel = decision.windowName ? ` ${decision.windowName}` : "";
-    const remaining = Number.isFinite(decision.quotaPercent)
-      ? remainingPercentFrom(decision.quotaPercent as number).toFixed(1)
-      : "?";
-    console.info(
-      `[QuotaPreflight] ${provider}/${connectionId}${windowLabel}: ${remaining}% remaining - switching`
-    );
-    return decision;
-  }
-  if (windows && Object.keys(windows).length > 0) {
-    return decision;
+    if (worstWindow !== null) {
+      const worstRemaining = remainingPercentFrom(worstUsedPercent);
+      console.info(
+        `[QuotaPreflight] ${provider}/${connectionId} ${worstWindow}: ${worstRemaining.toFixed(1)}% remaining — switching`
+      );
+      return {
+        proceed: false,
+        reason: "quota_exhausted",
+        quotaPercent: worstUsedPercent,
+        resetAt: worstResetAt,
+      };
+    }
+
+    return { proceed: true, quotaPercent: quota.percentUsed };
  }

  // Legacy single-signal path for fetchers that don't expose per-window data.
--- a/open-sse/services/usage.ts
+++ b/open-sse/services/usage.ts
@@ -61,8 +61,6 @@ import { getQoderUsage, parseQoderUserStatusUsage } from "./usage/qoder.ts";
 export { parseQoderUserStatusUsage } from "./usage/qoder.ts";
 import { getOpencodeUsage } from "./usage/opencode.ts";
 import { getDeepseekUsage } from "./usage/deepseek.ts";
-import { getMoonshotOpenPlatformUsage } from "./moonshotQuotaFetcher.ts";
-import { isMoonshotOpenPlatformConnection } from "./usage/moonshotOpenPlatform.ts";
 import { getDevinCliUsage } from "./usage/devinCli.ts";
 import { getBailianCodingPlanUsage } from "./usage/bailian.ts";
 import { getVertexUsage } from "./usage/vertex.ts";
@@ -113,10 +111,6 @@ export async function getUsageForProvider(
 ) {
  const { id, provider, accessToken, apiKey, providerSpecificData, projectId, email } = connection;

-  if (isMoonshotOpenPlatformConnection(connection)) {
-    return await getMoonshotOpenPlatformUsage(connection);
-  }
-
  switch (provider) {
    case "github":
      return await getGitHubUsage(accessToken, providerSpecificData);
@@ -174,9 +168,6 @@ export async function getUsageForProvider(
      return await getNanoGptUsage(apiKey || "");
    case "deepseek":
      return await getDeepseekUsage(id || "", apiKey || "");
-    case "moonshot":
-    case "kimi":
-      return await getMoonshotOpenPlatformUsage(connection);
    case "openrouter":
      return await getOpenrouterUsage(id || "", apiKey || "", providerSpecificData);
    case "opencode":
--- a/open-sse/services/usage/fetcherProviders.ts
+++ b/open-sse/services/usage/fetcherProviders.ts
@@ -45,8 +45,6 @@ export const USAGE_FETCHER_PROVIDERS = [
  "qwen-cloud-token-plan",
  "nanogpt",
  "deepseek",
-  "moonshot",
-  "kimi",
  "opencode",
  "opencode-zen",
  "xiaomi-mimo",
--- a/open-sse/services/usage/grokCli.ts
+++ b/open-sse/services/usage/grokCli.ts
@@ -239,12 +239,9 @@ export async function getGrokCliUsage(accessToken?: string) {
  const config = billing.config;
  const resetAt = config.currentPeriod?.end || null;
  const quotas: Record<string, ReturnType<typeof percentageQuota>> = {};
-  // SuperGrokPro (and proto3 omit-zero) billing configs often omit
-  // creditUsagePercent / productUsage. A present config object is a
-  // successful billing read, so treat a missing percent as 0% used and
-  // still render a weekly bar. A missing config still returns
-  // "Grok Build billing status unavailable" above — that path is unchanged.
-  quotas.weekly = percentageQuota(config.creditUsagePercent ?? 0, resetAt);
+  if (config.creditUsagePercent != null) {
+    quotas.weekly = percentageQuota(config.creditUsagePercent, resetAt);
+  }
  Object.assign(quotas, buildProductQuotas(config.productUsage, resetAt));

  const autoTopUpResponse = userId
--- a/open-sse/services/usage/moonshotOpenPlatform.ts
+++ b/open-sse/services/usage/moonshotOpenPlatform.ts
@@ -1,76 +0,0 @@
-/**
- * Moonshot Open Platform host recognition.
- *
- * Distinguishes prepaid Open Platform keys (api.moonshot.cn / api.moonshot.ai)
- * from Kimi Coding Plan (api.kimi.com/coding). Custom compatible nodes are
- * identified by baseUrl host, not by provider id (those ids are uuids).
- */
-
-import { moonshotProvider } from "../../config/providers/registry/moonshot/index.ts";
-import { kimiProvider } from "../../config/providers/registry/kimi/index.ts";
-
-export const MOONSHOT_OPEN_PLATFORM_HOSTS: ReadonlySet<string> = new Set([
-  "api.moonshot.cn",
-  "api.moonshot.ai",
-]);
-
-export type MoonshotOriginConnection = {
-  provider?: string;
-  providerSpecificData?: unknown;
-};
-
-function asRecord(value: unknown): Record<string, unknown> {
-  return value && typeof value === "object" && !Array.isArray(value)
-    ? (value as Record<string, unknown>)
-    : {};
-}
-
-export function parseMoonshotOrigin(baseUrl: string | null | undefined): string | null {
-  if (typeof baseUrl !== "string" || baseUrl.trim() === "") return null;
-  let url: URL;
-  try {
-    url = new URL(baseUrl.trim());
-  } catch {
-    return null;
-  }
-  if (url.protocol !== "https:" && url.protocol !== "http:") return null;
-  const host = url.hostname.toLowerCase();
-  if (!MOONSHOT_OPEN_PLATFORM_HOSTS.has(host)) return null;
-  const port = url.port ? `:${url.port}` : "";
-  return `${url.protocol}//${host}${port}`;
-}
-
-export function moonshotBalanceUrl(origin: string): string {
-  return `${origin}/v1/users/me/balance`;
-}
-
-function registryDefaultOrigin(provider: string | undefined): string | null {
-  if (provider === "moonshot") return parseMoonshotOrigin(moonshotProvider.baseUrl);
-  if (provider === "kimi") return parseMoonshotOrigin(kimiProvider.baseUrl);
-  return null;
-}
-
-export function resolveMoonshotOrigin(
-  connection: MoonshotOriginConnection,
-  nodeBaseUrl?: string | null,
-): string | null {
-  const psd = asRecord(connection.providerSpecificData);
-  const fromPsd = typeof psd.baseUrl === "string" ? parseMoonshotOrigin(psd.baseUrl) : null;
-  if (fromPsd) return fromPsd;
-  const fromNode = parseMoonshotOrigin(nodeBaseUrl);
-  if (fromNode) return fromNode;
-  return registryDefaultOrigin(connection.provider);
-}
-
-export function isMoonshotOpenPlatformConnection(
-  connection: MoonshotOriginConnection,
-  nodeBaseUrl?: string | null,
-): boolean {
-  return resolveMoonshotOrigin(connection, nodeBaseUrl) !== null;
-}
-
-/** Account-level empty wallet on Open Platform. Narrower than any compatible 429. */
-export function isMoonshotAccountBalanceExhausted(errorText: string | null | undefined): boolean {
-  const lower = String(errorText || "").toLowerCase();
-  return lower.includes("insufficient balance") || lower.includes("exceeded_current_quota");
-}
--- a/open-sse/services/usage/supportedProviders.ts
+++ b/open-sse/services/usage/supportedProviders.ts
@@ -41,8 +41,6 @@ export const USAGE_SUPPORTED_PROVIDERS: readonly string[] = [
  "crof",
  "nanogpt",
  "deepseek",
-  "moonshot",
-  "kimi",
  "xiaomi-mimo",
  "xiaomi-mimo-token-plan",
  "vertex",
--- a/open-sse/translator/response/openai-responses.ts
+++ b/open-sse/translator/response/openai-responses.ts
@@ -23,12 +23,12 @@ import {
 import { createEventEmitter } from "./openai-responses/eventEmitter.ts";
 import { buildResponsesToolCallItem } from "./responsesToolItem.ts";
 import { resolveRequestToolIdentity } from "./openai-responses/requestToolIdentity.ts";
-import { resolveLocalToolCallIndex } from "./openai-responses/toolCallLocalIndex.ts";
 import {
  synthesizeCompletedToolCalls,
  computeFinishReason,
  withAssistantRoleOnFirstDelta,
 } from "./openai-responses/synthesizeCompletedToolCalls.ts";
+
 // normalizeUpstreamFailure is re-exported for external importers (tests).
 export { normalizeUpstreamFailure } from "./openai-responses/pureHelpers.ts";

@@ -506,7 +506,7 @@ function toolCallOutputIndexBase(state) {

 function emitToolCall(state, emit, tc) {
  const tcIdx = tc.index ?? 0;
-  const outputIndex = toolCallOutputIndexBase(state) + resolveLocalToolCallIndex(state, tcIdx);
+  const outputIndex = toolCallOutputIndexBase(state) + normalizeOutputIndex(tcIdx);
  const newCallId = tc.id;
  const funcName = tc.function?.name;

@@ -609,7 +609,7 @@ function emitToolCall(state, emit, tc) {
 function closeToolCall(state, emit, idx, recordAsCompleted = true) {
  const callId = state.funcCallIds[idx];
  if (callId && !state.funcItemDone[idx]) {
-    const normalizedIndex = toolCallOutputIndexBase(state) + resolveLocalToolCallIndex(state, idx);
+    const normalizedIndex = toolCallOutputIndexBase(state) + normalizeOutputIndex(idx);
    const args = state.funcArgsBuf[idx] || "{}";
    const toolName = state.funcNames[idx] || "";
    // See emitToolCall()'s isCustomTool comment — must stay in sync (both compute the
--- a/open-sse/translator/response/openai-responses/toolCallLocalIndex.ts
+++ b/open-sse/translator/response/openai-responses/toolCallLocalIndex.ts
@@ -1,35 +0,0 @@
-/**
- * Remap a turn's raw upstream tool_calls delta `index` onto a local,
- * contiguous, 0-based sequence in first-seen order.
- *
- * Live incident (2026-09-02, minimax-m3:free via OpenRouter/GMICloud): the
- * upstream's own `index` doesn't reliably start at 0 or stay contiguous per
- * turn — this turn's two calls arrived with raw index 1 and 2 (never 0).
- * Adding that raw index straight onto toolCallOutputIndexBase() left a GAP
- * in the emitted output_index sequence (0 for the message, then 2 and 3 for
- * the calls — index 1 never used). A client that reads response.completed's
- * final `output[]` array by ARRAY POSITION and expects position to equal
- * output_index (the Responses API's own contract) reads output[1] (this
- * turn's first call, real output_index 2) while looking it up under
- * output_index 1, misses it, then reads output[2] (the second call, real
- * output_index 3) under output_index 2 — landing on the FIRST call's tracked
- * slot with a different call_id, which a spec-following client correctly
- * treats as "stream changed output item identity" and aborts.
- */
-
-export type ToolCallLocalIndexState = {
-  toolCallLocalIndex?: Record<string, number>;
-  toolCallLocalIndexNext?: number;
-};
-
-export function resolveLocalToolCallIndex(
-  state: ToolCallLocalIndexState,
-  tcIdx: string | number
-): number {
-  if (!state.toolCallLocalIndex) state.toolCallLocalIndex = {};
-  if (state.toolCallLocalIndex[tcIdx] === undefined) {
-    state.toolCallLocalIndex[tcIdx] = state.toolCallLocalIndexNext ?? 0;
-    state.toolCallLocalIndexNext = state.toolCallLocalIndex[tcIdx] + 1;
-  }
-  return state.toolCallLocalIndex[tcIdx];
-}
--- a/open-sse/utils/resourcePressure.ts
+++ b/open-sse/utils/resourcePressure.ts
@@ -67,47 +67,9 @@ function requireDuration(name: string, value: number): number {
  return value;
 }

-/**
- * Human-readable key=value detail appended to the rejection log line. Every
- * rejection (immediate heap trip AND cached-critical-state reuse) goes
- * through here, so this is the one place that needs the actual numbers —
- * the bare reason code alone ("psi_some") gives an operator nothing to act
- * on when deciding whether the guard is mistuned vs. genuinely saturated.
- */
-function formatPressureDetail(detail: Record<string, number | string | null | undefined>): string {
-  return Object.entries(detail)
-    .filter(([, value]) => value !== undefined)
-    .map(([key, value]) => `${key}=${value ?? "null"}`)
-    .join(" ");
-}
-
-/** Builds buildCriticalGuard's detail object for the cached-critical-state
- * reuse path in check() -- pulled out of check() itself so that function's
- * own cyclomatic complexity stays under the ratchet, not because this needs
- * to be reused anywhere else. */
-function describeCachedPressure(params: {
-  signals: ResourceSignals | null;
-  recoveryStreak: number;
-  cacheAgeMs: number;
-}): Record<string, number | string | null> {
-  const cgroup = params.signals?.cgroup;
-  return {
-    psiSomeAvg10: params.signals?.psi?.someAvg10 ?? null,
-    psiFullAvg10: params.signals?.psi?.fullAvg10 ?? null,
-    cgroupCurrentMb: cgroup?.currentBytes ? Math.round(cgroup.currentBytes / MB) : null,
-    cgroupMaxMb: cgroup?.maxBytes ? Math.round(cgroup.maxBytes / MB) : null,
-    recoveryStreak: params.recoveryStreak,
-    sampleAgeMs: params.cacheAgeMs,
-  };
-}
-
-function buildCriticalGuard(
-  reason: PressureReason,
-  detail: Record<string, number | string | null | undefined> = {}
-): ResourcePressureGuardResult {
-  const detailText = formatPressureDetail(detail);
+function buildCriticalGuard(reason: PressureReason): ResourcePressureGuardResult {
  console.warn(
-    `[resourcePressure] critical pressure guard tripped (reason=${reason}${detailText ? " " + detailText : ""}); returning 503`
+    `[resourcePressure] critical pressure guard tripped (reason=${reason}); returning 503`
  );
  return {
    success: false,
@@ -135,10 +97,7 @@ function immediateHeapGuard(
  if (thresholdMb == null) return null;
  const guard = checkHeapPressureGuard(heapUsedMb, thresholdMb);
  if (!guard) return null;
-  return buildCriticalGuard("v8_heap_absolute", {
-    heapUsedMb: Math.round(heapUsedMb),
-    thresholdMb: Math.round(thresholdMb),
-  });
+  return buildCriticalGuard("v8_heap_absolute");
 }

 export function createResourcePressureRuntime(
@@ -233,17 +192,9 @@ export function createResourcePressureRuntime(
        return immediate;
      }
      const cacheAge = lastSignals ? Math.max(0, now - lastRefreshAtMs) : Number.POSITIVE_INFINITY;
-      if (cacheAge > maxStaleMs || state.severity !== "critical") {
-        return null;
-      }
-      return buildCriticalGuard(
-        state.reason,
-        describeCachedPressure({
-          signals: lastSignals,
-          recoveryStreak: state.recoveryStreak,
-          cacheAgeMs: cacheAge,
-        })
-      );
+      return cacheAge <= maxStaleMs && state.severity === "critical"
+        ? buildCriticalGuard(state.reason)
+        : null;
    },
    getObservation: () => ({ signals: lastSignals, state }),
    whenRefreshSettled: async () => {
--- a/Show More
+++ b/Show More
				`@@ -1 +0,0 @@`
				- feat(providers): add SeekAi (`seekai.cc`) as an OpenAI-compatible New-API gateway — catalog id `seekai` (alias `ska`), `https://seekai.cc/v1`, live `/v1/models` via `passthroughModels`, aggregator-list membership so New-API balance detection can opt in. No referral/aff codes. ([#11786](https://github.com/diegosouzapw/OmniRoute/issues/11786))
				`@@ -1 +0,0 @@`
				`- feat(providers): import-from-file modal shows per-row API errors and ships a downloadable CSV template ([#12071](https://github.com/diegosouzapw/OmniRoute/issues/12071))`
				`@@ -1 +0,0 @@`
				- feat(providers): dashboard search matches connection name and `baseUrl` so imported OpenAI-compat nodes surface on the provider card ([#12108](https://github.com/diegosouzapw/OmniRoute/issues/12108))
				`@@ -1 +0,0 @@`
				- feat(settings): persist `headroomUrl` through Settings so status/start use the operator URL instead of only `HEADROOM_URL` ([#12306](https://github.com/diegosouzapw/OmniRoute/issues/12306))
				`@@ -1 +0,0 @@`
				- fix(grok-cli): treat omitted SuperGrokPro `creditUsagePercent` as 0% used so Provider Limits still renders a weekly bar (proto3 zero-elision) ([#12312](https://github.com/diegosouzapw/OmniRoute/pull/12312)) — thanks @HouMinXi
				`@@ -1 +0,0 @@`
				`- fix(quota): drop the generic quota cache (agy / Antigravity / Claude OAuth) on an upstream 429 so reset-aware scoring does not keep a 60s stale snapshot, and force-refresh the next usage fetch so inner provider caches cannot recache the same window ([#12325](https://github.com/diegosouzapw/OmniRoute/pull/12325)) — thanks @HouMinXi`
				`@@ -1 +0,0 @@`
				- chore(providers): bump the Claude Code wire identity and the Devin bridge image pin from `2.1.220` to `2.1.258` ([#12402](https://github.com/diegosouzapw/OmniRoute/pull/12402)) — thanks @ggiak
				`@@ -1 +0,0 @@`
				- chore(quality): rebaseline the file-size caps the hartmark batch grew past (`combos/page.tsx` via [#12355](https://github.com/diegosouzapw/OmniRoute/pull/12355), `open-sse/services/combo.ts` via [#12338](https://github.com/diegosouzapw/OmniRoute/pull/12338))