Root cause: GeminiWebExecutor.execute() opened the identical fixed
https://gemini.google.com/app URL and ran the identical Playwright
interaction sequence for every advertised gweb/<model> id. `model` was
read only AFTER the response was captured, purely to stamp the
OpenAI-shaped response — never to influence what was actually
clicked/typed, so two different advertised models produced
byte-identical automation and the response `model` field was a
caller-supplied label, not an observed fact.
Fix (owner decision, Option B): a new model -> Gemini UI mode map
(open-sse/executors/gemini-web/modeSelection.ts) drives an in-browser
selection step before anything is typed — try the mode control, read
back the active-mode indicator, and only proceed on a confirmed match.
An unconfirmed model, or a requested Extended Thinking control that
cannot be confirmed (#13381 follow-up comment), fails closed with 400
unsupported_control_for_provider instead of silently running the
account default under the requested label. The selectors involved are
UNVALIDATED (no live Gemini account from this checkout) — see the PR's
"Selector set is UNVALIDATED" section and the required live smoke.
Regression test: tests/unit/issue-13381-gemini-web-model-selection.test.ts
* fix(thinking): recognize adaptive thinking + parse/scrub DSML tool-call markers
Two defects combined to break DeepSeek-V4-Flash turns and raise 502
empty_response on Claude Code autocompact.
Defect 1 — DSML tool-call markers leaked as visible content:
DeepSeek-V4-Flash occasionally emits tool calls in a non-standard DSML
text format using full-width pipes instead of the OpenAI tool_calls JSON.
Two shapes appear in production call logs:
- complete block: <｜DSML｜:Read><path>...</path></｜DSML｜:Read>
- stray closers (truncated call): </｜DSML｜parameter></｜DSML｜invoke>
</｜DSML｜tool_calls>, sometimes trailing a system-prompt echo
The openai-compatible path never parsed these, so the markers leaked to
the client as visible content and the turn ended incomplete.
Fix: add open-sse/utils/dsmlToolCalls.ts — parseDsmlToolCalls() converts
complete DSML blocks into OpenAI tool_calls and strips stray closing
markers from content (streaming-safe via a holdback for partial openers).
Wire it into the response translator before extractXmlInvokeBlocks so
DSML and XML invoke tool calls share the same pending queue.
Defect 2 — adaptive thinking silently suppressed:
A prior inline === 'enabled' check on body.thinking.type silently
suppressed adaptive (the intent Claude Code actually sends), so
reasoning was dropped. The model then emitted DSML tool-call markers
as plain text, producing an incomplete stop finish. Fix: use
hasActiveClaudeThinking() (which recognizes enabled AND adaptive) to
set requestedThinking, thread it through stream.ts and translator
state, and gate thinking block emission on state.requestedThinking
so upstream reasoning_content only relays when the client opted in.
Tests: 29/29 (6 dsml-tool-calls, 5 thinking-active-claude-adapter,
3 translator-resp-dsml-integration, 15 translator-resp-openai-to-claude
incl. requestedThinking suppression regression). typecheck:core clean.
* fix(sse): strip echoed system-prompt preamble + preserve large analysis/summary blocks
DeepSeek-V4 and similar models echo the OMNIROUTE_SYSTEM_INSTRUCTION_APPEND
directive (appended to the system tail by claude-to-openai.ts) and whole chunks
of the system prompt (<analysis>/<system-reminder>/<summary> blocks, prose
reproductions of the superpowers skill section) verbatim at the START of their
reply — the 'system message leak' persisting after the request-side fix.
Add two streaming-safe preamble strippers in directivePreambleStripper.ts:
- createDirectivePreambleStripper(directive): drops a leading reproduction of
the exact configured directive across arbitrary SSE chunk boundaries.
- createSystemPreambleStripper(): removes <analysis>/<system-reminder>/
<summary> echo blocks and known prose heads (Phase B) from the very start
of a stream, only while the stream is still a preamble.
Wire both into openai-to-claude.ts content-delta path: chain the exact-directive
stripper then the system-echo stripper before DSML/XML-invoke parsing, so a
leading system echo is dropped before it reaches the client.
Preserve large blocks (>= SYSTEM_ECHO_THRESHOLD=1000 chars) and blocks with no
trailing content — these are the model's real response (e.g. a Claude Code
autocompact summary), not a short system-echo. Stops the autocompact
empty-response regression where a whole-summary <analysis> block was stripped
to empty (3a8515).
Regression: origin's markdown-boundary feature (bufferedPrefix /
splitMarkdownBoundary, commit 1b39873ea) is preserved — preamble strip runs
before the markdown buffer rehydration, and the scrubbed content flows into
the existing DSML/XML-invoke/markdown pipeline unchanged.
TDD: tests/unit/directive-preamble-strip.test.ts (7 cases),
system-preamble-strip.test.ts (12 cases incl. 3a8515 regression),
system-preamble-wiring.test.ts (3 integration cases); group F regression
24/24 green; typecheck:core 0 errors.
* fix(sse): gate thinking block on requestedThinking + synthesize text block for reasoning-only responses
Reasoning-content (thinking) blocks were emitted unconditionally to
Claude-format clients, leaking reasoning to thinking-opt-out clients
(Claude Code sends thinking:{type:"disabled"}) — the operator reported
'reasoning is exposed'. On reasoning-only upstream responses (GLM-5.2
autocompact pattern), the unconditional thinking block also caused either
a 502 'no content block' at flush, or — after a text-block fallback — an
autocompact 'empty response' rejection that looped the session forever.
Streaming translator (openai-to-claude.ts):
- Compute hasReasoning outside the emission gate; accumulate into
state._reasoningAccum always (so fix B can fire).
- Gate only the thinking-block EMISSION on requestedThinking === true.
- FIX B at finish: when no text block was started and requestedThinking
!== true, synthesize a text content block from _reasoningAccum so
autocompact can extract the summary (no 502, compact applies).
- Skip fix B when requestedThinking === true to avoid double-exposure
(thinking block + text block both carrying reasoning).
Non-streaming translator (responseTranslator.ts):
- Thread requestedThinking through translateNonStreamingResponse into
convertOpenAINonStreamingToClaude.
- suppressThinking = requestedThinking === false: drop the thinking block
when content is present (no leak); relay reasoning as a text block when
the response is reasoning-only (no 502). requestedThinking === undefined
keeps the legacy 'always a thinking block' relay.
chatCore.ts: pass hasActiveClaudeThinking(body) to the non-stream
translate call (inline, since the shared const is in the stream branch's
temporal dead zone here).
Tests: 25/25 (5 gate-restore, 1 gate-502-repro, 4 nonstream-leak,
15 resp-openai-to-claude incl. requestedThinking suppression regression).
Group E (22) + F (14) regression-free. typecheck:core clean.
* fix(sse): restore requestToolIdentityMap in the Codex CLI responses-translation path
The needsResponsesTranslation branch (openai-responses -> openai, used
when the client also speaks Responses) silently dropped the
requestToolIdentityMap argument to createSSETransformStreamWithLogger
when the requestedThinking parameter was added, reverting the #7936
tool-identity round-trip fix for that branch. The sibling
needsTranslation branch was updated correctly; restore the same
argument here.
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
* docs(changelog): add the 3 fragments documented in the PR body
The PR body already writes out the changelog.d/ entries for the DSML
parser (Group F), the directive-preamble stripper (Group E), and the
reasoning-gate thinking-leak fix (Group G), but none of the files
existed in the diff. Add them so the release aggregator picks them up.
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
* fix(sse): realign GLM's positional call after the new requestedThinking parameter
createSSETransformStreamWithLogger gained a new requestedThinking
parameter inserted before customToolNames. glm.ts's translateSseResponse
still called it with the pre-existing positional argument list, so the
new parameter silently absorbed the old customToolNames slot, and the
GLM_STREAM_BUFFER_BYTES tuning value (#12925) landed on
requestToolIdentityMap instead of streamBufferBytes — a TS2345 (number
is not assignable to Map<...> | null) caught by
check:open-sse-typecheck, and a real loss of GLM's 64KB stream buffer
budget. Insert an explicit `undefined` for requestedThinking to restore
the original alignment.
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
* fix(sse): make the system-preamble stripper opt-in and flush it at stream end
`createSystemPreambleStripper()` was wired DEFAULT-ON and unconditional in the
openai→claude streaming translator, unlike the exact-directive stripper right
above it, which only runs when the operator configured
OMNIROUTE_SYSTEM_INSTRUCTION_APPEND. Cause: the system-echo stripper recognises
its openers by English-prose heuristics ("# Skill usage", "# Verification
Process", <analysis>/<summary>/<system-reminder>), so leaving it always-on made
it mutate the response payload of EVERY openai→claude stream. A legitimate reply
opening with "# Skill usage: how to write one\n\nHere is the guide." lost that
whole section. It is now gated on OMNIROUTE_STRIP_SYSTEM_PREAMBLE=1, mirroring
the directive stripper's opt-in.
Second cause, same feature: neither stripper was ever flushed. Both buffer while
a construct is still undecided — a directive prefix that never completes, an
<analysis> block that never closes — and nothing released that buffer at the end
of the stream. A reply consisting of an unterminated echo block therefore reached
the client as an EMPTY message: the answer was held in the buffer and discarded
with the stripper. Both strippers now expose flush(), the finish handler calls it
for both, and the released text is emitted as a text block. A construct that WAS
finally classified as an echo is not resurrected (the drop is final).
Tests: tests/unit/system-preamble-gate-and-flush.test.ts pins the default-off
contract, the opted-in behaviour, the flush for both strippers (unit + wiring),
and the no-resurrection guard. system-preamble-wiring.test.ts now opts in
explicitly, since it exercises the stripping path.
* fix(sse): thread the client's thinking intent into the non-streaming path
The streaming and non-streaming translators disagreed on the default meaning of
`requestedThinking`, so the SAME request produced different shapes depending on
`stream`. Cause: chatCore computes the client's intent
(hasActiveClaudeThinking) and threads it into the SSE translator, which relays
reasoning as a thinking block only when it is explicitly `true` — but NO caller
ever passed it to translateNonStreamingResponse(). The non-streaming
OpenAI→Claude conversion therefore only ever saw `undefined`, its legacy
"always relay a thinking block" default, and leaked reasoning to a client that
had opted out with `thinking: {"type":"disabled"}`. The streaming plumbing also
coerced an omitted value into an explicit `false`, hiding the divergence behind
two different spellings of "no intent".
Fix (least destructive of the options): do NOT flip either gate — both encode a
deliberate, regression-tested contract — but give the non-streaming path the
same input the streaming path already has. runNonStreamingProviderLeg owns the
client body (`sourceBody`), so it computes the intent with the very same helper
and passes it down through translateNonStreamingClientResponse. `undefined`
keeps its documented back-compat relay for callers that cannot express intent
(issue-7856 / issue-6623), and stream.ts no longer defaults the parameter to
`false`, so "absent" now means the same thing in both signatures.
No content is lost by the suppression: a reasoning-ONLY response is still
relayed as an ordinary text block (no empty response, no 502) — exactly what the
streaming finish handler does.
Tests: tests/unit/nonstream-requested-thinking-parity.test.ts drives the real
provider leg with thinking disabled / enabled / adaptive.
---------
Co-authored-by: Jihyun Son <jihyun.son@sk.com>
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
* fix(claude-web): add charset=utf-8 to Content-Type headers to fix Arabic/Persian UTF-8 mojibake
Fixes#13416
The Claude Web endpoint and the outer SSE streaming pipeline were returning
Content-Type headers without an explicit charset parameter:
- stream.ts responseHeaders() returned 'application/json' and
'text/event-stream' without charset
- responseHeaders.ts buildStreamingResponseHeaders() returned
'text/event-stream' without charset
While RFC 8259 defaults JSON to UTF-8 and the SSE spec defaults
text/event-stream to UTF-8, some HTTP clients (notably VS Code Chat on
Windows) fall back to ISO-8859-1/Latin-1 when no charset is declared,
causing multi-byte UTF-8 characters to appear as mojibake.
For example, the Persian word for hello (سلام, UTF-8 bytes D8 B3 D9 84 D8 A7
D9 85) was decoded as Latin-1, producing the garbled output 'سلام'.
Fix:
- claude-web/stream.ts: append '; charset=utf-8' to the Content-Type
header in the responseHeaders() helper, with a guard to avoid double
appending if the caller already includes a charset
- chatCore/responseHeaders.ts: hardcode 'text/event-stream; charset=utf-8'
in buildStreamingResponseHeaders()
Tests:
- 11 new regression tests in claude-web-utf8-mojibake-13416.test.ts
covering Persian, Arabic, mixed-script, emoji, and chunk-boundary-split
scenarios across both streaming and buffered response paths
- All 7 existing claude-web-stream tests pass
- All 13 response header tests pass
- All 8 adaptive-admission-lifecycle tests pass
* test(sse): confirm streaming charset header and add byte-level UTF-8 repro (#13416)
Independently verified the mojibake root cause before trusting the charset
fix: OmniRoute's claude-web decoder already reconstructs a Persian/Arabic
multi-byte UTF-8 sequence split across a chunk boundary correctly because
it decodes with TextDecoder({ stream: true }); a naive per-chunk decode
(without stream state) is what actually produces the U+FFFD garbling.
Updated the 2 exact Content-Type assertions in chat-pipeline.test.ts to
match the new "text/event-stream; charset=utf-8" header, which is already
the convention used by the other streaming executors (uc.ts, maxai.ts,
codex-app-server.ts, etc).
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
---------
Co-authored-by: Koosha Pari <koosha@phenotype.ai>
Co-authored-by: diegosouzapw <8016841+diegosouzapw@users.noreply.github.com>
Root cause: when a user's Augment/Auggie quota is exhausted, the local
`auggie` CLI prints its "You have run out of usage for ..." warning to
stdout and exits with code 0. AuggieExecutor treated any clean exit as
a successful completion and wrapped that text as a normal 200 assistant
reply, so combo/fallback routing never saw a failure and kept sending
requests to the same exhausted connection.
Fix: detect the CLI's known quota-exhausted phrasing before wrapping
stdout as a completion. Non-streaming returns a 429 in-band error body
(mirroring blackbox-web.ts's precedent for HTTP-200 in-band errors);
streaming buffers the first ~2KB of stdout, and on a match emits the
existing {error:...} SSE envelope (reusing the #7880 combo quality-gate
detection) instead of forwarding the text as a delta.
Regression test: tests/unit/issue-12949-auggie-quota-exhausted-200.test.ts
Merged. The failure mode was concrete — a matched reasoning rule dropped on native Responses/Anthropic paths, model-suffix/account defaults, or fallback preparation, and `_omnirouteReasoningRule` leaking upstream as `Unsupported parameter` — and the fix is carried in the request-local credential context through dispatch, refreshed credentials and fallbacks, with forced effort winning over defaults and client-forged markers dropped at ingress. The 11-case integration suite exercises the real routing/translation modules.
Validated as a combined board first (this PR merged with the 4 siblings of the JxnLexn wave on the release tip): eslint with the frozen suppressions, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 77 passing / 0 failing focused node:test cases across the test files the wave touches. The wave's i18n fill (new keys carried to all 66 locales), free-tier doc counts and file-size rebaseline land in one follow-up PR right after the wave, as with #13904.
Thank you — and for keeping this a runtime-only change with the editor and service-tier work in their own PRs.
Merged. Root cause first: `publishers.models.list` was called with `pageSize=1000` against Google's hard maximum of 300, so every publisher answered 400 and discovery silently fell back to the stale static catalog. On top of that the PR separates API-key from Service-Account capabilities correctly (keys cannot list Model Garden — project-scoped curated catalog; SA tokens can — live catalog), stops treating the expected generativelanguage rejection of a Service Account as a discovery failure, replaces the speculative partner IDs with documented MaaS IDs, and rejects OAuth client-config JSON with a clear message instead of a misleading one.
The 5 ESLint errors flagged during the earlier fix sweep were fixed in your own follow-up commits; the branch was reconciled with the release tip and the 42 locale files were checked for lost keys before this merge.
Validated as a combined board first (this PR merged with the 4 siblings of the JxnLexn wave on the release tip): eslint with the frozen suppressions, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 77 passing / 0 failing focused node:test cases across the test files the wave touches. The wave's i18n fill (new keys carried to all 66 locales), free-tier doc counts and file-size rebaseline land in one follow-up PR right after the wave, as with #13904.
Thank you — the credential-capability distinction and the retired/non-chat filtering are what make the Vertex listing trustworthy.
Merged. Kimi emitting tool calls as history narration is a provider quirk we have to absorb rather than pass through; recovering them keeps the tool contract intact for clients that never see the quirk.
Validated as a combined board first (this PR merged with the 21 siblings of the same wave on the release tip): eslint with the frozen suppressions, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-counts, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 176 passing / 0 failing focused node:test cases across the 25 test files the wave touches and the dashboard test under Vitest (2/0). Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thank you.
Merged. Settling on `kv_after_text` while a trailing `exec_mcp` call is still pending drops the tool call entirely — the client then sees a finished turn that never ran the tool. Correct place to fix it.
Validated as a combined board first (this PR merged with the 21 siblings of the same wave on the release tip): eslint with the frozen suppressions, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-counts, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 176 passing / 0 failing focused node:test cases across the 25 test files the wave touches and the dashboard test under Vitest (2/0). Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thank you.
Merged after a maintainer rework that kept every one of @HouMinXi's commits intact.
**What the rework added on top of the contribution:** the new DB health-check behaviour is gated behind a default-off feature flag (`src/shared/constants/featureFlagDefinitions.ts`, `defaultValue: "false"`), documented in `docs/reference/FEATURE_FLAGS.md` with the description key carried into all 66 locales, so the release default is unchanged and the new bounds only apply when an operator opts in. The optional-FTS5 migration set was reconciled by hand with the "180" entry that landed meanwhile (`src/lib/db/migrationRunner/constants.ts`).
**Carried from your rebased head:** the `/api/db/health` local-only classification in `src/server/authz/routeGuard.ts` plus its `routeGuard` assertion — `runManagedDbHealthCheck()` forks native diagnostics into a child process, so Hard Rules #15/#17 apply. Re-verified here: 37 pass / 0 fail.
Validated as a combined board first (this PR merged with the 21 siblings of the same wave on the release tip): eslint with the frozen suppressions, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-counts, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 176 passing / 0 failing focused node:test cases across the 25 test files the wave touches and the dashboard test under Vitest (2/0). Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thank you for the depth of this one — the resource-bounds suite and the sql.js startup/backup coverage are the kind of tests that keep a database layer honest.
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
Merged in the 2026-09-16 sweep of the maintainer's own open PRs, at the owner's explicit instruction. No push was made to the PR branch: the merge took the head as the owning session left it (verified OPEN, non-draft and MERGEABLE against the release tip immediately before merging).
SenseNova DeepSeek V4 Flash now clamps `xhigh`/`max` down to `high` before the generic max-tier rewrite, and the model stops advertising `xhigh` — both values were rejected upstream. The clamp covers the first-party `sensenova` provider and openai-compatible connections that address the model as `snova/deepseek-v4-flash`.
Explicitly preserved: `sensenova/glm-5.2` and `deepseek-v4-flash` on `cmd`/`opencode-go`/`ollama-cloud` keep `max`.
Validated as a combined board first (this PR merged with the 11 siblings of the same batch on the release tip): eslint on every changed file with the suppressions file, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 275 passing / 0 failing focused node:test cases across the 28 test files the batch touches. Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thanks @HouMinXi!
Adds `claude` to the providers whose trailing text-only assistant turn is stripped before dispatch — official Claude rejects assistant prefill with `400 This model does not support assistant message prefill`.
Maintainer note: the strip applies to the whole `claude` provider family (API key as well as OAuth), matching what the Vertex-hosted Claude path (`open-sse/executors/antigravity.ts`), the Copilot path (`open-sse/executors/github.ts`) and the MITM handler already do unconditionally.
Validated as a combined board first (this PR merged with the 11 siblings of the same batch on the release tip): eslint on every changed file with the suppressions file, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 275 passing / 0 failing focused node:test cases across the 28 test files the batch touches. Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thanks @HouMinXi!
The Codex executor now whitelists the wire `reasoning` object to `effort`/`summary` before dispatch instead of spreading whatever the client sent, and maps `reasoning.enabled === false` to `effort: "none"` when no more specific effort was requested. OpenRouter-style keys (`enabled`, `max_tokens`, `exclude`) were reaching the Responses API and 400-ing the whole combo target with `Unknown parameter: 'reasoning.<key>'`.
The precedence chain keeps an explicit per-request effort ahead of `enabled: false`, and the strip matches the siblings already removed in the same function (`truncation`, `user`, `prompt_cache_retention`).
Validated as a combined board first (this PR merged with the 11 siblings of the same batch on the release tip): eslint on every changed file with the suppressions file, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 275 passing / 0 failing focused node:test cases across the 28 test files the batch touches. Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thanks @HouMinXi!
On a 429 the opencode executor now records the refused proxy's key in the request-local tried-set, exactly like the 403/451, 5xx, stall and network arms already did — so a second account sharing that same proxy is not dialed and refused again before the loop reaches a genuinely different route (direct, or another proxy).
Reviewed against the tip that already carries your 38 merges from this evening: this is additive to the flags that landed today (`OPENCODE_RATE_LIMITED_429_EARLY_STOP`, `PROXY_SKIP_RECENTLY_FAILED`, `OPENCODE_USER_BLOCKED_ROTATION`, `OPENCODE_TRANSIENT_FAILOVER_BACKOFF`) and does not double-skip when combined with them; direct accounts have a null proxy key and correctly record nothing.
Validated as a combined board first (this PR merged with the 11 siblings of the same batch on the release tip): eslint on every changed file with the suppressions file, typecheck:core, check:open-sse-typecheck, complexity, cognitive-complexity, changelog-integrity, i18n new-key coverage, docs-sync, migration-numbering, provider-consistency and a duplicate-identifier audit all green, plus 275 passing / 0 failing focused node:test cases across the 28 test files the batch touches. Then re-validated alone on the fresh tip before this merge: conflicts re-resolved, file sizes rebaselined for this PR's own growth, eslint and this PR's focused tests re-run.
Thanks @maxmad64bis!
The opencode executor classifies rate-limited 429 bodies (`classify429`, with real tests) and, when a whole account wave is exhausted, returns the last real upstream 429 — status, body, `Retry-After` and quota headers intact — so the provider error rules (monthly-quota cooldown) keep working.
Maintainer rework before merge (kept the idea, no default behavior change):
- The original stopped the cross-account wave at the first classified 429 and replaced the response with a synthetic one that dropped the body and headers; stopping early is now opt-in behind `OPENCODE_RATE_LIMITED_429_EARLY_STOP` (default off), the rate-limited account is still cooled down, the body is read as a bounded 8 KiB prefix from a clone and the original is never consumed, and the unused `status` input is gone.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Behind the new `OPENCODE_TRANSIENT_FAILOVER_BACKOFF` flag (default off), after two consecutive transient upstream failures the opencode rotation pauses before each later account (1.5s, 3s, 6s, capped at 10s per request) instead of hammering the upstream.
Maintainer rework before merge (kept the idea, no default behavior change):
- The pause honors the client abort signal (no dispatch after a disconnect), the failed attempt's body is cancelled before sleeping, `transientRetryDelayMs` now uses its arguments, and the sleep is injectable so the tests run without real timers.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Behind the new `OPENCODE_USER_BLOCKED_ROTATION` flag (default off), a 403 or 451 carrying `user_blocked` on a proxied opencode account rotates at most once to the next account instead of being returned as-is.
Maintainer rework before merge (kept the idea, no default behavior change):
- 403 and 451 are handled by one predicate (the original returned 451 without rotation), the refused account gets a cooldown and joins the tried-set, and the response body of the attempt rotated away from is cancelled.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Behind the new `OPENCODE_RESPONSES_STALL_ROTATION` flag (default off): a streamed Responses reply with no first body byte within `RESPONSES_FIRST_BYTE_TIMEOUT_MS` (15s) cools the account and rotates once; a second stall fails fast instead of waiting the 80s readiness timeout.
Maintainer rework before merge (kept the idea, no default behavior change):
- The TLS first-byte watchdog from #12656 is restored byte for byte (the PR had changed its pump, timer and cancel); the stall guard lives in its own module.
- Proxy-less multi-account setups now rotate the same way as proxied ones (the original threw for them), a client abort during the wait rethrows instead of rotating, and the env var is documented as flag-only.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Behind the new `PROXY_SKIP_RECENTLY_FAILED` flag (default off), pool rotation and the opencode account rotation remember a proxy that just failed (refused probe or 429) and skip it for a doubling cooldown instead of re-serving it immediately.
Maintainer rework before merge (kept the idea, no default behavior change):
- The original was on by default and re-queried the DB on every request while a member was set aside; selection now caches a refusal sequence number and re-runs the cascade once per set-aside event.
- `src/lib/db` no longer imports the heavy dispatcher for key normalization (a parity test guarantees the same key as `proxyConfigToUrl()`); `.env.example` and `ENVIRONMENT.md` document the default as false.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Registers Muse Spark 1.3 (with its effort aliases) on OpenCode Zen, OpenCode and OpenCode Go with `targetFormat: openai-responses` and a 1M context window, so the model no longer falls back to `/chat/completions` (#12674, #12698). Superset of #12675, #12973 and #13111, whose authors are credited in the PR.
Validated first on the combined board of all 38 PRs of this batch (10 merged as-is, 28 after the maintainer rework) on top of release/v3.8.51 c0f92ec: typecheck:core, check:open-sse-typecheck and check:dashboard-typecheck clean; ESLint clean on every changed file; file-size (rebaselined for the combined growth), complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync, migration-numbering and i18n new-key gates green; 735 focused node:test cases with the only batch-caused failure (a flag-count assertion) fixed. Then re-validated alone on the fresh release tip right before this merge: ESLint on the changed files, typecheck:core, check:open-sse-typecheck, the file-size/complexity/changelog gates and this PR's own tests.
Thanks @maxmad64bis!
Zed-hosted requests are re-mapped to the narrower enums the Zed proxy validates. Responses input items with `role: "developer"` go back to `system` (#13362). For Gemini, `safetySettings[].threshold: "OFF"` becomes `BLOCK_NONE` and `functionCallingConfig.mode` `VALIDATED`/`AUTO`/`ANY`/`NONE` become Zed's lowercase `auto`/`any`/`none` (#13363). Chosen over #13541, which carried the identical commit.
Maintainer changes: the inline mapping used five `as any` casts, which `no-explicit-any` rejects in `open-sse/`. It is now two typed helpers (`adaptGeminiRequestForZed`, `adaptResponsesRequestForZed`, exposed through `__test__`) plus a small mode table; behavior is unchanged. Added `tests/unit/zed-hosted-proxy-enums-13362.test.ts` (4 cases). Zed suites 47/47; `check:open-sse-typecheck` clean.
Validated in one consolidated batch of this series (37 PRs boarded together on `release/v3.8.51`): `typecheck:core`, `check:open-sse-typecheck` and `check:dashboard-typecheck` clean; ESLint clean on every changed file; file-size, complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync and migration-numbering gates green (only the pre-existing `open-sse/utils/stream.ts` file-size red remains, inherited from the base); 3,743 focused `node:test` cases plus 34 vitest cases green.
Thanks @KooshaPari!
`reka-flash-3` is marked `supportsReasoning` + `alwaysReasons` (with its 8192 output cap). `DefaultExecutor.ensureThinkingBudget` treats always-on reasoning as an implicit opt-in, so a small caller `max_tokens` gets the 4096 floor instead of being spent on reasoning and returning an empty answer (#13198). `alwaysReasons` is documented on `RegistryModel`.
Maintainer addition: `tests/unit/reka-flash-3-always-reasons-13198.test.ts` asserts the floor for `reka-flash-3` without any reasoning settings (fails on the release tip) and that `reka-flash` keeps the caller value. The branch also carried the #13532 Compare-scroll commit, which landed first; its hunks are identical, so this squash only adds the reka change.
Validated in one consolidated batch of this series (37 PRs boarded together on `release/v3.8.51`): `typecheck:core`, `check:open-sse-typecheck` and `check:dashboard-typecheck` clean; ESLint clean on every changed file; file-size, complexity, cognitive-complexity, changelog-integrity, docs-counts, docs-sync and migration-numbering gates green (only the pre-existing `open-sse/utils/stream.ts` file-size red remains, inherited from the base); 3,743 focused `node:test` cases plus 34 vitest cases green.
Thanks @KooshaPari!
Merged as part of the owner batch of 2026-09-11.
This PR had a live worktree in another session, so it sat outside the main 39. Merged on your explicit call, validated first rather than taken on trust: boarded with the other 10 worktree-held PRs into a consolidated worktree off `release/v3.8.51`.
- ESLint over every changed file: no errors
- `typecheck:core` clean; `check:dashboard-typecheck` OK; `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437
- 203 of 208 assertions green. The 5 remaining (`guide-settings-route` ×4, `hard-session-lease-bypass-inventory` ×1) reproduce on the pure tip with nothing from this batch applied.
- `imageGeneration.ts` rebaselined 3259 → 3293 for #12945's image-only-model guard, landed separately in #13392 so nothing was pushed onto a live branch.
⚠️ base-red inherited: #12732 — provider count 356 vs 358 and `open-sse/utils/stream.ts` 3115 > frozen 3098, both reproducing on the pure tip.
Merged as part of the 39-PR owner batch of 2026-09-11, validated as a unit.
Boarded into one consolidated worktree cut from `release/v3.8.51` with the other 38 — zero conflicts between them.
- ESLint over every changed file: no errors (the only finding was one suppression entry the batch emptied, pruned on #13243)
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437 — both under baseline
- 256 assertions green: 246 under node:test and 10 under vitest, which is where `tests/unit/**/*.test.tsx` actually runs
- `check-file-size`: `chatCore.ts` rebaselined 6144 → 6146 for #13278 and #13276, annotated and landed on #13243⚠️ base-red inherited: #12732 — the provider count (356 in the docs vs the 358 the modules define) and `open-sse/utils/stream.ts` at 3115 > frozen 3098 both reproduce on the pure tip with zero contribution from this batch.
Merged as part of the 39-PR owner batch of 2026-09-11, validated as a unit.
Boarded into one consolidated worktree cut from `release/v3.8.51` with the other 38 — zero conflicts between them.
- ESLint over every changed file: no errors (the only finding was one suppression entry the batch emptied, pruned on #13243)
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437 — both under baseline
- 256 assertions green: 246 under node:test and 10 under vitest, which is where `tests/unit/**/*.test.tsx` actually runs
- `check-file-size`: `chatCore.ts` rebaselined 6144 → 6146 for #13278 and #13276, annotated and landed on #13243⚠️ base-red inherited: #12732 — the provider count (356 in the docs vs the 358 the modules define) and `open-sse/utils/stream.ts` at 3115 > frozen 3098 both reproduce on the pure tip with zero contribution from this batch.
Merged as part of the 39-PR owner batch of 2026-09-11, validated as a unit.
Boarded into one consolidated worktree cut from `release/v3.8.51` with the other 38 — zero conflicts between them.
- ESLint over every changed file: no errors (the only finding was one suppression entry the batch emptied, pruned on #13243)
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437 — both under baseline
- 256 assertions green: 246 under node:test and 10 under vitest, which is where `tests/unit/**/*.test.tsx` actually runs
- `check-file-size`: `chatCore.ts` rebaselined 6144 → 6146 for #13278 and #13276, annotated and landed on #13243⚠️ base-red inherited: #12732 — the provider count (356 in the docs vs the 358 the modules define) and `open-sse/utils/stream.ts` at 3115 > frozen 3098 both reproduce on the pure tip with zero contribution from this batch.
Merged as part of the 39-PR owner batch of 2026-09-11, validated as a unit.
Boarded into one consolidated worktree cut from `release/v3.8.51` with the other 38 — zero conflicts between them.
- ESLint over every changed file: no errors (the only finding was one suppression entry the batch emptied, pruned on #13243)
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437 — both under baseline
- 256 assertions green: 246 under node:test and 10 under vitest, which is where `tests/unit/**/*.test.tsx` actually runs
- `check-file-size`: `chatCore.ts` rebaselined 6144 → 6146 for #13278 and #13276, annotated and landed on #13243⚠️ base-red inherited: #12732 — the provider count (356 in the docs vs the 358 the modules define) and `open-sse/utils/stream.ts` at 3115 > frozen 3098 both reproduce on the pure tip with zero contribution from this batch.
Merged as part of the 39-PR owner batch of 2026-09-11, validated as a unit.
Boarded into one consolidated worktree cut from `release/v3.8.51` with the other 38 — zero conflicts between them.
- ESLint over every changed file: no errors (the only finding was one suppression entry the batch emptied, pruned on #13243)
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK
- complexity 2821 / baseline 3218 and cognitive-complexity 1272 / baseline 1437 — both under baseline
- 256 assertions green: 246 under node:test and 10 under vitest, which is where `tests/unit/**/*.test.tsx` actually runs
- `check-file-size`: `chatCore.ts` rebaselined 6144 → 6146 for #13278 and #13276, annotated and landed on #13243⚠️ base-red inherited: #12732 — the provider count (356 in the docs vs the 358 the modules define) and `open-sse/utils/stream.ts` at 3115 > frozen 3098 both reproduce on the pure tip with zero contribution from this batch.
Three related things that genuinely belong together: the registry and suffix splitter not knowing Astra while the live catalog serves it, the Codex client pin Astra requires, and dropping `next/font/google` so a production image build does not reach fonts.googleapis.com.
---
Validated in one consolidated worktree cut from `release/v3.8.51`, boarded with the other 19 PRs of this batch. Two in-batch conflicts, both additive and resolved by keeping each side: the `ENVIRONMENT.md` table (#13035 + #13011) and the `chatHelpers.ts` import block (#12975 on the tip + #13017).
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, within baseline); `check:changelog-integrity` OK; `check:docs-counts` migrations ✓
- complexity 2816 / baseline 3218 and cognitive-complexity 1271 / baseline 1437 — both under baseline
- 531 of 532 focused assertions green across the batch's 46 test files
- `check-file-size` rebaselined for the batch's real growth (annotation `_rebaseline_2026_09_11_mergebatch_v3851_houminxi`, landed on #13038), attributed per PR
The single red is **not this batch**: `tests/unit/combo/quota-weighted-strategy.test.ts` → "A/B isolation: 7 hard-empty + 2 at 0.5% + 1 at 40%, floor=1" asserts an order between two connections of identical weight and flakes on the pure tip too — 2 failures in 4 runs at `origin/release/v3.8.51` with nothing from this batch applied.
⚠️ base-red inherited: #12732 — `Docs Gates`, `Merge integrity`, `No new ESLint warnings`, `Unit Tests fast-path` and `Fast Quality Gates` reproduce on the pure tip (provider count 356 vs the 358 the modules define, SKILL.md drift, and `open-sse/utils/stream.ts` at 3115 > frozen 3098, untouched here).
Thanks @HouMinXi — the live evidence on these (X500 logs, `storage.sqlite` state, real `/v1/models` probes, the 36-minute outage write-up) is what let a 20-PR batch be reviewed as a unit.
`DEVIN_AGENTIC_HOME` was unusable on every Windows host whatever its value, because a forward-slash `.sandbox` check can never match a backslash path. Normalizing separators before comparing, the same way `normalizeCommandToken` does, is the consistent fix; keeping `path.isAbsolute()` untouched keeps the guard intact.
---
Validated in one consolidated worktree cut from `release/v3.8.51`, boarded with the rest of this batch — zero conflicts between the 19 PRs.
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, all within the frozen baseline); `check:changelog-integrity` OK
- complexity 2802 / baseline 3218 and cognitive-complexity 1267 / baseline 1437 — both under baseline
- 226 of 228 focused assertions green across the batch's 23 test files. The 2 remaining belong to #12551, which is held separately.
Two batch-owned defects were found and fixed in flight, both pure base drift: `173_xp_action_counts.sql` collided with `173_call_logs_video_content_removed.sql` (renumbered to 176 on #12651 — it aborted every DB open, which is what 53 of the first run's failures were), and the feature-flag catalog was missing the `SERVER_OWNED_TOOL_LOOP_ENABLED` row the base gained after #12552 was written.
⚠️ base-red inherited: #12732 — `Docs Gates`, `Merge integrity`, `No new ESLint warnings`, `Unit Tests fast-path` and `Fast Quality Gates` reproduce on the pure tip (provider count 356 vs the 358 the modules define, SKILL.md drift, and `open-sse/utils/stream.ts` at 3115 > frozen 3098, which this batch does not touch).
Thanks @pacocartones — the `file:line` citations and the explicit out-of-scope notes on every one of these made a 19-PR batch reviewable in one pass.
Right fix: `thoughtsTokenCount` is real output the caller paid for, so folding it into `completion_tokens` and surfacing it as `completion_tokens_details.reasoning_tokens` matches what every other reasoning-capable provider reports.
---
Validated in one consolidated worktree cut from `release/v3.8.51`, boarded with the rest of this batch — zero conflicts between the 19 PRs.
- `typecheck:core` clean; `check:dashboard-typecheck` OK (206 pre-existing, all within the frozen baseline); `check:changelog-integrity` OK
- complexity 2802 / baseline 3218 and cognitive-complexity 1267 / baseline 1437 — both under baseline
- 226 of 228 focused assertions green across the batch's 23 test files. The 2 remaining belong to #12551, which is held separately.
Two batch-owned defects were found and fixed in flight, both pure base drift: `173_xp_action_counts.sql` collided with `173_call_logs_video_content_removed.sql` (renumbered to 176 on #12651 — it aborted every DB open, which is what 53 of the first run's failures were), and the feature-flag catalog was missing the `SERVER_OWNED_TOOL_LOOP_ENABLED` row the base gained after #12552 was written.
⚠️ base-red inherited: #12732 — `Docs Gates`, `Merge integrity`, `No new ESLint warnings`, `Unit Tests fast-path` and `Fast Quality Gates` reproduce on the pure tip (provider count 356 vs the 358 the modules define, SKILL.md drift, and `open-sse/utils/stream.ts` at 3115 > frozen 3098, which this batch does not touch).
Thanks @pacocartones — the `file:line` citations and the explicit out-of-scope notes on every one of these made a 19-PR batch reviewable in one pass.
Reconciled and merged. This branch was stacked on #12941, which has since landed, so it read as 1484 additions across 16 files and CONFLICTING. I merged the current `release/v3.8.51` into it rather than rewriting your branch: `open-sse/executors/opencode.ts` conflicted in six places where your side was a strict superset of the squashed #12941, and the tip had touched that file through nothing but #12941, so your side was taken whole. The PR now reads as its real 12 files, +839/-25.
The change itself is right: a flapping upstream 5xx aborting the whole agentic chain is exactly the case where rotating to the next healthy account is safe, and keeping it a separate arm from the 400-empty branch matters because that one has to clone-read the body while this one never touches it. Threading `correlationId` through so interleaved requests stay attributable — and never fabricating one when absent — is the right discipline.
Both geo-block regression suites pass alongside the new ones (43/43 across the five opencode test files), which is what proves the conflict resolution preserved #12941's behaviour.
I also tightened the batch's file-size rebaseline here: `src/sse/handlers/chat.ts` needed no bump at all (it lands at 2452, under its existing 2458 freeze) and `open-sse/executors/base.ts` needed only your +2. An earlier measurement had included a local prettier reformat that is not part of this branch.
---
Validated in one consolidated worktree cut from `release/v3.8.51`, boarded together with the rest of this batch — zero conflicts between them.
- `typecheck:core` clean; `check:changelog-integrity` OK
- complexity 2799 / baseline 3218 and cognitive-complexity 1265 / baseline 1437 — both under baseline
- 86 focused assertions green across the batch's 10 unit test files, plus 16/16 on the v1 plugin option schema and 16/16 on the v2 option tests
- `check-file-size` rebaselined for this batch's real growth (annotation `_rebaseline_2026_09_11_mergebatch_v3851_maxmad_opencode`, landed on #13141). `open-sse/utils/stream.ts` was deliberately left frozen: it is already 3115 > 3098 on the pure tip with zero contribution from this batch.
⚠️ base-red inherited: #12732 — `Docs Gates`, `Merge integrity`, `No new ESLint warnings`, `Unit Tests fast-path` and `Fast Quality Gates` all reproduce on the pure `release/v3.8.51` tip (provider count 356 vs the 358 the modules define, SKILL.md drift, and the `stream.ts` freeze above). None of them touch these diffs.
Thanks @maxmad64bis.
Rebased onto the tip and completed, per the maintainer's call to finish the wiring rather than merge the capability alone.
What changed since your version:
The tip had already cleared the TS2554 by deleting the 16th argument, leaving a comment that the highWaterMark stays at the helper default. So the base-red you found is gone, but the 64 KB #12179 asked for was still not applied and your new parameter had no caller. glm.ts now passes it, which is what turns the capability into the fix.
Your test file also hung the runner: every stream createSSEStream builds arms a 10s idle watchdog via setInterval in start, and nothing cancelled them, so node:test waited on a non-empty event loop long after the assertions passed. Cancelling each readable in an after hook runs the cancel handler that clears the timer — the file now reports in about 7 seconds. Worth knowing for future stream tests.
Your five assertions are unchanged and all pass. Reading the writable's desiredSize to measure the queue budget the stream was actually built with, rather than standing in for it, is the detail that makes this testable at all — and the 0-budget case pinning `??` against `||` is the kind of thing that silently rots otherwise.
Thank you also for separating your own red checks from the base's and reporting what you found there. That is how #12919's identical failures got explained instead of chased.
Boarded with 13 sibling PRs into one worktree off release/v3.8.51 and validated as a set: 132 focused tests pass across all 15 test files in the batch, typecheck:core is clean, check-changelog-integrity reports no lost base bullets, and check-file-size is green. Your PR merged without conflict against its siblings.
Thank you — the write-up made this reviewable: measuring the behaviour on the release tip and showing the before/after table meant the defect could be confirmed rather than taken on faith.
Validado numa worktree combinada com a onda de streaming desta leva sobre `release/v3.8.51`: typecheck:core limpo, check-api-typecheck OK (289), check-file-size OK após rebaseline, 88/88 nos testes focados.
Estender a rotação que já existe para 429 ao 403 de bloqueio geográfico é a generalização certa, e manter a rejeição de fingerprint (Cloudflare 1010) fora dela é o que impede a rotação de queimar todas as contas contra uma recusa que não é de egresso.
Nota: os checkboxes de validação do corpo ficaram em branco, mas o diff traz dois arquivos de teste — vale marcar da próxima para o revisor não precisar conferir.
Validado numa worktree combinada com as 16 PRs desta leva sobre `release/v3.8.51`: typecheck:core limpo, check-file-size e check-changelog-integrity OK, complexity 2788/3218 e cognitive 1261/1437 (ambos sob a baseline), ESLint 0 erros nos 152 arquivos alterados, 771 testes unitários focados, 49 de integração e a suíte vitest:ui completa (2149) verdes.
O `65536` era o 16º posicional de um helper com 15 parâmetros — `TS2554` vivo no tip (`open-sse/executors/glm.ts:244`, confirmado aqui antes do board). O teste de guarda de aridade é o que impede a reincidência: ele checa a assinatura do helper e o call site, não o comportamento, que é exatamente onde o erro morava.
Obrigado por isolar isso do #12711 em vez de deixar o `glm.ts` viajar junto com pin/combo-split/moonshot.
Validado em lote numa worktree combinada com os 4 PRs desta leva sobre o tip de `release/v3.8.51`: `typecheck:core` limpo e **119/119** nos 9 arquivos de teste que trazem.
Três dos quatro conflitavam apenas no `config/quality/file-size-baseline.json`, todos de forma aditiva (chaves `_rebaseline_` distintas que devem coexistir); resolvidos com validação de JSON a cada passo.
Registro que o **#12637 não é duplicata do #12566**, apesar do título quase idêntico: o autor documenta que aquele escopou o cooldown de preflight por família e este cobre o `genericQuotaFetcher`, que é o que o roteamento reset-aware efetivamente chama. Traz também validação ao vivo em VPS (imagem X500, `onmi-gemini3.6` → HTTP 200), satisfazendo a Hard Rule #18.
Obrigado, @HouMinXi.
Validado sobre o tip de `release/v3.8.51` depois de reconciliar com o #12620, que entrou primeiro nesta mesma sessão e ataca a mesma classe de problema por outra arquitetura.
**A colisão e como foi resolvida.** O #12620 consertou o GHSA-qv45-56jc-4wmj adicionando `RAW_CREDENTIAL_PATTERNS` a `error.ts` e importando-os em `upstreamErrorPassthrough.ts`. Este PR resolve o mesmo problema quebrando `error.ts` em `errorSanitization.ts` + `errorPathRedaction.ts`. Mantive a divisão em módulos deste PR, porque ao comparar os dois vocabulários o dele já era mais amplo: o `STRONG_CREDENTIAL_TOKEN` daqui cobre `sk-`/`sk_` **com lookbehind e uma variante para a forma embutida** (que pega `sk-proj-…`), mais Slack `xox-`, AWS `AKIA`/`ASIA`, `github_pat_`/`ghp_`/`glpat-` e JWT de três segmentos.
A única forma que o #12620 carregava e este conjunto não tinha era a chave do Google (`AIza…`) — adicionada aqui, com o mesmo quantificador limitado que os irmãos usam (AGENTS.md → PII §1, já que isso roda sobre corpos upstream não confiáveis).
**A verificação não foi por inspeção.** Rodei as suítes do próprio #12620 contra esta estrutura: **48/48** em `error-sanitizer-sk-key-qv45`, `bifrost-relay-response-leak-9m72`, `search-baseurl-client-override-3f8g` e `search-baseurl-ssrf-guard` — incluindo a asserção anti-drift daquela suíte, que é o oráculo certo aqui: *para todo corpo que a camada de passthrough recusa como vazante, o sanitizador de fallback não pode devolvê-lo intacto*. Ela passa, então a propriedade de segurança dos três GHSAs sobrevive à troca de arquitetura.
Os 21 arquivos de teste deste PR: **259/259**. `typecheck:core` limpo.
Validado após reconciliar com o #12465, que entrou primeiro e criou o mesmo arquivo novo `open-sse/utils/streamReadiness.ts` com desenho divergente de cancelamento.
Mantive a versão desta branch, que defere o release do lock para quando a leitura em voo termina e faz `reader.cancel()` fire-and-forget — assim uma promise de provider que nunca resolve não torna o cancelamento ilimitado. A escolha não foi por preferência: rodei as suítes dos **dois** PRs contra ela, 21/21 no readiness compartilhado e **22/22** incluindo o boundary do Perplexity do próprio #12465. typecheck:core limpo.
Validado em lote numa worktree combinada com os 14 PRs desta campanha de error-boundary sobre o tip de `release/v3.8.51`: `typecheck:core` limpo e **120/120** nos 23 arquivos de teste que os PRs trazem.
Um ponto que só apareceu no tree combinado: **#12465 e #12466 criam o mesmo arquivo novo** `open-sse/utils/streamReadiness.ts` (que não existe no tip) com desenhos divergentes de cancelamento — `cancelled` + `releaseLock` imediato num, `readInFlight`/`cancelRequested` com `cancelReader` fire-and-forget no outro. Adotei a versão do #12466, que difere e defere o release do lock para quando a leitura em voo termina, e validei a escolha rodando as suítes dos **dois** PRs contra ela: 21/21 no readiness compartilhado e 22/22 incluindo o boundary do Perplexity.
Validado em lote numa worktree combinada com os 14 PRs desta campanha de error-boundary sobre o tip de `release/v3.8.51`: `typecheck:core` limpo e **120/120** nos 23 arquivos de teste que os PRs trazem.
Um ponto que só apareceu no tree combinado: **#12465 e #12466 criam o mesmo arquivo novo** `open-sse/utils/streamReadiness.ts` (que não existe no tip) com desenhos divergentes de cancelamento — `cancelled` + `releaseLock` imediato num, `readInFlight`/`cancelRequested` com `cancelReader` fire-and-forget no outro. Adotei a versão do #12466, que difere e defere o release do lock para quando a leitura em voo termina, e validei a escolha rodando as suítes dos **dois** PRs contra ela: 21/21 no readiness compartilhado e 22/22 incluindo o boundary do Perplexity.
Validado em lote numa worktree combinada com os 14 PRs desta campanha de error-boundary sobre o tip de `release/v3.8.51`: `typecheck:core` limpo e **120/120** nos 23 arquivos de teste que os PRs trazem.
Um ponto que só apareceu no tree combinado: **#12465 e #12466 criam o mesmo arquivo novo** `open-sse/utils/streamReadiness.ts` (que não existe no tip) com desenhos divergentes de cancelamento — `cancelled` + `releaseLock` imediato num, `readInFlight`/`cancelRequested` com `cancelReader` fire-and-forget no outro. Adotei a versão do #12466, que difere e defere o release do lock para quando a leitura em voo termina, e validei a escolha rodando as suítes dos **dois** PRs contra ela: 21/21 no readiness compartilhado e 22/22 incluindo o boundary do Perplexity.
Validado em lote numa worktree combinada com os 14 PRs desta campanha de error-boundary sobre o tip de `release/v3.8.51`: `typecheck:core` limpo e **120/120** nos 23 arquivos de teste que os PRs trazem.
Um ponto que só apareceu no tree combinado: **#12465 e #12466 criam o mesmo arquivo novo** `open-sse/utils/streamReadiness.ts` (que não existe no tip) com desenhos divergentes de cancelamento — `cancelled` + `releaseLock` imediato num, `readInFlight`/`cancelRequested` com `cancelReader` fire-and-forget no outro. Adotei a versão do #12466, que difere e defere o release do lock para quando a leitura em voo termina, e validei a escolha rodando as suítes dos **dois** PRs contra ela: 21/21 no readiness compartilhado e 22/22 incluindo o boundary do Perplexity.
