The weight table said stability accounts for "low latency stdDev / error rate". Grep errorRate in scoring.ts and you find it declared on ProviderCandidate and read nowhere — while combo.ts pulls 24 hours of usage history behind a ten-sample floor, falls back to real-time metrics, and hands every candidate an errorRate the scorer ignores. Two candidates, one failing 1% of calls and one failing 99%, scored identically at 0.459486.
This declares reliability as a sixteenth factor: 1 - failureRate, using the same formula, field precedence and rate-bounding speedRanking.ts already applies, so a corrupt reading means "nothing observed" rather than "fails every call". It ships at weight 0, leaving the ranking unchanged to the digit — the honest default, since which weight this deserves is a product call backed by traffic the author does not have. Two declared-but-silent factors already ship (cacheAffinity, resetWindowAffinity), so the pattern is not new. The stability row now describes what that factor actually computes: latency variance.
The rest is the mechanical 15 → 16 across nineteen documents and the forty-two llm.txt mirrors — sourced from check:docs-counts rather than a grep, the first real use of the gate #12316 extended.
Protected-surface note: this PR touches AGENTS.md, llm.txt and its 42 mirrors, and skills/omni-combos-routing/SKILL.md. Every changed line in those 45 files is a digit substitution and nothing else — masking all digits makes the removed and added lines identical, with no sentence added, removed or reworded. Reviewed and approved on that basis before merging.
Verified on the author's rebased head: check:docs-counts green (the gate that now enforces the count this PR moves), typecheck:core clean, and 71/71 focused tests across scoring-reliability-factor, combo-scoring-weights-schema-coverage, check-docs-counts-sync, lkgp-enabled-context, intelligent-routing-options and the combo-matrix auto integration suite.
Thanks @maxmad64bis — shipping the factor at weight 0 and saying plainly that the weight is someone else's call is the right way to land this.
docs/routing/AUTO-COMBO.md documented four mode packs; six ship. It printed 0.14 where modePacks.ts says 0.1333. And nothing was watching: four documents stated a scoring-factor count and check:docs-counts covered none of them. Wiring them up turned the gate red on seven real drifts — ARCHITECTURE.md and REPOSITORY_MAP.md at "9-factor" (code: 15) and "4 mode packs" (code: 6), RESILIENCE_GUIDE.md and SKILL.md at 13, AUTO-COMBO-GUIDE.md at both 5 and 13. ARCHITECTURE.md did not merely have the wrong number: it named nine factors that are not the engine's, and its four "mode packs" were the auto/* request prefixes.
A product fact fell out of writing the table: no pack sets quality, and applying a pack replaces the weight map wholesale (weights = pack in engine.ts, not a merge), so quality carries 0.03 by default and normalizes to 0 under any pack — pick a mode pack and the observed-quality signal stops voting. Documented, not changed.
The gate reads pack names from the module through the tsx subprocess that already reads every other code-derived count, matching the three spellings the docs actually use; on the reference document a missing claim now fails rather than passing. The dashboard was behind too (four of six packs offered); the count is dropped from the strategy label rather than corrected, since nothing reads selector labels and a right-today number goes stale unnoticed.
Verified in a combined batch worktree: 174/174 focused tests across all 11 PRs of this batch, typecheck:core clean, and check:docs-counts green with the four newly-wired documents.
Thanks @maxmad64bis — finding the quality-under-a-pack behaviour while writing a docs table is the kind of thing a table is for.
Merged via merge-train (release/v3.8.50, batch1 2026-08-20) — static gates (typecheck/file-size/complexity/cognitive/changelog) green on the combined tree; test:unit reds observed in the boarded run were verified pre-existing on the pure release tip (unrelated flake), not caused by this PR. Thanks for the contribution!
* fix(build): exec native tool binaries directly in runBuildTool
#8858 routed every resolved local bin through process.execPath to avoid
Windows .cmd shims — but esbuild >=0.25 ships bin/esbuild as the NATIVE
platform executable (ELF on Linux), so Node parsed machine code as JS and
build:cli died with 'SyntaxError: Invalid or unexpected token', turning
dast-smoke red for every PR.
runBuildTool now sniffs the entry's magic bytes (ELF / Mach-O / PE) and
execs native binaries directly; JS entries keep going through this Node
binary (the .cmd-shim avoidance #8858 wanted).
Validation (RED->GREEN on this box):
- RED: node node_modules/esbuild/bin/esbuild --version -> SyntaxError (ELF)
- GREEN: the exact failing CI step reproduced via the new logic bundles
open-sse/mcp-server/server.ts successfully (4.2MB output, 1.3s).
* fix(docs): add MDX frontmatter to the 20 remaining docs without it
Same failure class as AGENTROUTER_WAF (#9503) and DOCKER_RELEASE_CHANNELS
(this run's dast-smoke red): any doc without frontmatter breaks the
fumadocs MDX loader during next build, killing build:cli/dast-smoke for
every PR. Swept ALL of docs/ (i18n mirrors excluded) in one pass so this
class cannot recur one file at a time.
* docs(env): document OMNIROUTE_INTERNAL_SERVICE_TOKEN(+_FILE), OPENROUTER_PROVIDER_STATS_* and embedded-Redis binding vars
Pre-existing env/docs contract drift from recently merged features made
check:env-doc-sync red for any docs-touching PR. Values and defaults read
from the defining modules (internalServiceAuth.ts, openrouterProviderStats.ts).
* fix(build): resolve bundled npm-cli.js in the standard Unix layout + safe npm fallback off-Windows
The opencode-plugin step hard-failed on GitHub runners because
resolveBundledNpmEntry only looked next to the node binary (Windows zip
layout); hostedtoolcache Node keeps npm at <prefix>/lib/node_modules/npm.
Added that candidate, and when neither exists on non-Windows the step now
falls back to plain 'npm' — the .cmd-shim hazard #8858 avoids is
Windows-only.
* test(mutation): register xai-agent-tools-passthrough.test.ts in stryker tap.testFiles
The test landed on release/v3.8.50 covering
open-sse/handlers/chatCore/passthroughHelpers.ts without the stryker
registration, so Fast Quality Gates' drift detection reds any PR that
carries it. Mechanical registration so its mutant kills count.
---------
Co-authored-by: diegosouzapw <diegosouzapw@users.noreply.github.com>
