mirror of
https://github.com/diegosouzapw/OmniRoute.git
synced 2026-08-19 13:42:09 +03:00
368d1c0e87b51b64e7c55d4dfd8f800c9bbafcf4
241 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
rafaeldrincon
|
66997a6404 |
fix(logs): avoid giant provider pills for failed auto family requests (#8867)
* fix(logs): avoid giant provider pills for failed auto family requests * refactor(logs): extract resolveRejectedComboProvider + cover it The provider label was decided inline in handleChat, which has no test harness — the change shipped untested and pushed chat.ts over its frozen size (1848 > 1845). Moved to rejectedRequestUsage.ts next to summarizeComboAttemptedModels, the helper it replaces on this path. chat.ts shrinks back under its baseline (no rebaseline needed) and the logic gets three cases in the suite that already covers its sibling: auto/* collapses to "auto", a named combo keeps its name, and bare "auto" (no slash) is NOT collapsed — that one is a combo request, not a family request. Also rebased on the current release tip and added the changelog fragment. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com> Co-authored-by: rafaeldrincon <rafaeldrincon@users.noreply.github.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
rafaeldrincon
|
bf77648071 |
fix(autoRouting): recognize auto/\<family\> combos in classifyAutoModel (#8866)
* fix(autoRouting): recognize auto/\<family\> combos in classifyAutoModel
classifyAutoModel() checks VALID_AUTO_VARIANTS and parseAutoSuffix but
never isValidModelFamily, so auto/glm, auto/minimax, auto/llama etc. are
rejected as "Unknown built-in auto combo" before chatHelpers.ts or
builtinCatalog.ts can handle them.
Fix: import isValidModelFamily and ModelFamily, add family to spec type,
check family suffixes before returning unrecognized. Mirrors the pattern
already in builtinCatalog.ts createBuiltinAutoCombo.
Closes: auto/\<family\> combos listed in /api/combos/auto but unusable
at /v1/chat/completions.
* test(autoRouting): cover auto/<family> classification + changelog fragment
The PR changed production code with no test — nothing in tests/ referenced
classifyAutoModel. Since it is module-private, the new suite exercises it through
the public resolveAutoRoutingState().
Verified it guards something real: against the release tip without this fix the
family case fails ("auto/glm should be a recognized built-in auto model"), and
passes with it. Also pins that a category suffix does not pick up spec.family and
that an unknown suffix stays unrecognized.
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
---------
Co-authored-by: rafaeldrincon <rafaeldrincon@users.noreply.github.com>
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
|
||
|
Wilson
|
f8487648c8 | fix(resilience): keep resource 404s from cooling models (#8756) | ||
|
glaze
|
01c0f8a7dd |
fix(providers): recover Kimi after quota reset (#8632)
* fix(providers): recover Kimi after quota reset * docs: add Kimi quota recovery changelog |
||
|
Éder Costa
|
9108955323 |
fix(claude): classify native subscription quota 429 (#8628)
Co-authored-by: Escalada Online <aescaladaonline@gmail.com> Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com> |
||
|
backryun
|
da24c58c3a |
Train 1D: merge via --admin on .113 validation
Squash merge from local merge-train (Hard Rule owner-approved). Tip 029cdf4215cf465f0e1716ac9f84a84692b1e881 validated on 192.168.0.113: 26631/26653 pass. |
||
|
backryun
|
847d1ad84c |
Train 1D: merge via --admin on .113 validation
Squash merge from local merge-train (Hard Rule owner-approved). Tip 029cdf4215cf465f0e1716ac9f84a84692b1e881 validated on 192.168.0.113: 26631/26653 pass. |
||
|
Diego Rodrigues de Sa e Souza
|
cd9a631464 |
feat: Claude Code discovery aliases (surface non-Claude models in the /model picker) (#8666)
* feat(db): cc discovery alias gate storage + EXPOSE_CC_DISCOVERY_ALIASES flag
Adds the gate for claude/<provider>/<model> discovery-alias mirror ids on
the /v1/models catalog: a new runtime feature flag (env forces on and wins
over the dashboard DB override), per-provider and per-model "on"/"off"/null
overrides stored in key_value under the ccDiscoveryAliases namespace, and a
pure precedence resolver (model > provider > global). Catalog wiring is a
separate follow-up task; this only lands the gate + storage.
* feat(sse): synthesize claude/ discovery aliases for the model catalog
* feat(api): advertise cc discovery aliases on /v1/models behind the 3-level gate
* feat(sse): resolve claude/ discovery aliases on the request path
* fix(sse): import getComboByName from db/combos, not the localDb barrel
* fix(sse): cover custom-node prefixes and the Codex WS bridge in cc discovery alias resolution
* feat(dashboard): cc discovery alias toggles + flag-screen env warning
Adds the operator-facing UI/API layer for the Claude Code discovery-alias
gate (claude/<provider>/<model> mirror ids on /v1/models): REST endpoint
for provider/model overrides, a provider-detail card with 3-state
(inherit/on/off) toggles, an info button on the Claude Code tool card
linking to Feature Flags, and an env-source warning on the
EXPOSE_CC_DISCOVERY_ALIASES flag card when it's forced on via env.
* feat(api): cc discovery usage metrics
* fix(api): record cc alias metric in the production wrapper + atomic counter upsert
* docs: document cc discovery aliases (Claude Code guide + feature flag catalog)
* fix(sse): don't mirror built-in auto/* combos as discovery aliases (advertised-but-unroutable)
* i18n(vi): translate the discovery-alias strings instead of shipping placeholders
vi is the one locale with a strict "no internal missing markers" test, so the 17
__MISSING__ entries this branch added (the provider ccAlias panel, the info
button, the feature-flag description and the env warning) would have turned that
test red the moment the base itself was repaired. Translated, keeping every ICU
placeholder ({modelId}, {error}) and the literal claude/<provider>/<model> id
shape intact.
* chore(quality): raise the frozen caps this feature legitimately grows
catalog.ts 1615 -> 1639: the alias synthesis is wired into the catalog builder,
which is where the per-key-filtered list is assembled — the only place the mirror
entries can be appended after model hiding has been applied.
localDb.ts 808 -> 810: two re-export lines for the new ccDiscoveryAliases db
module, which is exactly what the "Adding a New DB Module" recipe prescribes.
* refactor(dashboard,api): keep the complexity ratchets flat
The feature added four cyclomatic violations and one cognitive one, which the
ratchets reject — the baseline only moves when a metric improves. Split the new
code instead:
- appendCcDiscoveryAliases: the four skip-guards become isMirrorableId().
- resolveCcDiscoveryAliasStripWith: alias parsing and gate resolution become
parseCcAliasTarget() and resolveGateFor(), replacing a chain of ternaries that
each re-tested isComboAlias.
- FeatureFlagCard: the env-precedence warning becomes its own component instead
of a conditional branch inside an already-large render.
- ProviderCcAliasSection: the loader moves to useCcAliasData(), and the override
list and add-row become ModelOverrideList / AddOverrideRow, bringing both
oversized functions back under the 80-line rule.
Behavior unchanged — the 74 discovery-alias tests pass untouched. Both ratchets
now sit exactly at baseline (2188 / 971).
|
||
|
Diego Rodrigues de Sa e Souza
|
7f8a59ac66 |
fix: repair five base-red failures on release/v3.8.49 (#8706)
* fix: repair five base-red failures on release/v3.8.49 Every PR cut from this branch fails CI on the branch's own breakage. Five distinct causes, none introduced by the PRs that trip over them: 1. dast-smoke / Turbopack build — src/sse/handlers/chat.ts imported PROVIDER_BREAKER_FAILURE_STATUSES twice in one statement. A duplicate import specifier is an ECMAScript syntax error, so the production build never compiled. Introduced by #8258, whose export fix landed on top of an import that already existed. 2. Unit Tests — the #8393 verified-cooldown bypass was renamed exactCooldownVerified -> exactCooldownIsUpstreamReset during the #8254 conflict resolution, which also dropped the flag at the markAccountUnavailable call site entirely. The rename left the test passing the old key (so the flag was silently ignored and a verified upstream reset got clamped back to maxCooldownMs), and the dropped call site meant no real caller set it at all. Align the test on the surviving name, restore the call site, and restore the doc comment explaining #6863 vs #7940. 3. Unit Tests — #8526 added four common.* keys to en.json only, breaking the strict key-parity tests for pt-BR and vi. Translated into all 42 locales. 4. Unit Tests — vi carried 17 __MISSING__ placeholders from #8354 and #8463, and vi is the one locale with a no-placeholder test. Translated. 5. No new ESLint warnings — four suppressed `any`s in tests/unit/combo-routing-engine.test.ts no longer exist, and ESLint exits 2 on stale suppressions. Pruned (271 -> 267); no other entry moved. Also regenerates skills/cli-backup-sync/SKILL.md, which still documented the `backup status` flags #8512 removed — the merge-integrity gate compares the generated output against the tree. Not fixed here: the env/docs contract (NEXT_PUBLIC_OMNIROUTE_BASE_PATH and OMNIROUTE_BACKUP_SCHEDULE_JOB_INTERVAL_MS missing from .env.example), which #8690 already covers, and the quality baselines, which #8686 covers. * fix(quality): keep prettier off the generated SKILL.md files check:agent-skills-sync diffs the generator's output against the tree byte for byte, but lint-staged runs prettier over any staged *.md — and prettier inserts a blank line after the frontmatter that the generator does not emit. Committing a regenerated skill therefore made the gate fail again on the very file that was just brought back in sync. The 44 untouched skills only escape this because they never pass through lint-staged. The generator is the formatter of record for these files, so ignore them. * fix(i18n,quality): drop the stale zh-TW key; raise the auth.ts frozen cap #8463 renamed `oauthModal.googleOAuthWarning` away but left the old key behind in zh-TW, so the "the stale googleOAuthWarning key is GONE from every locale" guard fails on the branch. Removed it. The auth.ts frozen line cap goes 2486 -> 2492. Restoring the dropped exactCooldownIsUpstreamReset call site costs 7 lines, and staging the file makes lint-staged reformat four pre-existing over-100-column lines to prettier's rule — unavoidable without bypassing the hook, which hard rule #10 forbids. The file still sits 12 lines below where the cap was set relative to its actual size. |
||
|
Diego Rodrigues de Sa e Souza
|
4cd1bbc9f5 |
fix(sse): export PROVIDER_BREAKER_FAILURE_STATUSES — fix ReferenceError in chat.ts (base-red slice 4) (#8258)
* fix(sse): export PROVIDER_BREAKER_FAILURE_STATUSES so chat.ts stops throwing ReferenceError Base-red slice 4 (single root cause across the whole handleChat cluster). src/sse/handlers/chat.ts:1273 references PROVIDER_BREAKER_FAILURE_STATUSES to decide whether an all-rate-limited provider result should trip the provider breaker, but the constant was only a FILE-LOCAL const in chatPredicates.ts (a refactor extracted it out of chat.ts and never re-exported it). Every request that reached that branch threw `ReferenceError: PROVIDER_BREAKER_FAILURE_STATUSES is not defined`, so the global fallback and breaker-gate paths blew up — surfacing as "All models failed | PROVIDER_BREAKER_FAILURE_STATUSES is not defined" and breaking the handleChat coverage tests (combo-error passthrough, 503 for cooled-down/open-breaker, budget-error, model cooldown, body-derived retry-after, non-JSON rate-limit bodies). Fix: export the const from chatPredicates.ts and import it in chat.ts (one canonical definition, restoring the pre-refactor behavior). Validated: chat-route-coverage 15/0 (was 12/3), chat-cooldown-aware-retry 6/0, chat-rate-limit-body-lock 2/0; breaker guards (7907, combo-breaker-429, openrouter-6842) unchanged; typecheck:core clean. * test(nvidia): read PROVIDER_BREAKER_FAILURE_STATUSES from chatPredicates.ts Same root cause as the chat.ts import fix in this PR: the const was extracted out of chat.ts into chatPredicates.ts, so the nvidia-quota Phase-1 guard (which greps the source for the `= new Set([...])` declaration to prove 429 was not added to the breaker classification) must read chatPredicates.ts, not chat.ts. Now 13/0. --------- Co-authored-by: Probe Test <probe@example.com> |
||
|
Diego Rodrigues de Sa e Souza
|
7faec9339d |
fix(resilience,translator): three release/v3.8.49 base-red regressions + eslint baseline — conflict resolved (#8254)
Co-authored-by: herjarsa <herjarsa@users.noreply.github.com> |
||
|
MumuTW
|
cbdf1fc835 |
fix(backend): bound the client raw request snapshot instead of deep-cloning the body (#7847) (#8550)
buildClientRawRequest deep-cloned the ENTIRE request body on every chat request, unbounded. On the #7847 incident payload (3.05 MiB, 729 messages, 86 tools) that retains 3.19 MiB per request, and it is pure waste: every consumer of clientRawRequest.body is observability and none of them keeps the full payload. chatCore.ts -> reqLogger.logClientRawRequest no-op when the logger is disabled, otherwise re-clones via cloneBoundedForLog (0.08 MiB) chatCore.ts -> trackPendingRequest clientRequest, surfaced by /api/logs/[id] chat.ts -> recordRejectedRequestUsage requestBody None feeds dispatch, translation or the upstream request, so the snapshot is now taken with cloneBoundedForLog: 3.19 MiB -> 0.08 MiB, a 41x reduction, and retention no longer scales with history length. It stays a clone rather than an alias because body is rewritten downstream (plugin onRequest hook, compression) and the log must show what the client actually sent. Bounding at the entry means the logger re-bounds an already-bounded value, which exposed that cloneBoundedForLog was NOT idempotent -- each container exceeded its own bound once the marker was added, so a second pass truncated again: arrays [marker, ...24 items] is 25 entries > 24, so the marker and one real item were dropped and originalLength was rewritten as 25 instead of the true 729 objects 80 keys + _omniroute_truncated_keys is 81 > 80, so a real key was evicted to make room for the marker and the dropped count was reported as 1 instead of 20 strings the marker was appended AFTER slicing to maxLength, so the bounded string was longer than the bound Without this the persisted log payload would have changed shape versus before the fix. All three now keep the marker inside the budget and treat an already-bounded value as final; verified end to end -- the marker still reports originalLength 729. TDD: tests/unit/repro-7847-bound-client-raw-request.test.ts was written first and failed on three assertions (unbounded retention, retention scaling with history, and the idempotence precondition) before either change. |
||
|
backryun
|
909642879f | feat: add Claude Opus 5 support (#8464) | ||
|
Diego Rodrigues de Sa e Souza
|
c5c27b813a |
fix(sse): cap exact cooldowns only when synthetic — verified upstream resets pass uncapped (#8393)
Contract vs cap: #6863 requires a model lockout to honor a VERIFIED upstream quota reset exactly (e.g. Antigravity "Resets in 92h27m28s", shipped in v3.8.47). #7940 requires SYNTHETIC exact-cooldown estimates (the quota_exhausted until-midnight heuristic) to respect the operator's maxCooldownMs so they cannot balloon unbounded. Both are legitimate, non-conflicting contracts — they apply to different kinds of values. Root cause: #7980 (fixing #7940) changed recordModelLockoutFailure() in open-sse/services/accountFallback.ts to unconditionally clamp every exactCooldownMs against maxCooldownMs, with no way to distinguish a verified upstream reset from a synthetic estimate. A real ~92h reset got clamped to the operator's ~30min cap, and the router went on hammering 429 against quota that was known not to recover for days — regressing #6863's contract by omission, not by new policy (the "honor it exactly" docstrings on selectLockoutCooldownMs() and its call sites were left untouched and now describe dead code). Fix: add an opt-in `exactCooldownVerified` flag to recordModelLockoutFailure()'s options. When true, exactCooldownMs bypasses the maxCooldownMs clamp entirely; when false/omitted (the default), behavior is byte-identical to before this change. Set the flag only at the 4 call sites that already carry upstream provenance for the value they pass — usedUpstreamRetryHint / quotaResetHintMs from checkFallbackError(): - open-sse/services/combo.ts (2 sites): exactCooldownVerified mirrors lockoutHintMs > 0, which is only ever nonzero when it traces back to a genuine upstream signal. - src/sse/services/auth.ts (2 sites): exactCooldownVerified mirrors the same usedUpstreamRetryHint / quotaResetHintMs check already used to derive exactCooldownMs at each site. The quota_exhausted → until-midnight synthetic default and plain exponential backoff are untouched and stay capped, per #7940. The two other recordModelLockoutFailure call sites (combo.ts quality failure, auth.ts local-404/grok-web-403) never carry a verified hint and were left unmodified. Validation (TDD): tests/unit/combo-lockout-quota-reset-6863.test.ts red→green with its assertions unchanged (was clamping ~332,848,000ms to ~1,799,995ms; now honors the parsed reset). Added a boundary pair to tests/unit/model-lockout-exact-cooldown-cap.test.ts proving the same magnitude resolves differently by provenance: synthetic stays capped, verified passes through whole. Full existing suite in that file plus combo-model-lockout-honors-reset-1308.test.ts stay green unmodified. Swept 45 lockout/cooldown-adjacent test files (502/505 passing); the 3 failures reproduce byte-identical on a pristine origin/release/v3.8.49 checkout (PROVIDER_BREAKER_FAILURE_STATUSES ReferenceError in untouched chat.ts, and a documented timing-sensitive serial test) — confirmed pre-existing, out of this fix's scope. npm run typecheck:core and npm run lint are clean. Refs #6863 Refs #7940 Refs #7980 |
||
|
Diego Rodrigues de Sa e Souza
|
3b4f4afc9d |
fix(sse): re-export PROVIDER_BREAKER_FAILURE_STATUSES for the orphaned all-rate-limited breaker path (#8390)
Root cause: #8013 extracted shouldTripProviderBreakerForResult() from src/sse/handlers/chat.ts into the new src/sse/handlers/chatPredicates.ts, taking the (non-exported) const PROVIDER_BREAKER_FAILURE_STATUSES with it. A second, independent use of that const survived in chat.ts's handleSingleModelChat(), in the "all credentials rate-limited" block (~line 1340) — that reference was left orphaned by the extraction. Production impact: any request where every credential for a provider+model is simultaneously rate-limited throws `ReferenceError: PROVIDER_BREAKER_FAILURE_STATUSES is not defined` at runtime in that code path. Concretely this meant: - breaker._onFailure() was unreachable on the all-rate-limited path, so the provider circuit breaker could not trip from it - the ReferenceError propagated up and got mapped to a generic 502, masking the real 503 upstream-unavailable status in combo responses - the issue-agent route surfaced a generic 400 instead of the actual 429 provider-rate-limited response Fix: export PROVIDER_BREAKER_FAILURE_STATUSES from chatPredicates.ts and add it to chat.ts's existing import block from that module. No behavior change — the classification set ([408, 500, 502, 503, 504]) is unchanged, this only repairs the broken reference. Also re-points tests/unit/nvidia-quota-phase1.test.ts's regex-based declaration check at chatPredicates.ts, where the const now actually lives (it previously read chat.ts via fs+regex and silently failed to find the declaration). The regex and the classification assertions themselves are unchanged — this test still proves 429 is excluded from the whole-provider breaker. Refs #8013 |
||
|
Diego Rodrigues de Sa e Souza
|
3504050fcf | fix(providers): route noauth opencode-zen connections through their assigned proxy (#8324) | ||
|
Diego Rodrigues de Sa e Souza
|
35541c06cd | fix(providers): carve cookie-auth providers out of terminal 401 'expired' classification so one 401 cooldowns instead of killing the connection (#8321) | ||
|
Apostol Apostolov
|
f3ed4a49d4 |
feat(providers): add weekly quota tracking for grok-web (#8127)
* feat(providers): add weekly quota fetcher for grok-web (grok.com SSO) Implements a bespoke QuotaFetcher for the grok-web provider that: - Reads OIDC tokens from ~/.grok/auth.json (local Grok CLI login) - Refreshes tokens via auth.x.ai OIDC if expired - Calls https://cli-chat-proxy.grok.com/v1/billing?format=credits - Returns a single 'weekly' window with creditUsagePercent and resetAt - Caches results with 60s TTL (matching codexQuotaFetcher pattern) - Supports GROK_AUTH_PATH env var override for testing - Registers in chat.ts before registerGenericQuotaFetchers Tests cover: missing auth, successful fetch with header verification, 401-triggered token refresh with retry, 60s cache TTL, and preflight integration. Closes #6444 * chore(quality): rebaseline chat.ts for #8127 own-growth --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Markus Hartung
|
cc17b304ab |
fix(sse): Gemini TPM/RPD quota classification + combo cooldown-wait resilience (#8213)
* fix(sse): Gemini TPM classification, combo-cooldown-wait for auto/quota-share, and target-timeout floor
Gemini TPM/RPM 429s were misclassified as QUOTA_EXHAUSTED because
sanitizeErrorMessage() truncates to the first line, hiding Google's
metric name and retry hint on lines 2-3. Added a rawMessage field
(internal-only, never reaches the client) and classifyGeminiQuotaMetricFromText()
to classify from the untruncated text, reordered ahead of the generic
credits/daily-quota checks.
Widened comboCooldownWaitEnabled (wait out a short transient cooldown
instead of crystallizing a 429/503) from quota-share-only to also cover
auto-strategy combos, and raised the wait ceiling to 65s/130s-budget/90s-cap
to match Gemini's ~60s TPM/RPM windows.
The per-target timeout (DEFAULT_COMBO_TARGET_TIMEOUT_MS, 120s) was shorter
than the new 130s cooldown-wait budget, so a target could get cut off
mid-wait with a synthetic 524 instead of completing the retry. Added
resolveComboTargetTimeoutMsForCombo()/isComboCooldownWaitEligible() in
comboConfig.ts to raise the per-target floor to budgetMs+buffer only for
wait-eligible strategies (auto/quota-share), verified live: a 12-request
concurrent burst against a TPM-exhausted combo went from 2/12 succeeding
(10 x 524) to 12/12 succeeding with zero 503/524.
Also: liveGeminiShared.ts's sendAndValidate now fails fast on a 503
instead of retrying past it, and the health dashboard + request logger
surface TPM stats alongside RPM/RPD.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* fix(sse): combo-exhausted rejection logs now capture request body + attempted models
recordRejectedRequestUsage() (the fast path for combo requests that never
reach handleChatCore, e.g. all targets locked by resilience cooldown)
hardcoded provider: "-" and never passed a request body to saveCallLog(),
so /dashboard/logs entries for these failures were nearly useless for
debugging: no way to see the client's request or which models were tried.
- recordRejectedRequestUsage() now accepts requestBody and persists it
through the existing saveCallLog() artifact mechanism (same path
handleChatCore's own logging uses).
- Added summarizeComboAttemptedModels(), which reads the combo's own model
list (always available, unlike the response's combo-diagnostics headers —
a model-level resilience-lockout skip never touches the
exhaustedProviders/exhaustedConnections sets those headers are built
from) to populate a real "provider" value instead of "-".
- Wired both into the call site in src/sse/handlers/chat.ts.
NOTE: unrelated to the Gemini TPM/combo-cooldown-wait fix on this branch —
landed here per operator request, to be split into its own branch/PR.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* feat(sse): synthetic streaming keep-alive event + 5-minute Gemini cooldown-wait ceiling
Many clients enforce a first-SSE-byte timeout, which made it unsafe to wait
out a longer upstream rate-limit cooldown on a streaming request — the
client would abandon the connection before any bytes arrived. This landed
in two parts:
1. Synthetic startup "thinking" event (OpenAI chat/completions format):
the already-existing withEarlyStreamKeepalive wrapper (open-sse/utils/
earlyStreamKeepalive.ts, wired into /v1/chat/completions, /v1/messages,
/v1/responses since #2544) opens the SSE stream immediately once a
request runs past its threshold, but only ever sent empty/no-op
keepalive frames. Added a `startupFrame` option (defaults to
`keepaliveFrame` — zero behavior change unless a route opts in) so the
very first frame can carry real content instead. Wired
OPENAI_STARTUP_THINKING_FRAME (a reasoning_content delta: "OmniRoute:
got request, sending to provider") into /v1/chat/completions only —
Claude Messages and Responses API formats both require a preceding
envelope event (message_start / response.created) that a synthetic
pre-dispatch frame can't safely fabricate without risking a duplicate
envelope once the real stream arrives, so those two routes keep their
existing (safe, proven) keepalive frames unchanged.
2. Raised the "wait out a known cooldown, then retry" ceiling to 5 minutes
for both retry mechanisms, now that a client-side first-byte timeout is
no longer a risk on the (opted-in) route:
- comboCooldownWait (auto/quota-share combos, open-sse/services/combo.ts):
maxWaitMs hard clamp raised 90s -> 300s (src/lib/resilience/settings/
normalize.ts); defaults raised to maxWaitMs:90s/maxAttempts:5/
budgetMs:300s. comboConfig.ts's resolveComboTargetTimeoutMsForCombo
already derives the per-target timeout floor from budgetMs, so it
tracks the new ceiling with no further changes.
- waitForCooldown (direct, non-combo model requests, src/sse/handlers/
chat.ts): this mechanism had NO cumulative cap before — only a
per-wait cap (maxRetryWaitMs) and a retry count (maxRetries), so
maxRetries x maxRetryWaitMs could exceed 5 minutes with no ceiling.
Added a budgetMs field (mirrors comboCooldownWait) to
WaitForCooldownSettings/CooldownAwareRetrySettings, threaded a
requestRetryBudgetLeftMs tracker through chat.ts's requestAttemptLoop
(mirrors combo.ts's comboCooldownBudgetLeftMs), and made
getCooldownAwareRetryDecision refuse to wait once the cumulative
budget is exhausted even if the single wait is under maxRetryWaitMs.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* fix(sse): extend the synthetic keep-alive thinking event to /v1/responses
Live incident (OpenClaw, log id 1784407081908-cbc24f): a /v1/responses
request to gemini/gemma-4-31b-it took 56s to produce a first byte and the
client disconnected (499 request_signal_aborted) — the same client-first-byte-
timeout problem the previous commit fixed for /v1/chat/completions, but
/v1/responses only had the generic bare-comment keepalive (no content), so it
wasn't covered.
Added RESPONSES_STARTUP_THINKING_FRAME: a self-contained synthetic reasoning
item (response.output_item.added -> reasoning_summary_part.added ->
reasoning_summary_text.delta -> reasoning_summary_part.done), opened AND
closed within this one frame rather than left dangling — it never carries a
response_id, so it can't collide with the real upstream response's own
independent response.created lifecycle that follows. Mirrors the abbreviated
delta+part.done close pattern open-sse/utils/stream.ts's own
emitSyntheticResponsesReasoningSummary already uses for real mid-stream
reasoning content.
Wired into src/app/api/v1/responses/route.ts via the startupFrame option
added in the previous commit.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* fix(sse): combo cooldown-wait vars reset every setTry, crystallizing a bogus 503 instead of waiting
Live incident (log id 1784416706646-51): a request to the "default" combo
(strategy=auto, maxSetRetries=3) hit a real Gemini TPM 429 on both gemma-4
targets, correctly classified as a short 40s rate_limit lockout — then
crystallized a 503 "all upstream accounts are inactive" in 6.9s instead of
ever reaching the cooldown-aware wait.
Root cause: `lastError`/`earliestRetryAfter`/`lastStatus` were declared with
`let` INSIDE the `for (setTry...)` loop body, so they reset to null at the
start of every set-try. When both targets lock out on setTry 0, every
subsequent setTry (1..maxSetRetries) pre-skips both targets via the
isModelLocked check with no real dispatch — so on the FINAL setTry (the only
one whose values the post-loop decision reads, since it's gated behind
`if (setTry < maxSetRetries) continue`), lastStatus was null, hitting the
"!lastStatus" branch (ALL_ACCOUNTS_INACTIVE 503) and completely bypassing the
comboCooldownWaitEnabled / earliestRetryAfter wait logic — even though a
real 429 with a known ~40s retry-after WAS observed on setTry 0.
This bug predates today's Gemini TPM work (any combo with maxSetRetries > 0
whose targets all lock out on the first pass was affected) but was masked in
existing tests: the "auto strategy (2 models...)" regression test uses
maxSetRetries: 0, so it only ever runs ONE setTry iteration and never
exercises the reset-on-retry path. It also explains why the dedicated
12-concurrent-request burst test passed cleanly — with concurrent requests,
timing variance meant some request's FINAL setTry iteration still had a live
target to dispatch to, giving lastStatus/earliestRetryAfter fresh data. A
single isolated request has no such luck.
Fix: hoist lastError/earliestRetryAfter/lastStatus to just inside
dispatchWithCooldownRetry, before the setTry loop, so they persist across
set-tries (still reset fresh on each recursive dispatchWithCooldownRetry()
call after a wait, which is correct). recordedAttempts/fallbackCount/
exhaustedProviders etc. are intentionally left per-iteration (unrelated to
this bug).
New regression test in tests/unit/combo-quota-share-cooldown-wait.test.ts
reproduces the exact live scenario (2 targets, both lock out on setTry 0,
maxSetRetries: 3) — confirmed red (503) against the pre-fix code, green
(200, waits and retries) against the fix.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* test(sse): extend live Gemini workload to Responses API + add large-context TPM test
Two additions to the live Gemini test suite, both live-verified against the
dev instance:
1. sendAndValidate() (tests/integration/liveGeminiShared.ts) now accepts an
apiFormat: "chat" | "responses" parameter, building the Responses-API
request shape (input array, max_output_tokens) and parsing its SSE events
(response.output_text.delta / response.reasoning_summary_text.delta /
response.completed) via the new readResponsesSSEStream(). Wired into two
new tests in live-gemini-workload.test.ts ([30]/[31]), mirroring the
existing Chat Completions streaming coverage. Verified live: 24/25 + 5/5
payloads succeeded end-to-end through the new code path (the one failure
was a ~300s test-client fetch timeout unrelated to the Responses API code
itself — a separate, not-yet-addressed test-harness limitation).
2. genHugeContextMessage() builds a single message large enough (~4
chars/token estimate) to approach or exceed Gemini's free-tier TPM ceiling
(16000 input tokens/min for gemma-4) by itself. Every other prompt
generator in this file tops out around 1-2k tokens — nowhere near that
ceiling — so none of the existing workload tests ever exercised a REAL TPM
429, only RPM-style rate limiting. tests/integration/gemini-large-context-tpm.test.ts
sends two ~12-13k-token requests back-to-back (comfortably exceeding
16000/min together) to exercise the full path against production Gemini:
TPM classification, the comboCooldownWait retry, and the synthetic
keep-alive frame on a genuinely slow request.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* fix(sse): abandoned combo target dispatch now observes its own per-target timeout, fixing a permanent "pending" dashboard leak
Live incident (dashboard log id 1784418258231-14961a, reported as "an ongoing
request even though there's already a 200"): a combo target dispatch
abandoned by comboTargetTimeoutMs (open-sse/services/combo/targetTimeoutRunner.ts)
left a permanent phantom "pending" entry in the dashboard, even after the
overall combo request had already succeeded via a different retry.
Root cause: chatCore.ts's createStreamController — and everything downstream
that depends on it (withRateLimit's Promise.race against Bottleneck,
acquireAccountSemaphore) — only ever watches clientRawRequest.signal, which
is the ORIGINAL client's request signal (set once via buildClientRawRequest
and reused unchanged across every target dispatch in a combo). It has no
connection to targetTimeoutRunner.ts's OWN AbortController
(target.modelAbortSignal), which is what actually fires when
comboTargetTimeoutMs (300s) elapses. src/sse/handlers/chat.ts's
handleSingleModel bridge between combo.ts and handleSingleModelChat received
`target.modelAbortSignal` but silently dropped it — never forwarded it
anywhere. So when a target got abandoned (e.g. stuck inside a wedged
Bottleneck rate-limiter queue, see the WEDGED force-reset log line from the
same incident), its per-target timeout fired and let the COMBO move on and
retry successfully elsewhere — but the abandoned dispatch's own promise
chain never learned it had been superseded, so it hung forever waiting on a
signal that was never going to fire, and trackPendingRequest(false) (the
finalize call) never ran.
Fix: thread target.modelAbortSignal through as a new modelAbortSignal
runtimeOption, and merge it into clientRawRequest.signal (via the existing
mergeAbortSignals helper from open-sse/executors/base.ts) right before
dispatch, so an abandoned target's own promise chain now observes its abort
and can reach its cleanup path — new resolveDispatchClientRawRequest() makes
this mechanically testable in isolation.
Co-authored-by: Markus Hartung <markus.hartung@gmail.com>
* fix(sse): combo cooldown-wait state recording, rate-limit wedge recovery, OpenAI-format SSE error frames
Five related fixes surfaced by live incidents (dashboard log ids 1784457764961-73,
1784465227489-a2cbc0, 1784504040241-6f8b9a) while validating the Gemini TPM/cooldown-wait
work on this branch against real OpenClaw traffic:
- combo.ts: the model-lockout bail-out branches in dispatchWithCooldownRetry never
recorded lastStatus, so once every target in a set hit an existing lockout the final
check crystallized a bogus ALL_ACCOUNTS_INACTIVE 503 instead of reaching the
cooldown-wait decision, even with a real 429 + short retry-after observed.
- combo.ts/combo/types.ts: the "all credentials cooling down" pre-dispatch rejection
(buildModelCooldownBody) nests its retry hint as error.retry_after/reset_seconds, not
the top-level retryAfter every other 429 shape uses — combo's extraction only read the
latter, so earliestRetryAfter stayed null for this shape even after lastStatus was fixed.
- rateLimitManager.ts: the wedge-recovery watchdog used disconnect(), which releases the
heartbeat timer but never rejects jobs already QUEUED on that instance — orphaned
dispatches hung until the outer ~300s per-target timeout, well past real clients'
patience. Switched to stop({ dropWaitingJobs: true }), safe because the wedge condition
already requires RUNNING===0 && EXECUTING===0.
- earlyStreamKeepalive.ts: the in-band error frame emitted after committing to a 200 SSE
stream was hardcoded to Anthropic's `event: error` convention for every route, including
the OpenAI-format ones (/v1/chat/completions, /v1/responses) where that framing is
either invisible or malformed to a plain data-line parser. Added per-route
OPENAI_CHAT_ERROR_FRAME / OPENAI_RESPONSES_ERROR_FRAME and wired them in.
- chatCore.ts: persisted a synthetic clientResponse error body even when the client had
already disconnected (AbortError) before that body was ever computed — misleading the
dashboard into showing "what the client received" for a response that was never sent.
Also: RequestLoggerDetail.tsx — Provider/Client Event Stream panes lost their collapse
toggle when StreamSection replaced the collapsible PayloadSection (
|
||
|
backryun
|
888c872459 |
refactor(antigravity): align official clients and callable catalog (#8013)
* fix(antigravity): preserve protocol fidelity and fail closed * chore: add PR-numbered changelog fragment * test: split oversized Antigravity suites * refactor(antigravity): align official IDE and CLI identities * fix(antigravity): align catalog with callable models * test(antigravity): update 2 test files to renamed version-cache API (#8013 fix) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: nguyenha935 <208228297+nguyenha935@users.noreply.github.com> Co-authored-by: backryun <backryun@users.noreply.github.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> Co-authored-by: nguyenha935 <nguyenha935@users.noreply.github.com> Co-authored-by: Probe Test <probe@example.com> |
||
|
Innokentiy Solntsev
|
f909b1d45e |
fix(resilience): don't cool down accounts or trip the breaker on client aborts (#7908)
* fix(resilience): don't cool down accounts or trip the breaker on client aborts When the caller drops the connection mid-stream, the in-flight request surfaces request_signal_aborted, "Client disconnected", or a DOM AbortError with no upstream status code. These shapes were counted as provider failures: the serving connection went into cooldown, the provider circuit breaker accrued failures, and healthy accounts ended up marked unavailable from client-side cancellations alone. Treat client aborts as local stream lifecycle events (#4602 policy): extend isLocalStreamLifecycleError() to recognize abort shapes and skip connection disable and breaker accounting for them. Genuine upstream failures (5xx/429/401) are still counted. Fixes #7907. * fix(resilience): guard the two remaining breaker-trip call sites against client aborts (#7907) PR #7908 correctly wired isLocalStreamLifecycleError() into shouldSkipConnDisable() and chatHelpers.ts's onStreamFailure, but two separate breaker._onFailure()-triggering call sites were purely status-code gated and never checked it, so a client-side abort (no upstream status, defaults to 502, error='request_signal_aborted') still tripped the whole-provider circuit breaker — the highest blast-radius of the 3 resilience mechanisms: - src/sse/handlers/chat.ts: the single-model, non-combo terminal-failure path called breaker._onFailure() directly, bypassing the isFailure option (which only applies inside breaker.execute()). Extracted the predicate into shouldTripProviderBreakerForResult() and added the missing isLocalStreamLifecycleError guard. - open-sse/services/combo/comboPredicates.ts::shouldRecordProviderBreakerFailure(), used by handleComboChat's executeTarget (open-sse/services/combo.ts), gained the same guard via a new optional `error` field. Added tests/unit/circuit-breaker-abort-provider-trip-7907.test.ts exercising both real predicates directly (not just the isolated isLocalStreamLifecycleError() helper) — confirmed red on the unfixed code (missing export) and green after the fix, alongside the existing #4602/#7908/combo-breaker-429 suites (24/24 pass, no regressions). file-size-baseline.json: +1 combo.ts (irreducible call-site wiring for the new `error` field) and +1 chatHelpers.ts (own growth from the PR's already-verified onStreamFailure guard, surfaced only now since fast-gates PR->release skip check:file-size). Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouzapw@users.noreply.github.com> Co-authored-by: insoln <insoln@ya.ru> * test(quality): register circuit-breaker abort tests in stryker tap.testFiles The two new tests (circuit-breaker-abort-provider-trip-7907, circuit-breaker-client-abort) import mutated modules (circuitBreaker.ts, comboPredicates.ts) but were not listed in stryker.conf.json tap.testFiles, tripping check:mutation-test-coverage --strict. Co-authored-by: diegosouzapw <diegosouza.pw@gmail.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouzapw@users.noreply.github.com> |
||
|
Paijo
|
3238df3204 |
perf: lazy provider init, P2C quota cache, structuredClone elimination, getSettings→getCachedSettings (batch 2) (#7893)
* perf: startup parallelization, stream TextEncoder lift, auth middleware bottlenecks
Startup (~100-300ms faster cold start):
- Parallelize 4 early imports via Promise.all() in registerNodejs()
- Parallelize 10 independent background services via Promise.allSettled()
- Each service has independent try/catch — no failure domino effect
Streaming pipeline (8 fewer TextEncoder GC allocations per SSE event):
- Lift new TextEncoder() from per-chunk inside buildClaudeStreamingResponse
to function scope alongside existing decoder singleton
Auth middleware bottlenecks (from PerfBottleneckAnalysis):
- Backoff decay loop: replace updateProviderConnection (full CRUD:
SELECT+encrypt+cache-invalidate+backup) with resetConnectionBackoff
(targeted UPDATE of backoff/error columns only)
- Dual .filter() for quota: replace two passes calling
isQuotaExhaustedForRequest per connection with a single for loop
partitioning into withQuota/exhaustedQuota
- Debug-log filter recomputation: capture connectionFilterStatus Map
during the filter pass; debug loop reads 6 string comparisons instead
of 6 function calls per connection
Supporting:
- Add resetConnectionBackoff to src/lib/db/providers.ts (patterned after
clearConnectionErrorIfUnchanged, no CAS check)
- Re-export resetConnectionBackoff from src/lib/localDb.ts
- Update integration-wiring.test.ts regex for parallelized dynamic import
* perf: P2C quota cache, lazy provider init, structuredClone elimination, getSettings→getCachedSettings
- **auth.ts: P2C quota re-evaluation cache** — quotaResults Map threaded
through selectPoolSubset → compareP2CConnections → getP2CConnectionScore.
Populated during filter + partition passes, eliminating redundant
evaluateQuotaLimitPolicy / isQuotaExhaustedForRequest calls when the
P2C comparator re-evaluates previously-scored connections.
- **constants.ts: lazy PROVIDERS via Proxy** — replaces eager
generateLegacyProviders() + loadProviderCredentials() at module load
with Proxy delegating to deferred init on first property access.
- **providerModels.ts: lazy PROVIDER_MODELS + PROVIDER_ID_TO_ALIAS** —
same Proxy pattern for both exports; generateModels()/generateAliasMap()
deferred until first read.
- **stream.ts: structuredClone → minimal object spread** — replaces
O(n) deep clone of SSE response chunks with targeted reconstruction
of only mutated fields (usage, delta.content, finish_reason).
- **progressTracker.ts: TextDecoder lift** — module-level decoder
instead of per-chunk new TextDecoder().
- **Route files: getSettings() → getCachedSettings()** — 13 API route
files converted from uncached per-request DB reads to TTL-cached
wrapper (5s default), eliminating redundant queries on every request.
- **settings.ts: re-export getCachedSettings** from readCache for
non-localDb consumers.
- **Remove settingsCache.ts** — dead file, no imports reference it.
TS compile: 0 errors. Auth tests: 225/225 pass. Services: 269/269 pass.
* perf: Phase 1 tangible wins — egressCache eviction, mmap_size PRAGMA, composite indexes, proxyFallback lazy import
- egressCache: lazy TTL eviction on getCachedEgressIp access (bounds memory
leak to distinct proxy URLs, typically <100)
- mmap_size: apply stored PRAGMA from key_value table (256MiB default) after
applyStoredDatabaseOptimizationSettings — setting was stored but never applied
- schemaColumns: add idx_uh_provider_model_timestamp (covers getModelLatencyStats)
and idx_pc_provider_auth_type (covers 6+ provider_connections queries)
- proxyFallback: convert static import to dynamic import() inside error handler
(defers 210ms module load from startup to first proxy-retry scenario)
* perf: add dedup expression index, unref() sweep timers
- Add COALESCE expression index idx_uh_dedup on usage_history
matching the exact dedup query pattern. Eliminates FULL TABLE
SCAN on every saveRequestUsage insert.
- Add composite idx_uh_provider_model_timestamp on usage_history.
- Add composite idx_pc_provider_auth_type on provider_connections.
- Add .unref() to setInterval in batchProcessor.ts (polling loop).
- Add .unref() to setInterval in runtimeHeartbeat.ts (heartbeat).
* perf: bump SQLite cache_size default from 16MB to 64MB
New installs now start with 64MB page cache (was 16MB). Existing
users' stored settings are unchanged. Reduces disk reads for the
typical ~250MB database by keeping ~25% of pages in memory.
Also resolved pre-existing merge conflict in webhooks.ts.
* docs: add Redis production config guide and proxy port clash investigation report
- docs/redis-production-config.md: comprehensive Redis tuning guide
covering client options, server config, Docker settings, scaling,
and monitoring for all three Redis workloads (rate limiting,
auth cache, quota store)
- docs/proxy-port-clash-report.md: investigation confirming proxy
subsystem has no port binding issues; real EADDRINUSE history
traced to process supervisor crash-loop restart race (#4425) and
live-dashboard port clash (#6324), both already fixed
* fix: address PR #7893 review — add Proxy traps, extract migrations to reduce providers.ts size
- Add set trap to PROVIDER_ID_TO_ALIAS Proxy (providerModels.ts)
- Add deleteProperty traps to all three lazy Proxies (PROVIDERS,
PROVIDER_MODELS, PROVIDER_ID_TO_ALIAS)
- Extract autoMigrateLegacyEncryptedConnections and getGheCopilotHosts
from providers.ts (1129→1036 lines, -93) into providers/migrations.ts
- Both functions re-exported via providers.ts for backward compat
File-size ratchet resolved: src/lib/db/providers.ts now 1036 lines.
* fix: resolve merge conflict markers in 3 route/test files
- model-combo-mappings/route.ts: kept upstream version (Zod pagination
via validateBody + isValidationFailure), restored missing return
statement for GET handler
- playground/presets/route.ts: kept stashed version details (satisfies
type-narrowing + inlined Response) — functionally identical
- error-sanitization.test.ts: matches upstream exactly (no diff)
Test verification: same 7 pre-existing failures confirmed on upstream
baseline (
|
||
|
backryun
|
65e0aeda79 |
[Part 1/3]refactor(qwen): replace legacy Qwen Code and remove OAuth provider (#7866)
* refactor(cli): remove legacy Qwen Code integration * refactor(qwen): remove deprecated Qwen OAuth provider * feat(cli): rebuild Qwen Code integration for upstream V4 * fix(qwen): clear stale CLI auth on reset * test(qwen): align retired provider coverage * fix(db): renumber qwen-cleanup migration 129 -> 130 release/v3.8.49 tip took slot 129 via #7843 (usage_history_codex_strong_identity, itself renumbered from 128 during the #7838/#7840 base-red cleanup) after this branch forked; renumber remove_unregistered_qwen_data to 130. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Diego Rodrigues de Sa e Souza
|
51b118c2d3 |
feat(routing): read-only auto/* candidate transparency + per-API-key exclusions (#7819) (#7839)
Level 1: GET /v1/auto-combo/{channel}/candidates lists an auto/* channel's
candidate pool with live reachability (provider circuit breaker via
getStatus()/canExecute(), connection cooldown, model lockout).
Level 2: per-API-key candidate exclusions, persisted in a new
auto_candidate_overrides table and enforced at the virtualFactory.ts
candidate-pool chokepoint via a pure, fail-open filter — mirrors the #7622/
#7646 precedent exactly (zero touches to the frozen combo.ts god-file).
Levels 3 (weights/ordering) and 4 (policy pin) are deferred to a follow-up
issue, as is the dashboard UI (Step 4) and its i18n strings.
|
||
|
Erick Kinnee
|
eba6ecaf2b |
fix(compression): apply compression combo assignments to routing combos (#7779)
* fix(api): enumerate tiered auto combo endpoints in /api/combos/auto The backend already supports auto/<category>[:<tier>] routing via suffixComposition.ts + virtualFactory.ts, but GET /api/combos/auto only exposed 6 flat variants. This adds a second loop enumerating the 10 curated AUTO_SUFFIX_VARIANTS (auto/coding:free, auto/coding:cheap, auto/coding:pro, auto/reasoning, auto/vision, etc.). Fixes #7619 * fix(combos): enumerate template and family auto variants in GET /api/combos/auto The endpoint was missing 27 auto variants that /v1/models already advertises, causing 404s when clients tried to use them: - 20 template variants (auto/best-coding, auto/pro-*, auto/claude-*, auto/best-free, etc.) - 7 family variants (auto/glm, auto/minimax, auto/mimo, auto/zai, auto/gemma, auto/llama, auto/gemini) Fixes #7619 Refs #6453 * fix(combos): swap Phase B/C ordering to match catalog.ts Template variants (Phase C) now enumerate before suffix variants (Phase B) so that overlapping ids like auto/reasoning and auto/vision use template resolution (variant-based) rather than suffix resolution (category-based), matching the behavior in catalog.ts. * fix(combos): fix comment labels and redundant as const * fix(compression): apply compression combo assignments to routing combos Routing combos (e.g. codex, free-only, or-free) use provider-prefixed model strings like codex/gpt-5.5 and go through handleSingleModelChat, which passes comboName: null, isCombo: false. The compression combo assignment lookup in chatCore.ts was gated behind if (isCombo && comboName), so routing combos never had their compression combos applied. Fix: - Add routingComboId parameter threaded through handleSingleModelChat → executeChatWithBreaker → handleChatCore - In handleChat(), resolve the routing combo UUID from the model string's provider prefix via getComboByName - In chatCore.ts, change the gate to (isCombo && comboName) || routingComboId and add routingComboId to the lookup key array Fixes #7771 * fix(autoCombo): guarantee positive maxOutputTokens fallback in computeAdvertisedLimits GET /api/combos/auto now enumerates auto/<family> variants (auto/llama, auto/glm, etc). computeAdvertisedLimits() already guaranteed a positive contextLength for any non-empty candidate pool via getTokenLimit()'s fallback chain, but had no equivalent fallback for maxOutputTokens — candidates whose registry entry and models.dev sync data both lack that field (common for no-auth/free-tier providers matching a family filter, e.g. llama-* on groq/bazaarlink/etc) left maxOutputTokens null, which tests/unit/auto-combo-context-advertising.test.ts catches as a contract violation of the endpoint (opencode disables smart auto-compaction when a limit is falsy — the same bug class this module's docstring already describes for contextLength). Fall back to a conservative generic default (4096) when no candidate in the pool resolves a known maxOutputTokens, mirroring the existing contextLength guarantee. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * fix(autoCombo): align advertised max_output_tokens fallback with the catalog convention (8192) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * chore(quality): file-size baseline for chatHelpers routingComboId thread (876->877) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: Erick Kinnee <erick@ekinnee.dev> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Paijo
|
b61be8d2d0 |
feat(perf): IC2 — cache provider connections by ID + lazy-decrypt credentials (#7787)
* chore(ci): add .mergify.yml to main — Mergify only reads config from the default branch (#7168) * fix(ci): add the auto-enqueue pull_request_rule to the Mergify config (queue_conditions alone are eligibility-only) (#7179) * fix(ci): migrate Mergify auto-enqueue to merge_protections_settings.auto_merge_conditions (rules-based path is EOL 2026-07-16) (#7216) * fix(ci): drop Mergify batch settings (batching is a paid-tier feature; free plan queue is serial) (#7220) * fix(ci): merge queue tolerates the advisory dast-smoke failure (its GH-hosted build hang dequeued every attempt) (#7225) * test(ci): make the #6634 selfref guard hermetic — main's copy hard-fails every PR (#7341) main's copy of this test still does git I/O inside a unit test: const baseSrc = git(['show', 'origin/main:' + FILE]); Runners check out a shallow single ref, so origin/main does not resolve and the test dies with 'fatal: invalid object name origin/main'. Every PR into main fails Unit Tests (7/8) on it — today that is #7313, #7315, #7316, #7334, #7336 and #7337, six PRs red on a defect none of them introduced. #7313 has no other red at all. release/v3.8.49 already carries a fix ( |
||
|
Diego Rodrigues de Sa e Souza
|
7b85e1f6f7 | feat(api): sync upstream reasoning.supported_efforts into synced-model catalog (#7694) (#7767) | ||
|
Diego Rodrigues de Sa e Souza
|
d03fc19c58 |
fix(sse): wire settings.wildcardAliases into model resolution (#7693) (#7748)
Wildcard model aliases created via the Settings UI's "Wildcard Pattern" mode were persisted to settings.wildcardAliases but getCombinedModelAliases() never read that store, so the wildcard-matching step in getModelInfoCore() never saw the user's patterns. Every request fell through to provider inference and threw "Ambiguous model" for models multiple providers claim. Fold settings.wildcardAliases entries into the merged alias map (keyed by pattern string, folded in last so it never shadows exact aliases). |
||
|
Jan Leon
|
74c006e245 |
Add reasoning-based model and effort routing (#7607)
* feat(routing): add reasoning-based model and effort routing * refactor(routing): modularize reasoning and auto-routing pipeline * fix(routing): remove redundant DB re-export and prevent SQL scan false positives * fix(routing): resolve reasoning routing review blockers * fix(i18n): keep release ranking fallbacks outside reasoning * fix(db): renumber reasoning-routing migration past release tip (124→125) 124_generic_session_affinity_ttl.sql (#7274) has since landed on release/v3.8.49 at version 124, colliding with this PR's own 124_reasoning_routing_rules.sql. Renumbers to 125 (the next free slot past the current release tip) and updates the one filename reference in docs/routing/REASONING_ROUTING.md. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * chore(db): renumber reasoning-routing migration 125→126 (slot taken by #7360) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * refactor(api): compact temp-path decls in exportAll GET (complexity-ratchet lines budget) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * refactor(api): single-statement auth guard in exportAll GET (function under 80-line cap) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Dongwook
|
9fce7d0fbf |
fix(antigravity): allow cloudcode envelope through messages guard (#7582)
* fix(antigravity): allow cloudcode envelope through guard * fix(sse): dedupe antigravity source-format detection, shrink chat.ts under cap resolveChatSourceFormatForPath() in chat.ts duplicated the exact antigravity-path regex already in detectFormatFromEndpoint() (open-sse/services/provider.ts) — the added function pushed chat.ts to 1808 lines, over the frozen file-size cap of 1797, with no baseline bump. Remove the duplicate: add a thin detectFormatFromUrl(body, requestUrl) wrapper next to detectFormatFromEndpoint (single source of truth for the path/body-based format detection), and have chat.ts call it directly. Also drop the now-single-use FORMATS import (compare against the literal "antigravity", matching the existing convention in chatHelpers.ts) and remove an unneeded block-scope around the pre-existing #6402 messages guard (renamed its local to msgBody — a second, separate `const b` block further down for temperature/top_p/max_tokens/n validation is untouched and does not collide). Net effect: chat.ts 1808 -> 1797 lines (exactly at the frozen cap, no baseline change). Behavior is unchanged — same tests, same guard logic, same antigravity bypass. Re-verified full green: typecheck:core, eslint, file-size/complexity/cognitive-complexity/complexity-ratchets/changelog- integrity/test-discovery gates, and the PR's own regression suites (chat-messages-validation-6402.test.ts 26/26, mitm-server-antigravity- route-alias.test.ts 4/4), plus the adjacent format-detection and chat-pipeline test suites. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Isiah Wheeler
|
987b6448f7 |
feat(resilience): guard OmniRoute peer routing loops (#7555)
* feat(resilience): guard OmniRoute peer routing loops * refactor(resilience): fold peer-loop log+response into rejectPeerRequest helper (file-size budget on chat.ts) Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: Isiah Wheeler <2122839+isiahw1@users.noreply.github.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
backryun
|
9544fb6353 |
feat(kimi): sync Code, Web, and Moonshot providers (#7531)
* feat(kimi): sync Code, Web, and Moonshot providers * chore(quality): trim frozen file-size overflow in Kimi sync The Kimi/Moonshot provider sync added a net +1 line to both src/sse/services/auth.ts and ProviderDetailPageClient.tsx, pushing each 1 line past its frozen cap in file-size-baseline.json. Drop one optional blank line in each (prettier-neutral, no behavior change) to land back at/under the frozen baseline. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Jan Leon
|
53a6fea6fd |
Expose proxy controls for no-auth providers (#7419)
* fix(theoldllm): honor provider proxy for Vercel blocks * fix(theoldllm): fail closed when assigned proxy is unavailable * refactor(theoldllm): extract proxy guards into dedicated module * fix(ui): expose provider proxy for no-auth providers * fix(proxy): preserve provider-specific no-auth proxy routing * fix(ui): hide unsupported proxy control for VeoAI Free * fix(proxy): handle no-auth provider proxy assignments safely * fix(dns): expose agent-specific host resolution * fix(ui): gate no-auth proxy controls by provider capability * chore(ci): restart GitHub checks * fix(proxy): pass provider to no-auth count tokens resolution * chore(quality): rebaseline ProviderDetailPageClient 786->798 (3-PR irreducible wiring, campaign 2026-07-18) Three authorized PRs each add irreducible call-site wiring to ProviderDetailPageClient.tsx: #7360 +5 (ProviderQuotaVisibilityToggle render), #7419 +4 (NoAuthProviderControls wiring), #7062 +3 (Dahl provider hook) = 786->798. All three follow the extracted-component pattern; the frozen file only takes the wiring. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
NOXX - Commiter
|
46eac9813f |
fix(guardrails/chat): stop Vision Bridge hijacking credentialed models to opencode-zen (#7204)
* chore(ci): add .mergify.yml to main — Mergify only reads config from the default branch (#7168) * fix(ci): add the auto-enqueue pull_request_rule to the Mergify config (queue_conditions alone are eligibility-only) (#7179) * fix(guardrails/chat): stop Vision Bridge hijacking credentialed models to opencode-zen OpenCode (and similar clients) often send image parts in long sessions. Vision Bridge treated the request model as non-vision and whole-request- rerouted to getBestVisionModel(), which preferred opencode-* (priority 0). That landed on a noauth connection and returned 401 Missing API key — while proxies/combos still logged the original target (zai/glm-5.2, grok-cli, …). Also: after resolveRoutingModel(X-Route-Model), keep body.model aligned so the post-guardrail "body.model !== modelStr" path cannot undo the routing header. - visionBridge: skip whole-request reroute when original model has usable creds - visionBridge: refuse reroute to targets known unusable (noauth without key) - visionBridgeRouter: deprioritize opencode-* for auto vision pick - chat: alignBodyModelWithRouting + only adopt true guardrail model mutations - tests: VB-CRED-01/02 + alignBodyModelWithRouting coverage * fix(guardrails/chat): keep chat.ts under the file-size ratchet and update stale vision-bridge tests for the credential-aware reroute skip - Extract the routing-model reconciliation logic (X-Route-Model align, post-guardrail reroute policy re-check, hook model override) into RoutingModelOps helpers in resolveRoutingModel.ts, shrinking chat.ts back under the frozen 1796-line file-size baseline (was 1837). - Update tests/unit/guardrails/vision-bridge-callmodel.test.ts: the fallback mock must match whichever API shape the selected fallback model actually calls (OpenAI-compatible vs Anthropic), since the vision-bridge router priority fix in this PR can now legitimately select an Anthropic fallback model instead of always defaulting to an OpenAI-shaped opencode-* model. - Update tests/unit/vision-bridge-policy-reroute-6640.test.ts: per this PR's own VB-CRED-01 test, a credentialed original model is now intentionally never whole-request-rerouted (it always falls through to describe-then- forward) — so the pre-existing #6640 tests are updated to assert the final, user-facing answer always comes from the original credentialed model, matching the new intended behavior instead of the retired whole-request-reroute path. Co-authored-by: diegosouzapw <diegosouza.pw@gmail.com> * fix(guardrails/chat): reduce isProviderConnectionUsable cyclomatic complexity to satisfy the project-wide complexity ratchet The new isProviderConnectionUsable helper (complexity 21) regressed the project-wide complexity ratchet from 2056 to 2057. Refactor it to use Set membership checks and small extracted helpers (hasNonEmptyString, hasOAuthCredential) instead of chained === / || comparisons — same behavior, verified by the existing "isProviderConnectionUsable rejects noauth without api key" test, with complexity back under the 15-per-function threshold and the project-wide ratchet back at the 2056 baseline. Co-authored-by: diegosouzapw <diegosouza.pw@gmail.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <8016841+diegosouzapw@users.noreply.github.com> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> |
||
|
Jan Leon
|
c1a3d83c27 |
fix(combo): reject known context overflow without exhausting providers (#7177)
* Fix context-window exhaustion classification * fix(combo): keep chat.ts/comboStructure.ts under the file-size ratchet + fix context-overflow boundary bug - Extract getKnownContextOverflow (+ its KnownContextOverflow type) out of comboStructure.ts into a new open-sse/services/combo/knownContextOverflow.ts leaf, so the file-size ratchet (cap 800 for new files) passes. - Extract the skipConnectionDisable predicate out of handleSingleModelChat in chat.ts into open-sse/services/combo/comboPredicates.ts::shouldSkipConnDisable, and consolidate the new combo-failure-handling imports, to keep chat.ts under its frozen file-size baseline (1796) after the #7177 request-scoped-failure wiring. - Fix a real boundary bug in getKnownContextOverflow surfaced by the merge: estimateRequestInputTokens counted a caller-omitted `messages: []` (which some combo entrypoints default in) as real content, charging a few phantom "structural" JSON.stringify tokens toward the estimate. That was enough to falsely trip the new known-context-overflow rejection for a request with no real input when max_tokens exactly equals the target's context window (a common config where limit_input === limit_output === limit_context), regressing tests/unit/combo-routing-engine.test.ts's pre-existing #3587 "non-reasoning model does not get max_tokens buffer" case. Empty arrays/objects no longer count as estimable content. - Add a regression test for the exact-boundary empty-content case. Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> * refactor(combo): move overflow logic into knownContextOverflow module (file-size cap) comboStructure.ts is not frozen in the file-size baseline but is capped at 800 lines; this PR's net +29 on that file alone would push it over once merged. knownContextOverflow.ts already exists in this PR as the dedicated home for "known context limit" logic, so move the genuinely new pieces there instead of leaving them in comboStructure.ts: - hasEstimableContent (new): its own doc comment already frames it purely in terms of the known-context-overflow boundary check, so it belongs next to that check, not in the general request-compatibility file. - getKnownContextLimit (new, requestedOutputTokens-aware): this *is* the "how big is a target's known context window" primitive knownContextOverflow already consumes; hosting it there is a better fit than comboStructure.ts. - getLegacyKnownContextLimit: kept alongside its sibling rather than split across two files, since both are alternate implementations of the same concept (used only by comboStructure.ts's hasKnownCompatibleContextLimit). comboStructure.ts now imports all three back for its own internal callers (estimateRequestInputTokens, getTargetCompatibilityFailures, hasKnownCompatibleContextLimit). deriveRequestCompatibilityRequirements and the RequestCompatibilityRequirements type stay in comboStructure.ts exactly as this PR already has them (still consumed internally there), so knownContextOverflow.ts keeps importing those two, same as before. No behavior change — pure relocation, verified by the existing PR test suite (combo-context-window-filter, combo-breaker-429, combo-failure-log-message, combo-target-exhaustion, diagnostics) plus the pre-existing combo-vision-aware-routing/combo-context-requirements/combo-roundrobin-compat-fallback-6238 suites, all green. Net effect on open-sse/services/combo/comboStructure.ts vs. this PR's merge base: -2 lines (was +29). typecheck:core, lint, and the complexity ratchets (check:complexity, check:cognitive-complexity) are unchanged from this PR's current HEAD — the 4 pre-existing complexity/max-lines findings in valueContainsImagePart/filterTargetsByRequestCompatibility are untouched by this move (same violations, same total ratchet counts, just shifted line numbers). Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
Diego Rodrigues de Sa e Souza
|
d9f3699b4e |
feat(sse): quota tracking for AgentRouter, v0 (Vercel), FreeModel (#6850, #6845, #7075) (#7653)
Validated in merge-train --fast @ 6cafcbb (static gates + 9 changed test files + vitest green, 2m35s; full suite ran today on train 2c tip) |
||
|
Diego Rodrigues de Sa e Souza
|
9e084e18a7 |
feat: OpenRouter quota tracking (key/credits + free-window counter) (#6842) (#7651)
Validated in merge-train --fast @ 4ed4498 (static gates + changed tests 29/29 + vitest green, 2m39s; full suite ran today on trains 1/2c) |
||
|
Diego Rodrigues de Sa e Souza
|
735e2d0783 |
feat(sse): generalize session affinity TTL to all providers (#7274) (#7650)
Validated in local merge-train @ 8f27177d1 (full parity suite green: typecheck+file-size+complexity+cognitive+changelog+unit shards 1&2+vitest) |
||
|
Diego Rodrigues de Sa e Souza
|
21d5acbb40 |
feat(api): accept x-goog-api-key header for client-facing auth (#7034) (#7236)
gemini-cli (and any @google/genai-based client) sends its credential exclusively via x-goog-api-key and it is not client-configurable to use Authorization/x-api-key instead. Add it as an unconditional fallback, after Authorization: Bearer and x-api-key, before the path-scoped URL token, in both the real enforcement gate (src/server/authz/policies/clientApi.ts::extractBearer()) and the general extractor (src/sse/services/auth.ts::extractApiKey()). The header-read/trim logic is extracted into a new leaf module (src/sse/services/googApiKeyAuth.ts) shared by both call sites, so the frozen auth.ts file only takes the minimal chokepoint wiring (config/quality/file-size-baseline.json rebaselined 2458->2461 with justification, matching this repo's established extraction pattern). Closes #7034 |
||
|
Diego Rodrigues de Sa e Souza
|
265d00c0e1 |
fix(sse): honor per-model targetFormat override for zai/glm-coding-apikey (#7364) (#7584)
DefaultExecutor.buildUrl()'s "zai"/"glm-coding-apikey" case always returned the Anthropic Messages URL, ignoring a per-model targetFormat override (custom-model dropdown, #2905) that resolves to "openai" — e.g. for a vision model like glm-4.6v. chatCore/executionCredentials.ts now threads the resolved override onto providerSpecificData.targetFormat so buildUrl (via the new default/zaiFormatOverride.ts helper, extracted to respect the file-size ratchet) can route to the OpenAI-compatible endpoint instead. Separately, custom-model id lookup (lookupCustomModelMeta in src/sse/services/model.ts, getCustomModelRow in src/lib/db/models.ts) did an exact, case-sensitive match, so a model saved as "glm-4.6v" was invisible when looked up as "glm-4.6V". Both now fall back to a case-insensitive match after the exact match fails. Regression tests: tests/unit/zai-glm-target-format-override.test.ts (reused from the triage plan-file's RED probe) and tests/unit/zai-execution-credentials-target-format-7364.test.ts (production wiring in executionCredentials.ts). Gates run: check-file-size, check-complexity, check-cognitive-complexity, typecheck:core, eslint (suppressions), tests/unit/zai-glm-target-format-override.test.ts, tests/unit/zai-execution-credentials-target-format-7364.test.ts, tests/unit/executor-default-base.test.ts, tests/unit/custom-model-target-format.test.ts, tests/unit/chatcore-execution-credentials.test.ts, tests/unit/chatcore-target-format.test.ts, tests/unit/model-resolver.test.ts, tests/unit/model-alias-provider-resolution.test.ts, tests/unit/combo-custom-provider-resolution.test.ts — all green. Refs #7364 |
||
|
Diego Rodrigues de Sa e Souza
|
a798b4d5d9 | fix: preserve relayAuth for pool-referenced relay proxies (#5716) (#7182) | ||
|
Paijo
|
b4c47f5cbf |
feat(sse): add connection backpressure for chat handler (#6590)
Add checkConnectionCapacity guard with 429 + Retry-After in handleChat(). Introduce OMNI_MAX_CONCURRENT_CONNECTIONS env-bound cap, disabled (0) by default so existing deployments are unaffected until an operator opts in. Reconstructed from PR #6590, isolating only the backpressure change — the original branch also carried unrelated headroom/docker/perf work from the author's separate #6572 branch. Co-authored-by: oyi77 <oyi77@users.noreply.github.com> Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> |
||
|
backryun
|
67a0b99240 |
feat(providers): add GPT-5.6 model family (#6862)
* feat(providers): add GPT-5.6 model family * fix(chatgpt-web): resume temporary chat handoffs * fix(codex): auto-merge discovery, filter denylist, revalidate on lifecycle Restore live/GitHub auto-merge for Codex catalogs, drop models via explicit denylist (GPT-5.4 family), and run scrub+live re-sync once on first-start, app upgrade, or setup completion. Success log: kill deprecated models complete. * fix(codex): preserve live catalog reconciliation Expose remote-only Codex models without dropping user custom entries, and complete lifecycle revalidation only after a successful internal sync. Keep credentialed self-fetches pinned to the active dashboard listener. --------- Co-authored-by: backryun <backryun@daonlab.local> |
||
|
Diego Rodrigues de Sa e Souza
|
0834bb2c70 | fix(providers): strip redundant node prefix on connId-addressed custom models (#6772) (#6890) | ||
|
Diego Rodrigues de Sa e Souza
|
5e1a325e72 |
feat(combo): strict budget-cap fallback policy for auto/* combos (#3470) (#6816)
Auto-combo transparency + budget controls: the engine's budgetCap enforcement always degraded to the globally cheapest candidate when every candidate exceeded the cap - silently overspending instead of respecting the cap. - engine.ts: budgetFallback "cheapest" (default, legacy) | "strict" (BudgetExceededError when no candidate fits budgetCap) - requestControls.ts: X-OmniRoute-Budget-Fallback header + resolveRequestAutoControls() consolidating mode/budget/fallback parsing - resolveAutoStrategy.ts / autoConfig.ts: thread combo-level config.budgetFallback and catch BudgetExceededError into an HTTP 402 - chat.ts: switch to the consolidated resolveRequestAutoControls() helper (net line reduction, stays under the frozen file-size baseline) Regression guard: tests/unit/auto-combo-budget-fallback-3470.test.ts |
||
|
SeaXen
|
6557f44bd2 |
feat(settings): 9router-style Routing Strategy card + sticky parity (#6678)
* feat(dashboard): 9router-parity Routing Strategy card + provider/combo sticky override (#6678) Add a Routing Strategy settings card (Settings -> Routing) surfacing account round-robin/sticky-limit knobs plus a new combo-level sticky round-robin (comboStickyRoundRobinLimit), and a per-provider account-routing override (providerStrategies) wired into getProviderCredentials() ahead of the global fallback strategy. Rebased onto release/v3.8.47 (credit-preserving reconstruction: unrelated package.json/electron/proxyDispatcher drift from the PR's stale base was dropped, only the author's own 12 files were re-applied). Split ProviderAccountRoutingCard/RoutingStrategyCard into smaller hook+subcomponent pieces to stay under the frozen complexity/file-size gates; rebaselined ProviderDetailPageClient.tsx/auth.ts's frozen file-size caps for the small additive growth. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * chore(quality): register combo-rr-sticky-9router.test.ts in stryker tap.testFiles (#6678) CI's Fast Quality Gates -> check:mutation-test-coverage --strict flagged the new test as missing from stryker.conf.json's tap.testFiles (it covers the mutated module open-sse/services/combo/rrState.ts). Adds the single entry, alphabetized next to the existing combo-rr-fallback-advance-948.test.ts. Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * fix(changelog): restore CHANGELOG bullets eaten by release sync Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * fix(changelog): re-restore CHANGELOG bullet after further release sync Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> * fix(changelog): re-restore CHANGELOG bullet after further release sync Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> Co-authored-by: SeaXen <SeaXen@users.noreply.github.com> |
||
|
Hernan Javier Ardila Sanchez
|
912ff8d1c4 |
fix(vision-bridge): auto-reroute non-vision models to fastest vision model when images detected (#6640)
* fix(electron): bump electron 42→43 + build better-sqlite3 from source (ABI 148) (#6605) fix(electron): bump electron 42→43 + rebuild better-sqlite3 from source against the Electron ABI (148). Electron 43 raises NODE_MODULE_VERSION to 148; better-sqlite3@12.11.1 has no electron-v148 prebuild, so the packaged app died with 'Nenhum driver SQLite disponível'. prepare-electron-standalone now compiles better-sqlite3 from source against the electron headers into build/Release (where 'bindings' resolves it). Validated by Electron Package Smoke (green) + local (node_register_module_v148). Supersedes #6378. (--admin: the only reds are SonarQube/SonarCloud failing on a coverage-report artifact digest-mismatch — a GitHub Actions infra flake, not this diff; Sonar is green on main and the diff touches only the electron build.) * deps: bump the development group across 1 directory with 6 updates (#6588) deps: bump the development group (6 updates). Rebased onto current main; all checks green after the electron-smoke fix (#6605). * fix(proxy): force CONNECT tunnel for HTTP proxied requests (undici 8.7) + production deps bump (#6620) fix(proxy): force CONNECT tunnel for HTTP proxied requests (undici 8.7) + production deps bump. undici 8.6+ changed ProxyAgent to forward plain-HTTP via request-proxy instead of CONNECT, breaking OAuth refresh through a connection proxy (501). proxyDispatcher now passes proxyTunnel:true. Validated: Unit Tests 3/8 (the OAuth-proxy test) green, new regression test green (fails without the fix on undici 8.7), SonarQube green. Supersedes #6380. (--admin: the only red is Electron Package Smoke failing on a next-build artifact 'digest-mismatch' — a GitHub Actions infra flake corrupting the asar ('file data stream has unexpected number of bytes'); the better-sqlite3 rebuild itself succeeded (gyp ok) and the electron path is unchanged from #6605 which passed the smoke. Not this diff.) * fix(vision-bridge): auto-reroute non-vision models to fastest vision model when images detected The VisionBridgeGuardrail was describing images as text via a vision model and sending text to the original (non-vision) model. This defeated the purpose when the final target was already vision-capable (auto/vision, combos with vision targets) and never actually rerouted requests to a vision model. Changes: - Individual non-vision models + images → reroute to the fastest available vision-capable model (via getBestVisionModel), keeping images intact - Auto/ prefix models (auto/vision, auto) → skip guardrail entirely, letting the auto-combo resolver handle vision-capable model selection - Combo mappings with non-vision targets → keep existing describe behavior (fallback path via checkModelHasComboMapping) - chat.ts: sync modelStr from body.model after guardrail execution so downstream routing uses the rerouted model * fix(vision-bridge): use getBestVisionModel auto-routing instead of fixed model Address Gemini review feedback: getBestVisionConfig({}) with empty object bypassed auto-routing by always defaulting to a fixed model. Auto-select the best vision model from available providers instead. * fix: compact modelStr sync to stay under file-size cap (1632) * fix: remove debug log, orphaned brace to keep file under cap * chore: trigger CI re-run with file-size fix and PR evidence * chore: rebaseline chat.ts frozen cap to 1754 (PR #6640 +3 lines) * fix(auto-combo): respect hidden models from dashboard toggle getHiddenModelsByProvider() only queried modelCompatOverrides and customModels namespaces, missing the hiddenModels namespace used by the dashboard hide/unhide toggle. Auto-combo candidates now filter out models the user explicitly hid. * fix(changelog): restore CHANGELOG bullets eaten by release sync Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com> --------- Co-authored-by: Diego Rodrigues de Sa e Souza <8016841+diegosouzapw@users.noreply.github.com> Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com> Co-authored-by: herjarsa <herjarsa@users.noreply.github.com> Co-authored-by: Diego Rodrigues de Sa e Souza <diegosouza.pw@gmail.com> |
||
|
Diego Rodrigues de Sa e Souza
|
5e5447a2ba |
fix(sse): count gate/combo-rejected requests in per-api-key usage (#6698)
Requests rejected before handleChatCore — a pipeline-gate rejection (provider circuit breaker OPEN / model cooldown) or a combo whose targets were all exhausted — short-circuited in chat.ts and only wrote a call_logs row (dashboard/logs). They never reached persistFailureUsage, so no usage_history row was created and the per-api-key usage counter (getApiKeyUsageRows reads usage_history) never incremented. An API key whose traffic was entirely gate/breaker-rejected showed zero requests despite real usage. Route both rejection paths through recordRejectedRequestUsage(), which writes the call_logs row (unchanged visibility) AND a usage_history row attributed to the api key with success:false, mirroring persistFailureUsage. Regression guard: tests/unit/rejected-request-usage.test.ts. |
||
|
Diego Rodrigues de Sa e Souza
|
f570960958 | fix(oauth): persist and reuse rotated Codex OAuth refresh token (#6352) (#6602) | ||
|
Chirag Singhal
|
687a5aefc0 |
fix(providers): reject image-only models on /v1/chat/completions with clear error (#6457) (#6525)
fix(providers): reject image-only models on /v1/chat/completions with a clear error (#6457) Integrated into release/v3.8.47. (thanks @chirag127) |