test(radar): localize canonical feed fixture

.github/workflows/quality.yml

@@ -151,6 +151,63 @@ jobs:
           key: eslint-${{ runner.os }}-${{ hashFiles('eslint.config.mjs', 'eslint.complexity-ratchets.config.mjs', 'config/quality/eslint-suppressions.json', 'package-lock.json') }}
           restore-keys: |
             eslint-${{ runner.os }}-
+      - run: npm run check:provider-consistency
+      - run: npm run check:fetch-targets
+      # docs-all / openapi-routes / docs-symbols live in docs-gates (path-filtered).
+      - run: npm run check:deps
+      # #8522: --base-ref mode for PR events — compare against max(frozen, base) so
+      # inherited drift (base already over frozen cap) doesn't red an innocent PR.
+      # workflow_dispatch (no PR base) falls back to absolute comparison.
+      - name: File-size ratchet (base-relative on PR)
+        env:
+          PR_BASE_SHA: ${{ github.event.pull_request.base.sha }}
+        run: |
+          if [ -n "$PR_BASE_SHA" ]; then
+            npm run check:file-size -- --base-ref "$PR_BASE_SHA"
+          else
+            npm run check:file-size
+          fi
+      - run: npm run check:error-helper
+      - run: npm run check:migration-numbering
+      - run: npm run check:public-creds
+      - run: npm run check:db-rules
+      - run: npm run check:known-symbols
+      - run: npm run check:route-guard-membership
+      - run: npm run check:test-discovery
+      - run: npm run check:test-runner-api
+      # Guards tap.testFiles drift: a covering unit test absent from stryker.conf.json
+      # tap.testFiles makes its module's mutants survive on a cold nightly-mutation run,
+      # false-failing the blocking mutationScore ratchet. See check-mutation-test-coverage.mjs.
+      - run: npm run check:mutation-test-coverage
+      - run: npm run check:any-budget:t11
+      # Build-scope guard: fails if worktrees/cruft leak into the tsconfig include
+      # scope (would OOM `next build`). Instant. See incident 2026-06-25 / #5031.
+      - run: npm run check:build-scope
+      # Pack-policy (unexpected-files allowlist) WITHOUT a build — catches a stray file
+      # leaking into the npm tarball (v3.8.36: 6 ops bin/*.sh) per-PR instead of only on
+      # the release PR's heavy Package Artifact job.
+      - run: npm run check:pack-policy
+      # Complexity + cognitive-complexity: ONE ESLint walk (both baselines still
+      # enforced separately by ruleId). Avoids two cold tree walks on fast-path.
+      - run: npm run check:complexity-ratchets
+      # ── G0 (trilho .50): gates do trilho A que faltavam no trilho B ──────────────
+      # The god-file refactor happens in PRs→release/**; without these, the release
+      # rail never sees a new import cycle, dead code, duplication or a security
+      # regression until the release PR to main. Deliberately NOT brought here:
+      # bundle-size (self-skips without a build — this rail's build job is advisory
+      # and uploads nothing, so it would be dead configuration) and the coverage
+      # run (fast-unit already runs the full suite; the coverage ratchet stays on
+      # the main rail via --allow-missing in lint-guard).
+      - run: npm run check:cycles
+      - run: npm run check:lockfile
+      - name: Duplication ratchet
+        run: npm run check:duplication
+      - name: Dead-code ratchet (knip)
+        run: npm run check:dead-code
+      - name: Type coverage ratchet
+        run: npm run check:type-coverage
+      - name: Compression budget ratchet
+        run: npm run check:compression-budget
       # Security scanners — same hardened install as ci.yml quality-extended
       # (gh release download = authenticated, 5000 req/hr; curl to api.github.com
       # is rate-limited to 60/hr and silently no-ops when throttled). The blocking
@@ -194,63 +251,30 @@ jobs:
           "$HOME/.local/bin/osv-scanner" --version || true
           "$HOME/.local/bin/oasdiff" --version || true
           zizmor --version || true
-      # Quality gates (all, non-fail-fast) — #8542: replaces 17 bare check:* steps,
-      # 6 G0 gates, 4 ratchet gates, and 3 typecheck steps with a single aggregation
-      # step. Each gate runs in a loop with ::group::; failures are collected and
-      # reported at the end. set -uo pipefail (NOT set -e) so one failing gate does
-      # not abort the job and mask every later gate. Release-added gates are folded
-      # in: open-sse typecheck (#8781) and file-size base-relative mode (#8522).
-      - name: Quality gates (all, non-fail-fast)
+      - name: Secret scan (gitleaks, ratchet, blocking)
+        run: npm run check:secrets -- --ratchet
+      - name: Vulnerability ratchet (osv-scanner, ratchet, blocking)
+        run: npm run check:vuln-ratchet -- --ratchet
+      - name: Workflow lint (actionlint+zizmor, ratchet, blocking)
+        run: npm run check:workflows -- --ratchet
+      # BASE_REF is read by the script from the env (never interpolated into a
+      # shell body) — workflow-injection-safe. actions/checkout fetches remote
+      # refs, not a local branch named github.base_ref, so prefix origin/ or this
+      # gate self-skips every PR with reason=base-unresolved.
+      - name: OpenAPI breaking-change (oasdiff, ratchet, blocking)
         env:
-          # #8522: base-relative file-size mode on PR events — inherited drift (base
-          # already over frozen cap) must not red an innocent PR. Unset on
-          # workflow_dispatch (no PR base) → absolute comparison.
-          PR_BASE_SHA: ${{ github.event.pull_request.base.sha }}
           BASE_REF: ${{ github.base_ref && format('origin/{0}', github.base_ref) || '' }}
-        run: |
-          set -uo pipefail
-          gates=(
-            provider-consistency fetch-targets deps file-size error-helper
-            migration-numbering public-creds db-rules known-symbols
-            route-guard-membership test-discovery test-runner-api
-            mutation-test-coverage any-budget:t11 build-scope pack-policy
-            complexity-ratchets
-            cycles lockfile duplication dead-code type-coverage compression-budget
-            # #8781: open-sse workspace typecheck gate — the workspace imports @/ which
-            # escapes to src/ via undeclared path aliases. See check-open-sse-typecheck.mjs.
-            open-sse-typecheck
-          )
-          ratchet_gates=(
-            secrets vuln-ratchet workflows openapi-breaking
-          )
-          failed=()
-          for g in "${gates[@]}"; do
-            echo "::group::check:$g"
-            # #8522: file-size is base-relative on PR events (compare against
-            # max(frozen, base)) so inherited drift doesn't red an innocent PR;
-            # workflow_dispatch (no PR base) falls back to absolute comparison.
-            if [ "$g" = "file-size" ] && [ -n "${PR_BASE_SHA:-}" ]; then
-              npm run "check:$g" -- --base-ref "$PR_BASE_SHA" || failed+=("$g")
-            else
-              npm run "check:$g" || failed+=("$g")
-            fi
-            echo "::endgroup::"
-          done
-          for g in "${ratchet_gates[@]}"; do
-            echo "::group::check:$g (ratchet)"
-            npm run "check:$g" -- --ratchet || failed+=("$g")
-            echo "::endgroup::"
-          done
-          echo "::group::typecheck:core"
-          npm run typecheck:core || failed+=("typecheck:core")
-          echo "::endgroup::"
-          echo "::group::check:dashboard-typecheck"
-          npm run check:dashboard-typecheck || failed+=("check:dashboard-typecheck")
-          echo "::endgroup::"
-          if (( ${#failed[@]} )); then
-            printf '::error::%d gate(s) failed: %s\n' "${#failed[@]}" "${failed[*]}"
-            exit 1
-          fi
+        run: npm run check:openapi-breaking -- --ratchet
+      - name: Typecheck (core)
+        run: npm run typecheck:core
+      # #7033: dashboard-scoped typecheck gate — src/app/(dashboard) TSX is not
+      # covered by typecheck:core's curated allowlist. See check-dashboard-typecheck.mjs.
+      - name: Typecheck (dashboard)
+        run: npm run check:dashboard-typecheck
+      # #8781: open-sse workspace typecheck gate — the workspace imports @/ which
+      # escapes to src/ via undeclared path aliases. See check-open-sse-typecheck.mjs.
+      - name: Typecheck (open-sse)
+        run: npm run check:open-sse-typecheck
       # WS4.2 (v3.8.49 plan): TypeScript 7 native-compiler SHADOW — advisory only.
       # TS7 went GA 2026-07-08 with 8-12x type-check speedups; its Compiler API only
       # arrives in 7.1, so typescript-eslint / type-coverage / Stryker stay on 6.x

@omniroute/opencode-plugin/src/index.ts

@@ -1033,11 +1033,7 @@ export const OmniRoutePlugin: Plugin = async (_input, options) => {
   // Config hook: keep existing catalog shim, and register slash command
   // templates that ask the agent to call the force-sync tool (OpenCode has no
   // Pi-style registerCommand API; tools + command templates are the native path).
-  const baseConfigHook = createOmniRouteConfigHook(resolved, {
-    cache: sharedCache,
-    diskSnapshotReader: defaultDiskSnapshotReader,
-    diskSnapshotWriter: defaultDiskSnapshotWriter,
-  });
+  const baseConfigHook = createOmniRouteConfigHook(resolved, { cache: sharedCache });
   const configWithSyncCommand = async (input: Config) => {
     await baseConfigHook(input);
     const cfg = input as Config & {
@@ -4407,11 +4403,11 @@ export function buildStaticProviderEntry(
       entry.release_date = raw.release_date;
     }
 
-    // #9175: OC's `getModel` looks the model up by BARE id — the part after
-    // the first `/` in the user's request — so a dict key with an embedded
-    // provider prefix (`<providerId>/<raw-id>`) is unreachable. Keys are the
-    // raw id verbatim; ids that already contain `/` (e.g. `cc/claude-opus-4-7`)
-    // keep it because the slash is part of the upstream model id itself.
+    // OC's static-catalog reader parses each key on `/` and rejects the
+    // entire provider block if ANY key resolves to a parsed providerID that
+    // has no corresponding provider block. So bare keys (no `/`) MUST be
+    // prefixed with the resolved providerId. Already-prefixed keys
+    // (e.g. `cc/claude-opus-4-7`) are left as-is to avoid double-prefixing.
     models[raw.id] = entry;
   }
 
@@ -4745,7 +4741,7 @@ export type OmniRouteDiskSnapshotWriter = (
 export type OmniRouteDiskSnapshotReader = (
   providerId: string,
   identityFingerprint: string
-) => Promise<(Omit<OmniRouteFetchCacheEntry, "expiresAt"> & { writtenAt?: number }) | undefined>;
+) => Promise<Omit<OmniRouteFetchCacheEntry, "expiresAt"> | undefined>;
 
 /**
  * Bind a snapshot to the endpoint and effective credential tuple without
@@ -4828,36 +4824,15 @@ export const defaultDiskSnapshotReader: OmniRouteDiskSnapshotReader = async (
         ? parsed.rawCompressionCombos
         : [],
       rawConnections: Array.isArray(parsed.rawConnections) ? parsed.rawConnections : [],
-      writtenAt: typeof parsed.writtenAt === "number" ? parsed.writtenAt : undefined,
     };
   } catch {
     return undefined;
   }
 };
 
-/** No-op disk-cache pair — used by tests to avoid filesystem side effects.
- * Also used as the default in createOmniRouteConfigHook so that tests
- * that don't pass a diskSnapshotReader don't read real snapshot files
- * from the user's ~/.local/share/opencode/plugins/ directory.
- * The OmniRoutePlugin function passes the real defaultDiskSnapshotReader
- * explicitly. */
-export const noopDiskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
+/** No-op disk-cache pair — used by tests to avoid filesystem side effects. */
 export const noopDiskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
 
-/**
- * In-flight refresh guard: prevents concurrent refreshes for the same
- * cacheKey. When a warm snapshot is served, the refresh runs detached; if
- * a second hook invocation arrives before the refresh completes, it should
- * piggyback on the in-flight promise rather than starting a second one.
- * Cleared on settle so it doesn't leak.
- */
-const _inflightRefresh: Map<string, Promise<void>> = new Map();
-
-/** Reset the in-flight refresh guard (for test isolation). */
-export function _resetInflightRefresh(): void {
-  _inflightRefresh.clear();
-}
-
 // ────────────────────────────────────────────────────────────────────────────
 // Debug logging (features.debugLog)
 // ────────────────────────────────────────────────────────────────────────────
@@ -5092,6 +5067,7 @@ export function createDebugLoggingFetch(
     }
   };
 }
+export const noopDiskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
 
 export type OmniRouteReadAuthJson = () => Promise<AuthJsonShape | undefined | null>;
 
@@ -5194,8 +5170,8 @@ export function createOmniRouteConfigHook(
   const compressionMetaFetcher =
     deps.compressionMetaFetcher ?? defaultOmniRouteCompressionMetaFetcher;
   const providersFetcher = deps.providersFetcher ?? defaultOmniRouteProvidersFetcher;
-  const diskSnapshotReader = deps.diskSnapshotReader ?? noopDiskSnapshotReader;
-  const diskSnapshotWriter = deps.diskSnapshotWriter ?? noopDiskSnapshotWriter;
+  const diskSnapshotReader = deps.diskSnapshotReader ?? defaultDiskSnapshotReader;
+  const diskSnapshotWriter = deps.diskSnapshotWriter ?? defaultDiskSnapshotWriter;
   const now = deps.now ?? Date.now;
   const cache: OmniRouteFetchCache = deps.cache ?? new Map();
   const logger = deps.logger ?? console;
@@ -5290,12 +5266,12 @@ export function createOmniRouteConfigHook(
     const t = now();
     const cached = cache.get(cacheKey);
 
-    let rawModels: OmniRouteRawModelEntry[] = [];
-    let rawCombos: OmniRouteRawCombo[] = [];
-    let rawAutoCombos: OmniRouteRawAutoCombo[] = [];
-    let rawEnrichment: OmniRouteEnrichmentMap = new Map();
-    let rawCompressionCombos: OmniRouteCompressionCombo[] = [];
-    let rawConnections: OmniRouteProviderConnection[] = [];
+    let rawModels: OmniRouteRawModelEntry[];
+    let rawCombos: OmniRouteRawCombo[];
+    let rawAutoCombos: OmniRouteRawAutoCombo[];
+    let rawEnrichment: OmniRouteEnrichmentMap;
+    let rawCompressionCombos: OmniRouteCompressionCombo[];
+    let rawConnections: OmniRouteProviderConnection[];
 
     if (cached && cached.expiresAt > t) {
       rawModels = cached.rawModels;
@@ -5305,275 +5281,160 @@ export function createOmniRouteConfigHook(
       rawCompressionCombos = cached.rawCompressionCombos;
       rawConnections = cached.rawConnections;
     } else {
-      // ─────────────────────────────────────────────────────────────────────
-      // Warm startup: read the disk snapshot before fetching so the provider
-      // registers immediately with the last-known-good catalog. The live
-      // fetch then refreshes in the background (detached) and updates the
-      // cache + snapshot. Gated by features.diskCache (default-on).
-      // ─────────────────────────────────────────────────────────────────────
-      let warmSnapshot: Omit<OmniRouteFetchCacheEntry, "expiresAt"> | undefined;
-      if (wantDiskCache) {
-        const snapshotResult = await diskSnapshotReader(resolved.providerId, snapshotFingerprint);
-        if (snapshotResult && snapshotResult.rawModels.length > 0) {
-          warmSnapshot = snapshotResult;
-          // Log snapshot age (accept any age — instant beats empty).
-          const age = (snapshotResult as { writtenAt?: number }).writtenAt;
-          const ageLabel = typeof age === "number" ? `${Math.round((Date.now() - age) / 3_600_000)}h` : "unknown";
+      // Fail-open fetcher errors: on /v1/models throw, fall back to empty
+      // catalog (still publish a stub block so OC has a complete-shape
+      // entry); on /api/combos throw, publish models-only. Disk-cache
+      // fallback below recovers the last-known-good catalog when the
+      // fetcher threw (network down / 403 / timeout) AND features.diskCache
+      // !== false. A 0-entry SUCCESS (fresh tenant) does NOT trigger
+      // disk fallback — that's a valid empty catalog.
+      let modelsFetchThrew = false;
+      try {
+        rawModels = await fetcher(baseURL, apiKey, 10_000);
+      } catch (err) {
+        logger.warn(
+          "[omniroute-plugin] config shim: /v1/models fetch failed; publishing stub provider entry",
+          err
+        );
+        rawModels = [];
+        modelsFetchThrew = true;
+      }
+      const modelsFetchOk = !modelsFetchThrew && rawModels.length > 0;
+
+      rawCombos = [];
+      try {
+        rawCombos = await combosFetcher(baseURL, managementReadToken, 10_000);
+      } catch (err) {
+        logger.warn(
+          "[omniroute-plugin] config shim: /api/combos fetch failed; publishing models-only static catalog",
+          err
+        );
+      }
+
+      rawAutoCombos = [];
+      if (wantAutoCombos) {
+        try {
+          rawAutoCombos = await autoCombosFetcher(baseURL, managementReadToken, 5_000);
+        } catch {
+          // Already handled inside the default fetcher
+        }
+      }
+
+      // Eagerly fetch enrichment so the static block can overlay human
+      // display names on raw model ids. On OC ≤1.15.5 the dynamic
+      // `provider.models` hook never fires in `serve` mode, so the static
+      // block IS what reaches `/provider` and the TUI model picker.
+      // Gated by `features.enrichment` (default-on). Soft-fail on error —
+      // we still publish a name-less catalog if /api/pricing/models is
+      // unreachable.
+      rawEnrichment = new Map();
+      if (wantEnrichment) {
+        try {
+          rawEnrichment = await enrichmentFetcher(baseURL, managementReadToken, 10_000);
+        } catch (err) {
           logger.warn(
-            `[omniroute-plugin] config shim: warm startup from disk snapshot (${snapshotResult.rawModels.length} models, age ${ageLabel})`
+            "[omniroute-plugin] config shim: /api/pricing/models fetch failed; publishing raw-id static catalog",
+            err
           );
         }
       }
 
-      // ─────────────────────────────────────────────────────────────────────
-      // Parallel refresh: all six fetchers run concurrently via
-      // Promise.allSettled. Each wrapper never rejects (catches internally)
-      // so partial failure is tolerated — same soft-fail semantics as the
-      // old sequential chain, but ~6x faster.
-      // ─────────────────────────────────────────────────────────────────────
-      const doRefresh = async (): Promise<void> => {
-        let modelsFetchThrew = false;
-        let localRawModels: OmniRouteRawModelEntry[] = [];
-        let localRawCombos: OmniRouteRawCombo[] = [];
-        let localRawAutoCombos: OmniRouteRawAutoCombo[] = [];
-        let localRawEnrichment: OmniRouteEnrichmentMap = new Map();
-        let localRawCompressionCombos: OmniRouteCompressionCombo[] = [];
-        let localRawConnections: OmniRouteProviderConnection[] = [];
-
-        // Each wrapper keeps the existing try/catch, default value, and
-        // exact warn message so per-endpoint fallbacks are preserved.
-        const doModels = async (): Promise<void> => {
-          try {
-            localRawModels = await fetcher(baseURL, apiKey, 10_000);
-          } catch (err) {
-            logger.warn(
-              "[omniroute-plugin] config shim: /v1/models fetch failed; publishing stub provider entry",
-              err
-            );
-            localRawModels = [];
-            modelsFetchThrew = true;
-          }
-        };
-
-        const doCombos = async (): Promise<void> => {
-          try {
-            localRawCombos = await combosFetcher(baseURL, managementReadToken, 10_000);
-          } catch (err) {
-            logger.warn(
-              "[omniroute-plugin] config shim: /api/combos fetch failed; publishing models-only static catalog",
-              err
-            );
-          }
-        };
-
-        const doAutoCombos = async (): Promise<void> => {
-          if (!wantAutoCombos) return;
-          try {
-            localRawAutoCombos = await autoCombosFetcher(baseURL, managementReadToken, 5_000);
-          } catch {
-            // Already handled inside the default fetcher
-          }
-        };
-
-        const doEnrichment = async (): Promise<void> => {
-          if (!wantEnrichment) return;
-          try {
-            localRawEnrichment = await enrichmentFetcher(baseURL, managementReadToken, 10_000);
-          } catch (err) {
-            logger.warn(
-              "[omniroute-plugin] config shim: /api/pricing/models fetch failed; publishing raw-id static catalog",
-              err
-            );
-          }
-        };
-
-        const doCompression = async (): Promise<void> => {
-          if (!wantCompressionMeta) return;
-          try {
-            localRawCompressionCombos = await compressionMetaFetcher(baseURL, managementReadToken, 10_000);
-          } catch (err) {
-            logger.warn(
-              "[omniroute-plugin] config shim: /api/context/combos fetch failed; publishing combos without compression suffix",
-              err
-            );
-          }
-        };
-
-        const doConnections = async (): Promise<void> => {
-          if (!wantUsableOnly) return;
-          try {
-            localRawConnections = await providersFetcher(baseURL, managementReadToken, 10_000);
-          } catch (err) {
-            logger.warn(
-              "[omniroute-plugin] config shim: /api/providers fetch failed; usableOnly filter disabled for this refresh",
-              err
-            );
-          }
-        };
-
-        await Promise.allSettled([
-          doModels(),
-          doCombos(),
-          doAutoCombos(),
-          doEnrichment(),
-          doCompression(),
-          doConnections(),
-        ]);
-
-        const modelsFetchOk = !modelsFetchThrew && localRawModels.length > 0;
-
-        // Disk-cache fallback (cold first run, no warm snapshot): when the
-        // live fetch returned no models AND features.diskCache !== false,
-        // hydrate from the last-known-good snapshot so OC still surfaces a
-        // usable catalog (e.g. IP whitelist drop, offline laptop).
-        if (modelsFetchThrew && wantDiskCache && !warmSnapshot) {
-          const snapshot = await diskSnapshotReader(resolved.providerId, snapshotFingerprint);
-          if (snapshot && snapshot.rawModels.length > 0) {
-            logger.warn(
-              `[omniroute-plugin] config shim: /v1/models unreachable; using stale disk cache (${snapshot.rawModels.length} models)`
-            );
-            localRawModels = snapshot.rawModels;
-            localRawCombos = snapshot.rawCombos;
-            localRawAutoCombos = snapshot.rawAutoCombos ?? [];
-            localRawEnrichment = snapshot.rawEnrichment;
-            localRawCompressionCombos = snapshot.rawCompressionCombos;
-            localRawConnections = snapshot.rawConnections;
-          }
+      // Compression-metadata fetch — opt-in via features.compressionMetadata.
+      // When on, the default pipeline is appended to every combo `name` so
+      // the TUI picker advertises which compression a combo applies.
+      rawCompressionCombos = [];
+      if (wantCompressionMeta) {
+        try {
+          rawCompressionCombos = await compressionMetaFetcher(baseURL, managementReadToken, 10_000);
+        } catch (err) {
+          logger.warn(
+            "[omniroute-plugin] config shim: /api/context/combos fetch failed; publishing combos without compression suffix",
+            err
+          );
         }
+      }
 
-        // Cache even partial results — a subsequent provider-hook call should
-        // not re-burn the timeout window on the same broken endpoint.
-        cache.set(cacheKey, {
-          rawModels: localRawModels,
-          rawCombos: localRawCombos,
-          rawAutoCombos: localRawAutoCombos,
-          rawEnrichment: localRawEnrichment,
-          rawCompressionCombos: localRawCompressionCombos,
-          rawConnections: localRawConnections,
-          expiresAt: now() + resolved.modelCacheTtl,
+      // Provider-connections fetch — opt-in via features.usableOnly. When
+      // on, the static catalog filters out models/combos whose canonical
+      // provider has no active connection. Soft-fail (empty list) disables
+      // the filter for this refresh, never hiding the whole catalog.
+      rawConnections = [];
+      if (wantUsableOnly) {
+        try {
+          rawConnections = await providersFetcher(baseURL, managementReadToken, 10_000);
+        } catch (err) {
+          logger.warn(
+            "[omniroute-plugin] config shim: /api/providers fetch failed; usableOnly filter disabled for this refresh",
+            err
+          );
+        }
+      }
+
+      // Disk-cache fallback: when the live fetch returned no models AND
+      // features.diskCache !== false, hydrate from the last-known-good
+      // snapshot so OC still surfaces a usable catalog (e.g. IP whitelist
+      // drop, offline laptop). The snapshot is whatever we last wrote on
+      // a healthy refresh; staleness is bounded only by how recently the
+      // user was online.
+      if (modelsFetchThrew && wantDiskCache) {
+        const snapshot = await diskSnapshotReader(resolved.providerId, snapshotFingerprint);
+        if (snapshot && snapshot.rawModels.length > 0) {
+          logger.warn(
+            `[omniroute-plugin] config shim: /v1/models unreachable; using stale disk cache (${snapshot.rawModels.length} models)`
+          );
+          rawModels = snapshot.rawModels;
+          rawCombos = snapshot.rawCombos;
+          rawAutoCombos = snapshot.rawAutoCombos ?? [];
+          rawEnrichment = snapshot.rawEnrichment;
+          rawCompressionCombos = snapshot.rawCompressionCombos;
+          rawConnections = snapshot.rawConnections;
+        }
+      }
+
+      // Cache even partial results — a subsequent provider-hook call should
+      // not re-burn the timeout window on the same broken endpoint.
+      cache.set(cacheKey, {
+        rawModels,
+        rawCombos,
+        rawAutoCombos,
+        rawEnrichment,
+        rawCompressionCombos,
+        rawConnections,
+        expiresAt: t + resolved.modelCacheTtl,
+      });
+
+      // Startup diagnostics (file-based) — fires at startup via config hook
+      if (resolved.features?.startupDebug === true) {
+        await writeStartupDiagnostics({
+          providerId: resolved.providerId,
+          baseURL,
+          modelCount: rawModels.length,
+          comboCount: rawCombos.length,
+          enrichmentSize: rawEnrichment.size,
+          autoComboCount: rawAutoCombos.length,
+          enrichment: rawEnrichment,
+          autoCombos: rawAutoCombos,
+          features: resolved.features,
         });
+      }
 
-        // Startup diagnostics (file-based) — fires at startup via config hook
-        if (resolved.features?.startupDebug === true) {
-          await writeStartupDiagnostics({
-            providerId: resolved.providerId,
-            baseURL,
-            modelCount: localRawModels.length,
-            comboCount: localRawCombos.length,
-            enrichmentSize: localRawEnrichment.size,
-            autoComboCount: localRawAutoCombos.length,
-            enrichment: localRawEnrichment,
-            autoCombos: localRawAutoCombos,
-            features: resolved.features,
-          });
-        }
-
-        // Disk-cache write: persist the last successful (or any non-empty)
-        // catalog so a subsequent cold start with a failed fetch can recover.
-        // Best-effort; soft-fail keeps us moving when the data dir isn't
-        // writable (e.g. read-only container). A failed refresh never
-        // overwrites the snapshot (modelsFetchOk gate).
-        if (modelsFetchOk && wantDiskCache) {
-          await diskSnapshotWriter(
-            resolved.providerId,
-            {
-              rawModels: localRawModels,
-              rawCombos: localRawCombos,
-              rawAutoCombos: localRawAutoCombos,
-              rawEnrichment: localRawEnrichment,
-              rawCompressionCombos: localRawCompressionCombos,
-              rawConnections: localRawConnections,
-            },
-            snapshotFingerprint
-          );
-        }
-
-        // Re-publish a fresh block via the shared cache so OC >=1.14.49's
-        // dynamic provider hook picks it up from the cache. When the models
-        // fetch threw and a warm snapshot was served, keep the warm block
-        // (no downgrade to stub).
-        if (modelsFetchOk || !warmSnapshot) {
-          const freshBlock = buildStaticProviderEntry(
-            localRawModels,
-            localRawCombos,
-            resolved,
-            baseURL,
-            apiKey,
-            localRawEnrichment,
-            localRawCompressionCombos,
-            localRawConnections,
-            localRawAutoCombos
-          );
-          const inputWithProvider2 = input as { provider?: Record<string, unknown> };
-          if (inputWithProvider2.provider) {
-            inputWithProvider2.provider[resolved.providerId] = freshBlock;
-          }
-        }
-      };
-
-      if (warmSnapshot) {
-        // Warm startup: publish the snapshot block immediately, then run
-        // the refresh detached (never a floating unhandled rejection).
-        rawModels = warmSnapshot.rawModels;
-        rawCombos = warmSnapshot.rawCombos;
-        rawAutoCombos = warmSnapshot.rawAutoCombos ?? [];
-        rawEnrichment = warmSnapshot.rawEnrichment;
-        rawCompressionCombos = warmSnapshot.rawCompressionCombos;
-        rawConnections = warmSnapshot.rawConnections;
-
-        // In-flight guard: if a refresh is already running for this
-        // cacheKey, piggyback on it instead of starting a second one.
-        const existing = _inflightRefresh.get(cacheKey);
-        if (existing) {
-          // Another refresh is in-flight — don't start a second one.
-          // The existing refresh will update the cache when it completes.
-        } else {
-          const refreshP = doRefresh()
-            .catch((err: unknown) => {
-              logger.warn("[omniroute-plugin] config shim: background refresh failed", err);
-            })
-            .finally(() => {
-              _inflightRefresh.delete(cacheKey);
-            });
-          _inflightRefresh.set(cacheKey, refreshP);
-        }
-      } else {
-        // Cold first run (no warm snapshot): await the refresh so the
-        // first publish is always correct. In-flight guard still applies.
-        const existing = _inflightRefresh.get(cacheKey);
-        if (existing) {
-          await existing;
-          // After the in-flight refresh completes, the cache has the data.
-          const fresh = cache.get(cacheKey);
-          if (fresh) {
-            rawModels = fresh.rawModels;
-            rawCombos = fresh.rawCombos;
-            rawAutoCombos = fresh.rawAutoCombos;
-            rawEnrichment = fresh.rawEnrichment;
-            rawCompressionCombos = fresh.rawCompressionCombos;
-            rawConnections = fresh.rawConnections;
-          }
-        } else {
-          const refreshP = doRefresh()
-            .catch((err: unknown) => {
-              logger.warn("[omniroute-plugin] config shim: refresh failed", err);
-            })
-            .finally(() => {
-              _inflightRefresh.delete(cacheKey);
-            });
-          _inflightRefresh.set(cacheKey, refreshP);
-          await refreshP;
-          // After the refresh, the cache has the data.
-          const fresh = cache.get(cacheKey);
-          if (fresh) {
-            rawModels = fresh.rawModels;
-            rawCombos = fresh.rawCombos;
-            rawAutoCombos = fresh.rawAutoCombos;
-            rawEnrichment = fresh.rawEnrichment;
-            rawCompressionCombos = fresh.rawCompressionCombos;
-            rawConnections = fresh.rawConnections;
-          }
-        }
+      // Disk-cache write: persist the last successful (or any non-empty)
+      // catalog so a subsequent cold start with a failed fetch can recover.
+      // Best-effort; soft-fail keeps us moving when the data dir isn't
+      // writable (e.g. read-only container).
+      if (modelsFetchOk && wantDiskCache) {
+        await diskSnapshotWriter(
+          resolved.providerId,
+          {
+            rawModels,
+            rawCombos,
+            rawAutoCombos,
+            rawEnrichment,
+            rawCompressionCombos,
+            rawConnections,
+          },
+          snapshotFingerprint
+        );
       }
     }
 

@omniroute/opencode-plugin/tests/config-shim.test.ts

@@ -33,7 +33,6 @@ import {
   createOmniRouteProviderHook,
   OmniRoutePlugin,
   resolveOmniRoutePluginOptions,
-  _resetInflightRefresh,
   type OmniRouteCombosFetcher,
   type OmniRouteEnrichmentEntry,
   type OmniRouteEnrichmentFetcher,
@@ -48,16 +47,6 @@ import {
   type OmniRouteStaticProviderEntry,
 } from "../src/index.js";
 
-// ────────────────────────────────────────────────────────────────────────────
-// Test isolation: reset the module-level in-flight refresh guard between
-// tests so a detached refresh from a previous test doesn't leak into the
-// next one.
-// ────────────────────────────────────────────────────────────────────────────
-
-test.beforeEach(() => {
-  _resetInflightRefresh();
-});
-
 // ────────────────────────────────────────────────────────────────────────────
 // Fixtures
 // ────────────────────────────────────────────────────────────────────────────
@@ -238,7 +227,7 @@ test("config: with valid auth.json + apiKey + baseURL → mutates input.provider
   // Stripped per-model shape: name + cap flags + modalities + (optional)
   // cost. OC's SDK static schema accepts only `limit.{context,output}` —
   // `limit.input` is NOT in the SDK shape and gets dropped silently.
-  const claude = entry.models["claude-sonnet-4-6"];
+  const claude = entry.models["opencode-omniroute/claude-sonnet-4-6"];
   assert.ok(claude, "claude model surfaced");
   assert.equal(claude.name, "claude-sonnet-4-6");
   assert.equal(claude.attachment, true);
@@ -259,7 +248,7 @@ test("config: with valid auth.json + apiKey + baseURL → mutates input.provider
 
   // Combo surfaces under bare key + LCD'd
   // (gemini's reasoning=false → combo reasoning=false).
-  const combo = entry.models["claude-tier"];
+  const combo = entry.models["omniroute/claude-tier"];
   assert.ok(combo, "combo surfaced under bare key");
   assert.equal(combo.name, "Claude Tier");
   assert.equal(combo.reasoning, false, "LCD: any member reasoning=false → combo reasoning=false");
@@ -482,10 +471,10 @@ test("config: combos fetcher throws → emit models-only catalog (no combos in m
   assert.ok(entry);
   const ids = Object.keys(entry.models).sort();
   assert.deepEqual(ids, [
-    "claude-sonnet-4-6",
-    "gemini-3-flash",
+    "opencode-omniroute/claude-sonnet-4-6",
+    "opencode-omniroute/gemini-3-flash",
   ]);
-  assert.equal(entry.models["claude-tier"], undefined, "no combo entry");
+  assert.equal(entry.models["omniroute/claude-tier"], undefined, "no combo entry");
   assert.ok(
     logger.entries.some((e) => String(e[0]).includes("/api/combos fetch failed")),
     "combos-fetch breadcrumb emitted"
@@ -734,7 +723,7 @@ test("buildStaticProviderEntry: stripped per-model shape matches sibling @omniro
   }
 
   // Sanity: claude entry has all expected stripped fields.
-  const claude = block.models["claude-sonnet-4-6"];
+  const claude = block.models["opencode-omniroute/claude-sonnet-4-6"];
   assert.equal(typeof claude.name, "string");
   assert.equal(typeof claude.attachment, "boolean");
   assert.equal(typeof claude.reasoning, "boolean");
@@ -759,8 +748,8 @@ test("buildStaticProviderEntry: hidden combos are excluded", () => {
     "https://or.example/v1",
     "sk-test"
   );
-  assert.equal(block.models["claude-tier"], undefined);
-  assert.ok(block.models["claude-sonnet-4-6"]);
+  assert.equal(block.models["omniroute/claude-tier"], undefined);
+  assert.ok(block.models["opencode-omniroute/claude-sonnet-4-6"]);
 });
 
 // ────────────────────────────────────────────────────────────────────────────
@@ -776,7 +765,7 @@ test("buildStaticProviderEntry: emits modalities.input from raw.input_modalities
     "https://or.example/v1",
     "sk-test"
   );
-  const claude = block.models["claude-sonnet-4-6"];
+  const claude = block.models["opencode-omniroute/claude-sonnet-4-6"];
   assert.deepEqual(claude.modalities?.input, ["text", "image"]);
   assert.deepEqual(claude.modalities?.output, ["text"]);
 });
@@ -790,7 +779,7 @@ test("buildStaticProviderEntry: never emits limit.input (OC SDK rejects it)", ()
     "https://or.example/v1",
     "sk-test"
   );
-  const claude = block.models["claude-sonnet-4-6"];
+  const claude = block.models["opencode-omniroute/claude-sonnet-4-6"];
   assert.equal((claude.limit as Record<string, unknown>).input, undefined);
   assert.equal(typeof claude.limit?.context, "number");
   assert.equal(typeof claude.limit?.output, "number");
@@ -818,7 +807,7 @@ test("buildStaticProviderEntry: emits cost when enrichment carries pricing", ()
     "sk-test",
     enrichment
   );
-  const claude = block.models["claude-sonnet-4-6"];
+  const claude = block.models["opencode-omniroute/claude-sonnet-4-6"];
   assert.equal(claude.cost?.input, 3);
   assert.equal(claude.cost?.output, 15);
   assert.equal(claude.cost?.cache_read, 0.3);
@@ -839,8 +828,8 @@ test("buildStaticProviderEntry: emits release_date when raw carries it; omits wh
     "https://or.example/v1",
     "sk-test"
   );
-  assert.equal(block.models["claude-with-date"].release_date, "2026-02-19");
-  assert.equal(block.models["gemini-3-flash"].release_date, undefined);
+  assert.equal(block.models["opencode-omniroute/claude-with-date"].release_date, "2026-02-19");
+  assert.equal(block.models["opencode-omniroute/gemini-3-flash"].release_date, undefined);
 });
 
 test("buildStaticProviderEntry: combo modalities = intersection of members (LCD)", () => {
@@ -869,7 +858,7 @@ test("buildStaticProviderEntry: combo modalities = intersection of members (LCD)
     "https://or.example/v1",
     "sk-test"
   );
-  const combo = block.models["mixed-tier"];
+  const combo = block.models["omniroute/mixed-tier"];
   assert.ok(combo, "combo emitted under slug key");
   // claude has text+image, text-only has text → intersection drops image.
   assert.deepEqual(combo.modalities?.input, ["text"]);
@@ -978,10 +967,10 @@ test("config: enrichment fetched + name overlaid on raw-model entries", async ()
     "opencode-omniroute"
   ];
   assert.ok(entry);
-  assert.equal(entry.models["claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
-  assert.equal(entry.models["gemini-3-flash"].name, "Gemini 3 Flash");
+  assert.equal(entry.models["opencode-omniroute/claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
+  assert.equal(entry.models["opencode-omniroute/gemini-3-flash"].name, "Gemini 3 Flash");
   // Combo names still come from /api/combos — enrichment overlay does NOT touch combos.
-  assert.equal(entry.models["claude-tier"].name, "Claude Tier");
+  assert.equal(entry.models["omniroute/claude-tier"].name, "Claude Tier");
   assert.equal(enrichmentFetcher.callCount(), 1);
 });
 
@@ -1011,7 +1000,7 @@ test("config: features.enrichment=false skips enrichment fetch + keeps raw-id na
   assert.ok(entry);
   assert.equal(enrichmentFetcher.callCount(), 0, "enrichment fetch suppressed by feature flag");
   assert.equal(
-    entry.models["claude-sonnet-4-6"].name,
+    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "claude-sonnet-4-6",
     "raw id retained"
   );
@@ -1038,7 +1027,7 @@ test("config: enrichment fetcher throws → soft-fail (warn + raw-id static cata
   ];
   assert.ok(entry, "static block still published on enrichment failure");
   assert.equal(
-    entry.models["claude-sonnet-4-6"].name,
+    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "claude-sonnet-4-6",
     "raw id retained"
   );
@@ -1240,20 +1229,17 @@ test("config: diskCache hydrates stale snapshot when /v1/models throws", async (
     "opencode-omniroute"
   ];
   assert.ok(
-    entry.models["claude-sonnet-4-6"],
+    entry.models["opencode-omniroute/claude-sonnet-4-6"],
     "stale snapshot hydrated into static block"
   );
   assert.equal(
-    entry.models["claude-sonnet-4-6"].name,
+    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude Sonnet 4.6 (cached)",
     "stale enrichment also reused"
   );
   assert.equal(writes, 0, "disk write skipped when live fetch failed");
   assert.ok(
-    logger.entries.some((e) =>
-      String(e[0]).includes("using stale disk cache") ||
-      String(e[0]).includes("warm startup from disk snapshot")
-    ),
+    logger.entries.some((e) => String(e[0]).includes("using stale disk cache")),
     "disk-cache hydration breadcrumb emitted"
   );
 });
@@ -1295,7 +1281,7 @@ test("config: cached rawEnrichment from earlier provider hook is reused (no refe
   const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
     "opencode-omniroute"
   ];
-  assert.equal(entry.models["claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
+  assert.equal(entry.models["opencode-omniroute/claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
 });
 
 // ─────────────────────────────────────────────────────────────────────
@@ -1346,12 +1332,12 @@ test("config: providerTag (default-on) prepends '<provider> - ' to enriched raw-
   ];
   assert.ok(entry);
   assert.equal(
-    entry.models["claude-sonnet-4-6"].name,
+    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude - Claude Sonnet 4.6"
   );
-  assert.equal(entry.models["gemini-3-flash"].name, "Gemini - Gemini 3 Flash");
+  assert.equal(entry.models["opencode-omniroute/gemini-3-flash"].name, "Gemini - Gemini 3 Flash");
   // Combos stay untouched — `Combo: ` prefix already conveys multi-upstream.
-  assert.equal(entry.models["claude-tier"].name, "Claude Tier");
+  assert.equal(entry.models["omniroute/claude-tier"].name, "Claude Tier");
 });
 
 test("config: providerTag=false suppresses the suffix", async () => {
@@ -1378,7 +1364,7 @@ test("config: providerTag=false suppresses the suffix", async () => {
     "opencode-omniroute"
   ];
   assert.equal(
-    entry.models["claude-sonnet-4-6"].name,
+    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude Sonnet 4.6",
     "enriched name kept, provider tag suppressed"
   );
@@ -1410,7 +1396,7 @@ test("config: providerTag falls back to UPPER(alias) when providerDisplayName mi
   const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
     "opencode-omniroute"
   ];
-  assert.equal(entry.models["claude-sonnet-4-6"].name, "CC - Claude Sonnet 4.6");
+  assert.equal(entry.models["opencode-omniroute/claude-sonnet-4-6"].name, "CC - Claude Sonnet 4.6");
 });
 
 test("config: providerTag skipped entirely when neither providerDisplayName nor providerAlias set", async () => {
@@ -1437,7 +1423,7 @@ test("config: providerTag skipped entirely when neither providerDisplayName nor
   const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
     "opencode-omniroute"
   ];
-  assert.equal(entry.models["claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
+  assert.equal(entry.models["opencode-omniroute/claude-sonnet-4-6"].name, "Claude Sonnet 4.6");
 });
 
 test("config: providerTag is idempotent — second hook call doesn't double-suffix", async () => {
@@ -1465,7 +1451,7 @@ test("config: providerTag is idempotent — second hook call doesn't double-suff
     "opencode-omniroute"
   ];
   assert.equal(
-    entryA.models["claude-sonnet-4-6"].name,
+    entryA.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude - Claude Sonnet 4.6"
   );
 
@@ -1476,7 +1462,7 @@ test("config: providerTag is idempotent — second hook call doesn't double-suff
     "opencode-omniroute"
   ];
   assert.equal(
-    entryB.models["claude-sonnet-4-6"].name,
+    entryB.models["opencode-omniroute/claude-sonnet-4-6"].name,
     "Claude - Claude Sonnet 4.6"
   );
 });
@@ -1530,7 +1516,7 @@ test("buildStaticProviderEntry: nested combo-ref context is the bottleneck acros
   );
   // Pre-fix: Parent would advertise 200_000 (only raw-big counted).
   // Post-fix: Parent should advertise 8_000 (TinyCombo bottleneck).
-  const parent = block.models["parent"];
+  const parent = block.models["omniroute/parent"];
   assert.ok(parent, "Parent combo must be in the static catalog");
   assert.equal(parent.limit?.context, 8_000);
 });

@omniroute/opencode-plugin/tests/provider-id-routing.test.ts

@@ -111,9 +111,7 @@ test("#6859: createOmniRouteProviderHook end-to-end — catalog keys/providerID
 // `opencode-omniroute`. Confirmed against the issue's own curl repro
 // (`model: "opencode-omniroute/hermes-smart-stack"` → "No active
 // credentials for provider: opencode-omniroute").
-// #9175 tightened this further: OC's `getModel` looks models up by BARE id,
-// so combo dict keys now carry NO prefix at all (not even `omniroute/`).
-test("#7976/#9175: buildStaticProviderEntry keys combos by bare slug (no prefix at all — never the OC-gate providerId)", () => {
+test("#7976: buildStaticProviderEntry keys bare-slug combo ids with the unprefixed omnirouteProviderId (no double OC-gate prefix)", () => {
   const resolved = resolveOmniRoutePluginOptions({ providerId: "omniroute" });
   assert.equal(resolved.providerId, "opencode-omniroute");
   assert.equal(resolved.omnirouteProviderId, "omniroute");
@@ -133,7 +131,7 @@ test("#7976/#9175: buildStaticProviderEntry keys combos by bare slug (no prefix
     "sk-test"
   );
 
-  assert.deepEqual(Object.keys(block.models), ["hermes-smart-stack"]);
+  assert.deepEqual(Object.keys(block.models), ["omniroute/hermes-smart-stack"]);
   assert.equal(
     block.models["opencode-omniroute/hermes-smart-stack"],
     undefined,

@omniroute/opencode-plugin/tests/warm-startup.test.ts

@@ -1,827 +0,0 @@
-/**
- * Warm-startup + parallel-refresh tests for the opencode-plugin config shim.
- *
- * Covers `createOmniRouteConfigHook(opts, deps)`:
- *   - (a) Warm startup: cache miss + matching snapshot → provider block
- *     populated from snapshot data (not live fetch data).
- *   - (b) Fingerprint mismatch: reader returns undefined → no warm publish,
- *     falls through to awaited fetch (cold-start behavior).
- *   - (c) Successful parallel refresh: all fetchers resolve → cache updated,
- *     disk snapshot written.
- *   - (d) Failed refresh keeps the snapshot: warm-served + models fetcher
- *     rejects → no disk overwrite, block stays at warm-snapshot shape.
- *   - (e) Parallelism: all six fetchers start concurrently (not sequential).
- *   - (f) Soft-fail parity under Promise.allSettled: per-endpoint
- *     fallbacks + logger.warn breadcrumbs preserved.
- *   - (g) No double-refresh: concurrent hook invocations on the same cacheKey
- *     trigger only one refresh (in-flight guard).
- *   - (h) features.diskCache: false disables the warm read entirely.
- *
- * Mocking strategy: every dependency is DI-injected at hook construction
- * (same pattern as config-shim.test.ts). No global monkey-patching.
- */
-
-import test from "node:test";
-import assert from "node:assert/strict";
-import type { Config } from "@opencode-ai/plugin";
-
-import {
-  createOmniRouteConfigHook,
-  resolveOmniRoutePluginOptions,
-  _resetInflightRefresh,
-  type OmniRouteAutoCombosFetcher,
-  type OmniRouteCombosFetcher,
-  type OmniRouteCompressionMetaFetcher,
-  type OmniRouteEnrichmentEntry,
-  type OmniRouteEnrichmentFetcher,
-  type OmniRouteEnrichmentMap,
-  type OmniRouteFetchCache,
-  type OmniRouteModelsFetcher,
-  type OmniRouteProviderConnection,
-  type OmniRouteProvidersFetcher,
-  type OmniRouteRawAutoCombo,
-  type OmniRouteRawCombo,
-  type OmniRouteRawModelEntry,
-  type OmniRouteReadAuthJson,
-  type OmniRouteStaticProviderEntry,
-  type OmniRouteDiskSnapshotReader,
-  type OmniRouteDiskSnapshotWriter,
-  type OmniRouteCompressionCombo,
-} from "../src/index.js";
-
-// ────────────────────────────────────────────────────────────────────────────
-// Test isolation: reset the module-level in-flight refresh guard between
-// tests so a detached refresh from a previous test doesn't leak into the
-// next one (same cacheKey, different cache instance).
-// ────────────────────────────────────────────────────────────────────────────
-
-test.beforeEach(() => {
-  _resetInflightRefresh();
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// Fixtures
-// ────────────────────────────────────────────────────────────────────────────
-
-const MODEL_CLAUDE: OmniRouteRawModelEntry = {
-  id: "claude-sonnet-4-6",
-  capabilities: {
-    tool_calling: true,
-    reasoning: true,
-    vision: true,
-    thinking: false,
-    temperature: true,
-  },
-  context_length: 200_000,
-  max_output_tokens: 64_000,
-  max_input_tokens: 180_000,
-  input_modalities: ["text", "image"],
-  output_modalities: ["text"],
-};
-
-const MODEL_GEMINI: OmniRouteRawModelEntry = {
-  id: "gemini-3-flash",
-  capabilities: { tool_calling: true, reasoning: false, vision: true, thinking: false },
-  context_length: 1_000_000,
-  max_output_tokens: 8_192,
-  input_modalities: ["text", "image"],
-  output_modalities: ["text"],
-};
-
-const COMBO_CLAUDE_TIER: OmniRouteRawCombo = {
-  id: "combo-claude-tier",
-  name: "Claude Tier",
-  models: [
-    { id: "s1", kind: "model", model: "claude-sonnet-4-6", weight: 100 },
-    { id: "s2", kind: "model", model: "gemini-3-flash", weight: 50 },
-  ],
-};
-
-const AUTO_COMBO: OmniRouteRawAutoCombo = {
-  id: "auto",
-  name: "Auto",
-};
-
-const COMPRESSION_COMBO: OmniRouteCompressionCombo = {
-  id: "ctx-combo-1",
-  name: "Context Combo",
-  pipeline: "gzip",
-};
-
-const CONNECTION_CLAUDE: OmniRouteProviderConnection = {
-  id: "c1",
-  provider: "claude",
-  isActive: true,
-  testStatus: "active",
-};
-
-// ────────────────────────────────────────────────────────────────────────────
-// DI stub helpers
-// ────────────────────────────────────────────────────────────────────────────
-
-function stubReadAuthJson(
-  value: Record<string, unknown> | undefined | null
-): OmniRouteReadAuthJson {
-  return async () => value as never;
-}
-
-function immediateFetcher<T extends (...args: unknown[]) => Promise<unknown>>(
-  payload: ReturnType<T> extends Promise<infer U> ? U : never
-): T & { callCount: () => number; startedAt: () => number | undefined } {
-  let n = 0;
-  let start: number | undefined;
-  const f = async (..._args: unknown[]) => {
-    start = Date.now();
-    n++;
-    return payload;
-  };
-  return Object.assign(f as T, { callCount: () => n, startedAt: () => start });
-}
-
-function throwingFetcher<T extends (...args: unknown[]) => Promise<unknown>>(
-  msg = "ECONNREFUSED"
-): T & { callCount: () => number } {
-  let n = 0;
-  const f = async (..._args: unknown[]) => {
-    n++;
-    throw new Error(msg);
-  };
-  return Object.assign(f as T, { callCount: () => n });
-}
-
-interface WarnCapture {
-  warn: (...args: unknown[]) => void;
-  entries: unknown[][];
-}
-
-function captureWarn(): WarnCapture {
-  const entries: unknown[][] = [];
-  return {
-    warn: (...args: unknown[]) => {
-      entries.push(args);
-    },
-    entries,
-  };
-}
-
-function makeInput(initialProvider: Record<string, unknown> = {}): Config {
-  return { provider: initialProvider } as unknown as Config;
-}
-
-/** Build a valid auth.json stub for the default providerId. */
-function authStub() {
-  return stubReadAuthJson({
-    "opencode-omniroute": {
-      type: "api",
-      key: "sk-test",
-      baseURL: "https://or.example.com/v1",
-    },
-  });
-}
-
-// ────────────────────────────────────────────────────────────────────────────
-// (a) Warm startup: cache miss + matching snapshot → provider block populated
-//     from snapshot data (not live fetch data)
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: snapshot data used when snapshot is present", async () => {
-  // Live fetch returns MODEL_CLAUDE, but snapshot has MODEL_GEMINI.
-  // With warm startup, the block should contain the snapshot data.
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const autoCombosFetcher = immediateFetcher<OmniRouteAutoCombosFetcher>([]);
-  const enrichmentFetcher = immediateFetcher<OmniRouteEnrichmentFetcher>(new Map());
-  const compressionMetaFetcher = immediateFetcher<OmniRouteCompressionMetaFetcher>([]);
-  const providersFetcher = immediateFetcher<OmniRouteProvidersFetcher>([]);
-  const logger = captureWarn();
-
-  const snapshot: Omit<import("../src/index.js").OmniRouteFetchCacheEntry, "expiresAt"> = {
-    rawModels: [MODEL_GEMINI],
-    rawCombos: [],
-    rawAutoCombos: [],
-    rawEnrichment: new Map(),
-    rawCompressionCombos: [],
-    rawConnections: [],
-  };
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => snapshot;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      autoCombosFetcher,
-      enrichmentFetcher,
-      compressionMetaFetcher,
-      providersFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const provider = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider;
-  const entry = provider["opencode-omniroute"];
-  assert.ok(entry, "provider entry published");
-
-  // With warm startup, the block should contain the snapshot data (GEMINI),
-  // not the live fetch data (CLAUDE). This is the key assertion: the warm
-  // snapshot is served first, and the live refresh updates the cache in the
-  // background. On the next hook invocation, the cache will have the fresh data.
-  const hasGemini = entry.models["opencode-omniroute/gemini-3-flash"] !== undefined;
-  const hasClaude = entry.models["opencode-omniroute/claude-sonnet-4-6"] !== undefined;
-  assert.ok(
-    hasGemini || hasClaude,
-    "provider block has at least one model"
-  );
-
-  // The warm-startup breadcrumb should be emitted.
-  assert.ok(
-    logger.entries.some((e) =>
-      String(e[0]).includes("warm startup from disk snapshot")
-    ),
-    "warm-startup breadcrumb emitted"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (b) Fingerprint mismatch: reader returns undefined → no warm publish,
-//     falls through to awaited fetch
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: fingerprint mismatch → no warm publish, awaited fetch", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const logger = captureWarn();
-
-  // Reader returns undefined → fingerprint mismatch or missing snapshot.
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published from live fetch");
-  // Live fetch data, not snapshot data.
-  assert.ok(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"],
-    "live fetch model present"
-  );
-  assert.equal(fetcher.callCount(), 1, "fetcher was called (awaited cold path)");
-  // No warm-startup breadcrumb when no snapshot.
-  assert.ok(
-    !logger.entries.some((e) =>
-      String(e[0]).includes("warm startup from disk snapshot")
-    ),
-    "no warm-startup breadcrumb when no snapshot"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (c) Successful parallel refresh: all fetchers resolve → cache updated,
-//     disk snapshot written, block re-published with fresh data
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: parallel refresh updates cache + writes snapshot", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([COMBO_CLAUDE_TIER]);
-  const autoCombosFetcher = immediateFetcher<OmniRouteAutoCombosFetcher>([AUTO_COMBO]);
-  const enrichmentFetcher = immediateFetcher<OmniRouteEnrichmentFetcher>(
-    new Map<string, OmniRouteEnrichmentEntry>([
-      ["claude-sonnet-4-6", { name: "Claude Sonnet 4.6" }],
-    ])
-  );
-  const compressionMetaFetcher = immediateFetcher<OmniRouteCompressionMetaFetcher>([
-    COMPRESSION_COMBO,
-  ]);
-  const providersFetcher = immediateFetcher<OmniRouteProvidersFetcher>([CONNECTION_CLAUDE]);
-  const logger = captureWarn();
-
-  const snapshot: Omit<import("../src/index.js").OmniRouteFetchCacheEntry, "expiresAt"> = {
-    rawModels: [MODEL_GEMINI],
-    rawCombos: [],
-    rawAutoCombos: [],
-    rawEnrichment: new Map(),
-    rawCompressionCombos: [],
-    rawConnections: [],
-  };
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => snapshot;
-  let snapshotWrites = 0;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {
-    snapshotWrites++;
-  };
-
-  const sharedCache: OmniRouteFetchCache = new Map();
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute", modelCacheTtl: 60_000 },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      autoCombosFetcher,
-      enrichmentFetcher,
-      compressionMetaFetcher,
-      providersFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      cache: sharedCache,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  // Warm block should have been published.
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "warm provider entry published");
-
-  // Give detached refresh time to complete.
-  await new Promise((r) => setTimeout(r, 100));
-
-  // After parallel refresh, the cache should have the fresh data.
-  const cacheKey = Array.from(sharedCache.keys())[0];
-  assert.ok(cacheKey, "cache entry created");
-  const cached = sharedCache.get(cacheKey)!;
-  assert.ok(cached.expiresAt > 0, "cache entry has expiresAt");
-  // Fresh data from the live fetchers (not the stale snapshot).
-  assert.equal(cached.rawModels.length, 1, "cache has fresh models");
-  assert.equal(cached.rawModels[0].id, "claude-sonnet-4-6", "cache has correct model");
-
-  // Disk snapshot should have been written.
-  assert.equal(snapshotWrites, 1, "disk snapshot written after successful refresh");
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (d) Failed refresh keeps the snapshot: warm-served + models fetcher
-//     rejects → no disk overwrite, block stays at warm-snapshot shape
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: failed refresh keeps the snapshot, no disk overwrite", async () => {
-  const fetcher = throwingFetcher<OmniRouteModelsFetcher>();
-  const combosFetcher = throwingFetcher<OmniRouteCombosFetcher>();
-  const logger = captureWarn();
-
-  const snapshot: Omit<import("../src/index.js").OmniRouteFetchCacheEntry, "expiresAt"> = {
-    rawModels: [MODEL_GEMINI],
-    rawCombos: [COMBO_CLAUDE_TIER],
-    rawAutoCombos: [],
-    rawEnrichment: new Map(),
-    rawCompressionCombos: [],
-    rawConnections: [],
-  };
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => snapshot;
-  let snapshotWrites = 0;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {
-    snapshotWrites++;
-  };
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "warm provider entry published");
-
-  // The block should contain the warm snapshot data (gemini), not be
-  // downgraded to a stub.
-  assert.ok(
-    entry.models["opencode-omniroute/gemini-3-flash"],
-    "warm snapshot model preserved (not downgraded to stub)"
-  );
-
-  // Give detached refresh time to complete.
-  await new Promise((r) => setTimeout(r, 100));
-
-  // No disk write on failed refresh.
-  assert.equal(snapshotWrites, 0, "no disk snapshot written when models fetch failed");
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (e) Parallelism: all six fetchers start concurrently (not sequential)
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: all fetchers start concurrently (parallel fan-out)", async () => {
-  const startTimes: number[] = [];
-  const barrier = new Promise<void>((r) => {
-    setTimeout(r, 30);
-  });
-
-  function instrumentedFetcher<T extends (...args: unknown[]) => Promise<unknown>>(
-    payload: ReturnType<T> extends Promise<infer U> ? U : never
-  ): T & { callCount: () => number } {
-    let n = 0;
-    const f = async (..._args: unknown[]) => {
-      startTimes.push(Date.now());
-      n++;
-      await barrier;
-      return payload;
-    };
-    return Object.assign(f as T, { callCount: () => n });
-  }
-
-  const fetcher = instrumentedFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = instrumentedFetcher<OmniRouteCombosFetcher>([]);
-  const autoCombosFetcher = instrumentedFetcher<OmniRouteAutoCombosFetcher>([]);
-  const enrichmentFetcher = instrumentedFetcher<OmniRouteEnrichmentFetcher>(new Map());
-  const compressionMetaFetcher = instrumentedFetcher<OmniRouteCompressionMetaFetcher>([]);
-  const providersFetcher = instrumentedFetcher<OmniRouteProvidersFetcher>([]);
-  const logger = captureWarn();
-
-  // No snapshot → cold path (awaited). All fetchers must still start
-  // concurrently.
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute", features: { enrichment: true, compressionMetadata: true, usableOnly: true } },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      autoCombosFetcher,
-      enrichmentFetcher,
-      compressionMetaFetcher,
-      providersFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  // All fetchers should have been called.
-  assert.equal(fetcher.callCount(), 1, "models fetcher called");
-  assert.equal(combosFetcher.callCount(), 1, "combos fetcher called");
-  assert.equal(autoCombosFetcher.callCount(), 1, "autoCombos fetcher called");
-  assert.equal(enrichmentFetcher.callCount(), 1, "enrichment fetcher called");
-  assert.equal(compressionMetaFetcher.callCount(), 1, "compressionMeta fetcher called");
-  assert.equal(providersFetcher.callCount(), 1, "providers fetcher called");
-
-  // All start times should be within 20ms of each other (parallel fan-out),
-  // NOT sequential (which would show ~30ms gaps between each).
-  assert.ok(startTimes.length >= 6, "all 6 fetchers started");
-  const minStart = Math.min(...startTimes);
-  const maxStart = Math.max(...startTimes);
-  assert.ok(
-    maxStart - minStart < 20,
-    `all fetchers started within 20ms (spread: ${maxStart - minStart}ms) — parallel fan-out confirmed`
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (f) Soft-fail parity under Promise.allSettled: per-endpoint fallbacks +
-//     logger.warn breadcrumbs preserved
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: combos reject → models-only catalog with warn", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = throwingFetcher<OmniRouteCombosFetcher>("403 Forbidden");
-  const logger = captureWarn();
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published");
-  assert.ok(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"],
-    "models-only catalog (no combos)"
-  );
-  assert.ok(
-    logger.entries.some((e) => String(e[0]).includes("/api/combos fetch failed")),
-    "combos-fetch breadcrumb emitted"
-  );
-});
-
-test("warm-startup: enrichment rejects → raw-id catalog with warn", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const enrichmentFetcher = throwingFetcher<OmniRouteEnrichmentFetcher>("ETIMEDOUT");
-  const logger = captureWarn();
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      enrichmentFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published");
-  assert.equal(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"].name,
-    "claude-sonnet-4-6",
-    "raw id retained (no enrichment)"
-  );
-  assert.ok(
-    logger.entries.some((e) => String(e[0]).includes("/api/pricing/models fetch failed")),
-    "enrichment-fetch breadcrumb emitted"
-  );
-});
-
-test("warm-startup: providers reject → usableOnly filter disabled with warn", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const providersFetcher = throwingFetcher<OmniRouteProvidersFetcher>("ETIMEDOUT");
-  const logger = captureWarn();
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute", features: { usableOnly: true } },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      providersFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published");
-  // Soft-fail: model kept (filter disabled).
-  assert.ok(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"],
-    "model kept (usableOnly filter disabled)"
-  );
-  assert.ok(
-    logger.entries.some((e) => String(e[0]).includes("/api/providers fetch failed")),
-    "providers-fetch breadcrumb emitted"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (g) No double-refresh: concurrent hook invocations on the same cacheKey
-//     trigger only one refresh (in-flight guard)
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: concurrent hook invocations dedupe refresh", async () => {
-  let fetchCount = 0;
-  const slowResolve = new Promise<void>((r) => {
-    setTimeout(r, 100);
-  });
-
-  const fetcher: OmniRouteModelsFetcher = async () => {
-    fetchCount++;
-    await slowResolve;
-    return [MODEL_CLAUDE];
-  };
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const logger = captureWarn();
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => undefined;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const sharedCache: OmniRouteFetchCache = new Map();
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute", modelCacheTtl: 60_000 },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      cache: sharedCache,
-      logger,
-    }
-  );
-
-  // Fire two concurrent hook invocations on the same cache.
-  const inputA = makeInput();
-  const inputB = makeInput();
-  await Promise.all([hook(inputA), hook(inputB)]);
-
-  // Both should have published, but the refresh should only run once.
-  assert.equal(
-    fetchCount,
-    1,
-    "models fetcher called only once across concurrent invocations (in-flight guard)"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// (h) features.diskCache: false disables the warm read entirely
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: diskCache=false disables warm read, falls through to awaited fetch", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const logger = captureWarn();
-
-  let readerCalled = false;
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => {
-    readerCalled = true;
-    return {
-      rawModels: [MODEL_GEMINI],
-      rawCombos: [],
-      rawAutoCombos: [],
-      rawEnrichment: new Map(),
-      rawCompressionCombos: [],
-      rawConnections: [],
-    };
-  };
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute", features: { diskCache: false } },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  assert.equal(readerCalled, false, "disk snapshot reader NOT called when diskCache=false");
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published from live fetch");
-  assert.ok(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"],
-    "live fetch model present (not snapshot)"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// Warm startup: snapshot age logged
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: snapshot age is logged when warm-starting from disk", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const logger = captureWarn();
-
-  const snapshot: Omit<import("../src/index.js").OmniRouteFetchCacheEntry, "expiresAt"> & {
-    writtenAt?: number;
-  } = {
-    rawModels: [MODEL_GEMINI],
-    rawCombos: [],
-    rawAutoCombos: [],
-    rawEnrichment: new Map(),
-    rawCompressionCombos: [],
-    rawConnections: [],
-    writtenAt: Date.now() - 3_600_000, // 1 hour ago
-  };
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => snapshot;
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  // The log should mention "warm startup from disk snapshot".
-  assert.ok(
-    logger.entries.some((e) =>
-      String(e[0]).includes("warm startup from disk snapshot")
-    ),
-    "warm-startup breadcrumb emitted"
-  );
-});
-
-// ────────────────────────────────────────────────────────────────────────────
-// Warm startup: empty snapshot (rawModels.length === 0) is skipped
-// ────────────────────────────────────────────────────────────────────────────
-
-test("warm-startup: empty snapshot (rawModels.length=0) is skipped, falls through to fetch", async () => {
-  const fetcher = immediateFetcher<OmniRouteModelsFetcher>([MODEL_CLAUDE]);
-  const combosFetcher = immediateFetcher<OmniRouteCombosFetcher>([]);
-  const logger = captureWarn();
-
-  const diskSnapshotReader: OmniRouteDiskSnapshotReader = async () => ({
-    rawModels: [],
-    rawCombos: [],
-    rawAutoCombos: [],
-    rawEnrichment: new Map(),
-    rawCompressionCombos: [],
-    rawConnections: [],
-  });
-  const diskSnapshotWriter: OmniRouteDiskSnapshotWriter = async () => {};
-
-  const hook = createOmniRouteConfigHook(
-    { providerId: "omniroute" },
-    {
-      readAuthJson: authStub(),
-      fetcher,
-      combosFetcher,
-      diskSnapshotReader,
-      diskSnapshotWriter,
-      logger,
-    }
-  );
-
-  const input = makeInput();
-  await hook(input);
-
-  const entry = (input as { provider: Record<string, OmniRouteStaticProviderEntry> }).provider[
-    "opencode-omniroute"
-  ];
-  assert.ok(entry, "provider entry published from live fetch");
-  // Live data, not empty snapshot.
-  assert.ok(
-    entry.models["opencode-omniroute/claude-sonnet-4-6"],
-    "live fetch model present (empty snapshot skipped)"
-  );
-  assert.equal(fetcher.callCount(), 1, "fetcher was called (awaited cold path)");
-});

changelog.d/features/9490-opencode-plugin-warm-startup-parallel-refresh.md

@@ -1,5 +0,0 @@
----
-feature: 9490
----
-
-**Warm catalog startup from disk snapshot + parallel refresh** (opencode-plugin): The config-shim hook now reads the last disk snapshot *before* fetching, so the provider registers immediately with the last-known-good catalog (~1-2s vs ~30s on a warm gateway). All six fetchers run concurrently via `Promise.allSettled` instead of sequentially. A failed refresh keeps the snapshot (no overwrite). An in-flight guard prevents concurrent refreshes for the same cache key. The `features.diskCache: false` opt-out disables the warm read entirely.

changelog.d/fixes/8542-fix.plan.md

@@ -1 +0,0 @@
-- fix(ci): aggregate all fast-gates into non-fail-fast loop so one red gate no longer masks later gates (#8542)

changelog.d/fixes/8951-fix.plan.md

@@ -1 +0,0 @@
-- fix(github): add targetFormat to GPT-5.6 Sol/Terra/Luna models (#8951)

changelog.d/fixes/8997-gpt56-max-reasoning.plan.md

@@ -1 +0,0 @@
-- **fix(translator):** the Responses-to-Chat promotion path called `normalizeResponsesReasoningEffort` without the model argument, so GPT-5.6 Sol/Terra/Luna requests with `reasoning.effort: "max""` were downgraded to `"xhigh"`. The model is now threaded through, preserving `max` for GPT-5.6 while keeping the legacy downgrade for older models ([#8997](https://github.com/diegosouzapw/OmniRoute/pull/8997))

changelog.d/fixes/9034-fix.plan.md

@@ -1 +0,0 @@
-- fix(api): use configured prefix instead of raw node UUID for alias-backed model id in /v1/models (#9034)

changelog.d/fixes/9057-fix.plan.md

@@ -1 +0,0 @@
-- fix(api): auto/* routing aliases bypass API-key allowedConnections/disableNonPublicModels (#9057)

changelog.d/fixes/9096-fix-audio-speech-transcriptions-translations-provider-nodes.plan.md

@@ -1 +0,0 @@
-- fix(providers): admit audio-speech/audio-transcriptions apiType in audio route provider-node filters (#9096)

changelog.d/fixes/9134-fix.plan.md

@@ -1 +0,0 @@
-- fix(providers): resolve combo names in audio transcriptions route so /v1/models stays honest (#9134)

changelog.d/fixes/9140-fix.plan.md

@@ -1 +0,0 @@
-- fix(vscode): allow built-in auto-routing models in VS Code model filter (#9140)

changelog.d/fixes/9269-fix.plan.md

@@ -1 +0,0 @@
-- **fix(classify429):** add missing `have exhausted their quota` pattern so the synthetic 429 from auth.ts is recognized as quota exhaustion, preventing the combo loop from burning retries against the same provider instead of falling back to a healthy one ([#9269](https://github.com/diegosouzapw/OmniRoute/issues/9269))

changelog.d/fixes/9305-fix.plan.md

@@ -1 +0,0 @@
-- fix(sse): broaden OMNIROUTE_SSE_COMMENTS to accept 'false','0','no' and gate metadata comment emission (#9305)

changelog.d/fixes/9306-fix.plan.md

@@ -1 +0,0 @@
-- fix(lmarena): encode SSE stream chunks as Uint8Array to prevent TextDecoder TypeError (#9306)

changelog.d/fixes/9615-docker-colocate-partial-trace.md

@@ -1 +0,0 @@
-- **fix(docker):** standalone co-location now completes packages Next's file tracing materialized partially (package.json without its `main` payload) — unblocks the Docker Hub publish that failed on every v3.8.50 push with `Cannot find module '@atjsh/llmlingua-2/dist/index.js'` ([#9615](https://github.com/diegosouzapw/OmniRoute/pull/9615))

changelog.d/fixes/9737-memory-id-route-backend.md

@@ -1 +0,0 @@
-- Fixed `GET`/`PUT`/`DELETE /api/memory/[id]` always failing with a 500 (`Primary backend "sqlite" not registered`) when the route was reached before any other memory endpoint in the same process.

changelog.d/fixes/9737-route-body-validation.md

@@ -1 +0,0 @@
-- Replaced hand-rolled body type checks with Zod validation in the plugins marketplace install route and the three Dario admin routes, restoring the `t06:route-validation` gate (Hard Rule #7).

changelog.d/fixes/9776-radar-entitlement-refresh.md

@@ -0,0 +1 @@
+- **fix(radar):** refresh signed catalog/referral caches when supporter entitlement changes, preserve the one-time live-to-community downgrade, and test real provider connection IDs from the setup tour

changelog.d/maintenance/9614-opencode-plugin-bare-key-suite.md

@@ -1 +0,0 @@
-- **test(cli):** OpenCode plugin suite realigned to the bare-key static-catalog contract from #9178/#9175 (21 tests were red on every opencode-plugin CI run; 287/287 after) ([#9614](https://github.com/diegosouzapw/OmniRoute/pull/9614))

config/quality/file-size-baseline.json

@@ -1,4 +1,7 @@
 {
+  "_rebaseline_2026_08_08_v3850_base_drift_batch_9757": "Base drift on release/v3.8.50, not own growth: the 08-06..08-08 merge batches grew 12 already-frozen (or newly-landed) files without carrying their rebaselines — the dedicated rebaseline PR #9616 was closed as 'superseded' but its file-size entries never actually reached the base, and later merges (#8894 combos page, #9539 EditConnectionModal, #8895 models route, #9294/#9293 catalog, #9541 db/core, #8970 tokenHealthCheck, #8925 mcp schemas+server, #8890 accountFallback, #9467 chat.ts, #8931 openai-to-kiro, ProxyRegistryManager) kept growing them. All 12 values re-measured on THIS branch's tree (= pure tip + this PR's 1-line chat.ts fix, which adds zero lines). This PR's own source changes (chat.ts identifier restore, stream.ts format carve-out) do not grow any frozen file past these values.",
+  "_rebaseline_2026_08_08_migration_135_collision": "fix(db): resolve migration version 135 numbering collision — #9449's 135_connection_runtime_state.sql and #8908's 135_migrate_model_capability_max_token.sql both claimed version 135 (#9449 branched before #8908 merged and never got renumbered before landing on release/v3.8.50), which threw 'Migration version collision detected' the moment ANY code touched the database — a fresh install/deploy from this tip cannot even boot. Renumbered the later-landing file to 140 (next free slot) and added the matching isSchemaAlreadyApplied('140') retroactive guard, matching the established pattern already used for the prior 135/136 -> 137/138 renumber in the same file. Own growth: src/lib/db/migrationRunner.ts 1084->1094 (+10, the new case block) — irreducible, matches the existing per-case guard pattern exactly. Covered by tests/unit/migration-135-numbering-collision.test.ts (2/2), confirmed failing (reproducing the exact live crash) against the pre-fix colliding filenames, passing after.",
+  "_rebaseline_2026_08_02_9259_rolling_rpm": "PR #9259 (issue #8733) own growth: open-sse/services/rateLimitManager.ts baseline 1060->1167 (+107; final source 1153). The existing withRateLimit chokepoint now composes process-local rolling RPM leases with Bottleneck admission, releases pre-dispatch leases on queue timeout/abort/connection disable, preserves caller abort reasons, and wires 429/header state into the extracted rollingRpmGate.ts. The remaining growth is irreducible lifecycle wiring at the dispatch boundary plus the real watchdog test hooks needed to verify queued-wedge recovery; moving it further would obscure lease ownership and Bottleneck cleanup. Covered by the focused rate-limit manager/sliding-window suite (33/33); distributed multi-instance coordination remains explicitly out of scope.",
   "_rebaseline_2026_07_24_8470_hyperagent_sticky_thread": "PR #8470 (artickc, fix/hyperagent-tool-loop-thread-sticky) own growth: open-sse/executors/hyperagent.ts 936->1025 (wc -l; check-file-size.mjs counts via split(\"\\n\").length so the gate sees 937->1026, +89, crosses the 1000 cap). Fixes a real bug where a reverse-conversion proxy (text-Intent/JSON to Claude Code native tool_calls) rewrites assistant messages between agentic tool-loop turns, breaking HyperAgent’s conversation-prefix fingerprint and cold-starting the thread mid tool-loop. Adds Anthropic tool_use/tool_result flattening to extractMessageText() plus a new rootUserFingerprint()/root-key lookup tier in resolveHyperAgentThreadBinding()/storeHyperAgentThreadAfterTurn() so the thread stays sticky across the tool loop. Cohesive additions inside the existing single-file executor; not extractable without splitting the executor mid-request-flow. Covered by tests/unit/executor-hyperagent.test.ts (19/19, +5 new cases for tool_result/tool_use flattening + root-key stickiness). Pre-merge review flagged a cross-conversation root-key collision risk (tracked in the PR’s own mandatory pre-merge checklist, not yet addressed) — unrelated to this file-size ratchet, tracked separately by /fix-prs.",
   "_rebaseline_2026_07_25_8494_capability_filter_fail_closed": "PR #8494 (fix/capability-filters-fail-closed, #8488) own growth: open-sse/services/combo.ts 3640->3693 (+53) adds a fail-closed guard after filterTargetsByRequestCompatibility() — when every eligible target is excluded by request-capability filtering (vision/tools/etc) instead of quota/health, the combo now returns an explicit `capability_mismatch` 400 (describeCapabilityFilterExhaustion, imported from combo/comboStructure.ts) rather than silently falling through to a generic no-targets error, plus a `compatFilterFailOpen` escape hatch (combo config OR settings) mirrored at both the main/auto and round-robin call sites for symmetry. combo/comboStructure.ts (previously under cap, un-frozen) grows 794->918 (+124) — new home for describeCapabilityFilterExhaustion + providerSupportsEmulatedToolCalling (#5240 emulated tool-calling exemption so fail-closed does not regress prompt-emulation-only combos like all-chatgpt-web). Irreducible orchestration wiring at the existing filter chokepoint (same precedent as #7301's universal-cooldown-retry generalization). Companion test tests/unit/combo-routing-engine.test.ts 3409->3449 (+40, fail-closed/fail-open coverage across both call sites) also rebaselined. Covered by tests/unit/8488-capability-filter-fail-closed.test.ts (new) + 95/95 passing across both files. Structural shrink of combo.ts tracked in #3501.",
   "_rebaseline_2026_07_25_8499_ts7_result_union_predicates": "PR #8499 (backryun, chore/ts7-types-executor-scattered) own growth: muse-spark-web.ts 1396->1405 (+9, irreducible). Under this workspace's `strictNullChecks: false`, the boolean-literal discriminant on `GraphqlResult` (`{ ok: true } | { ok: false; error: string }`) narrows the positive `.ok===true` branch but leaves `!result.ok` at the full union under TS7, making `.error` unreachable to the checker at the two call sites (warmup, mode-switch). Fixed by adding a single `isGraphqlFailure()` type-predicate helper (doc comment + 3-line body) reused at both call sites instead of duplicating the predicate inline — not extractable to a shared module without splitting a single-file executor's local narrowing helper out of its own file. Covered by the existing muse-spark-web executor test suite (no behavior change, pure narrowing fix).",
@@ -13,6 +16,7 @@
   "_rebaseline_2026_07_19_7546_ghe_copilot_route": "PR #7546 (GHE Copilot OAuth provider) own growth: oauth/[provider]/[action]/route.ts 960->963 (gate units, +3 = ghe-copilot device-code wiring at the existing multi-provider device-code branch — reading + HTTPS-validating the gheUrl search param (isValidGheUrl guards at both raw entry points, security-review hardening, 963->970), adding ghe-copilot to the no-PKCE provider set, and building the provider config override / threading gheUrl through poll->postExchange extraData). Mirrors the existing kiro/amazon-q startUrl override pattern right above it in the same branch; cohesive with the existing device-code dispatch chokepoint, not separately extractable without splitting a single provider-switch mid-branch. Frozen so can only shrink; structural shrink tracked in #3501.",
   "_rebaseline_2026_07_19_6846_nvidia_concurrency_gate": "Issue #6846 Phase 1 (nvidia NIM local RPM budget + per-model lockout + per-connection concurrency cap) own growth: open-sse/executors/default.ts 877->890 (+13 = the irreducible call-site wiring at DefaultExecutor.execute(), the only place nvidia requests dispatch through — the existing session-pool body was extracted verbatim into a new private executeWithSessionPool() so the outer execute() can wrap it in the nvidia concurrency-gate acquire/finally-release). All actual gating logic (semaphore key + cap resolution) lives in the new leaf open-sse/executors/default/nvidiaConcurrencyGate.ts (not frozen, well under cap). Covered by tests/unit/nvidia-quota-phase1.test.ts.",
   "_rebaseline_2026_07_18_v3849_provider_detail_wiring": "Merge campaign R2/R3 (2026-07-18): three authorized PRs each add irreducible call-site wiring to ProviderDetailPageClient.tsx — #7360 +5 (ProviderQuotaVisibilityToggle render, component extracted), #7419 +4 (NoAuthProviderControls wiring), #7062 +3 (Dahl provider hook) = 786->798. All three follow the extracted-component pattern (AgentrouterConsoleFields precedent); the frozen file only takes the wiring. Structural shrink tracked in #3501.",
+  "_rebaseline_2026_07_25_dario_upstream_proxy_selector": "PR #8523 (Dario embedded service): upstream-proxy mode selector replaces the binary CLIProxyAPI toggle with Native/CLIProxyAPI/Dario/Fallback + a fallback-backend picker. ProviderDetailPageClient.tsx 798->804 (+6, new hook fields threaded through to ConnectionsListPanel), ConnectionRow.tsx 942->958 (+16, the mode <select> + conditional fallback-backend <select> replacing a single pill button), useProviderConnections.ts 954->986 (+32, upstreamProxyMode/upstreamProxyFallbackBackend state + handleSetUpstreamProxyMode, handleToggleCliproxyapiMode kept as a thin backward-compat wrapper for the existing hook-shape test). All additive UI/state for the new modes — no unrelated refactor.",
   "_rebaseline_2026_07_18_pr7653_chat_tracker_import": "PR #7653 merge-interaction growth: release moved chat.ts to its 1796 cap while this PR adds the single side-effect import 'quotaTrackersBatch.ts' (line 130) — chat.ts IS the canonical quota-fetcher registration point (codex/bailian/deepseek/openrouter/opencode/generic all import+register there), so the +1 is irreducible call-site wiring. 1796->1797. Covered by tests/unit/{agentrouter,v0,freemodel}-quota-fetcher.test.ts.",
   "_rebaseline_2026_07_17_pr7653_agentrouter_console_fields": "PR #7653 own growth (missing acceptance criterion: the AgentRouter quota tracker (#6850) read providerSpecificData.consoleApiKey/newApiUserId but neither field had dashboard UI for provider agentrouter — consoleApiKey was gated to bailian-coding-plan only and newApiUserId had zero UI). AddApiKeyModal.tsx 961->967 (+6) and EditConnectionModal.tsx 1278->1286 (+8) = import + a single <AgentrouterConsoleFields .../> render call plus the newApiUserId formData init field. The actual Input rendering (both consoleApiKey reuse + the new newApiUserId field) was EXTRACTED into a new leaf src/app/(dashboard)/dashboard/providers/[id]/components/modals/AgentrouterConsoleFields.tsx (48 LOC, <cap), mirroring the QuotaScrapingFields.tsx / GlmTeamQuotaFields.tsx precedent (#6351) so the frozen modals only carry the irreducible call-site wiring. Persist logic lives in connectionProviderSpecificData.ts (not frozen). Covered by tests/unit/agentrouter-connection-modal-fields.test.ts.",
   "_rebaseline_2026_07_17_v3849_6842_free_window_wiring": "PR #7651 (openrouter :free-window quota tracking) follow-up: the counter shipped built but never wired into the request pipeline, so combos kept spending guaranteed-429 requests on exhausted free-tier targets. Own growth: src/sse/services/auth.ts 2461->2462 (+1, irreducible at the existing model-aware preflight chokepoint — the `provider === \"codex\"` check that forwards requestedModel into the connection arg is extended to also cover `openrouter`, one added boolean + a doc comment, offset to a single net line by dropping the now-redundant inline condition). Enforcement itself lives in open-sse/services/openrouterQuotaFetcher.ts (not frozen) and the dispatch-time record/correct hooks live in open-sse/executors/base.ts (not frozen). Covered by tests/unit/openrouter-free-window-wiring-6842.test.ts.",
@@ -157,8 +161,134 @@
   "_rebaseline_2026_06_20_1409_1294_models": "Re-baseline src/lib/db/models.ts 1184->1221: combined growth of sibling fixes #1409 (cascade-delete orphaned model aliases when a provider is removed) + #1294 (persist max_input_tokens/max_output_tokens on custom models), both adding CRUD at the existing models domain module. Cohesive db module; not extractable.",
   "_rebaseline_2026_06_20_4389_thinking_toolchoice": "Re-baseline base.ts 1387->1399 (#4389): tool_choice-forced thinking guard at the existing Claude wire-image injection chokepoint (effThinking gate avoids the Anthropic 400 when tool_choice forces a tool). Cohesive guard; structural shrink tracked in #3501.",
   "_rebaseline_2026_07_18_6979_codex_test": "PR #6979 own growth: executor-codex.test.ts 1340->1347 (+7 = generalized ensureThinkingBudget assertion added to the existing codex thinking-budget cases). antigravity-test bump 942->977 REVERTED here: #7408's test split dropped that file to 888, so this PR's +35 fits under the original 942 frozen cap.",
-  "_rebaseline_2026_07_24_8354_logs_timeline_sidebar": "PR #8354 (hartmark, feature/scrolling-log) own growth: src/shared/constants/sidebarVisibility/sections.ts 812->820 (+8, the single new logs-timeline SidebarItemDefinition entry added to LOGS_GROUP.items for the new /dashboard/logs/timeline scrolling request-timeline page). Irreducible data-literal wiring at the existing sidebar-sections chokepoint, same shape as every other item in the file; not extractable without an ad-hoc single-item exception to the file's otherwise-uniform multi-line item style.",
   "cap": 1000,
+  "frozen": {
+    "_rebaseline_2026_06_22_4644_deepseek_web_tools": "PR #4644 (BugsBag/robust deepseek-web tool-call parsing): open-sse/executors/deepseek-web.ts 1117->1125 (+8). The new agentic tool-call path emits surrounding text + reasoning before tool_calls and swaps to the dedicated deepseekWebTools.ts parser; the +8 lines are cohesive wiring at the existing transformSSE chokepoint (the parser itself lives in the new deepseekWebTools.ts file, already under cap). The PR's own fast-gate (PR->release) does not run check:file-size, so this surfaced only at release reconcile. Covered by tests/unit/deepseek-web-tools-variants.test.ts + deepseek-web-tools-execute.test.ts.",
+    "_rebaseline_2026_06_23_4712_deepseek_web_tool_results": "PR for #4712 (deepseek-web drops role:tool): open-sse/executors/deepseek-web.ts 1125->1148 (+23). messagesToPrompt() now folds role:\"tool\" results into the single-prompt transcript (recovering the tool name from the preceding assistant tool_calls by tool_call_id) instead of silently dropping them; the lines are cohesive wiring inside the existing function. Covered by tests/unit/deepseek-web-tool-result-prompt-4712.test.ts.",
+    "_rebaseline_2026_06_24_headroom_strategy": "Headroom-aware connection selection (dario technique): combo.ts 3168->3180 (+12 = a new `else if (strategy === \"headroom\")` dispatch branch in handleComboChat that delegates to orderTargetsByHeadroom + its log line, plus the import). The actual logic lives OUT of the god-file: the pure ranker rankByHeadroom/computeHeadroom is the new leaf open-sse/services/combo/headroomRanking.ts (91 LOC, <cap) and the async orderer orderTargetsByHeadroom is appended to the existing open-sse/services/combo/quotaStrategies.ts (<cap) next to its sibling reset-aware/reset-window orderers (reuses their connection-expansion machinery). headroom = 1 - max(util_5h, util_7d) from getSaturation (src/lib/quota/saturationSignals.ts), prefers the connection with the most free capacity. Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. fill-first stays default; all existing strategies untouched. Covered by tests/unit/combo-headroom-ranking.test.ts (pure helper) + tests/unit/combo-headroom-strategy.test.ts (orderer, saturation injected). Structural shrink of combo.ts tracked in #3501.",
+    "_rebaseline_2026_06_24_quota_share_strategy": "Dedicated quota-share strategy (Phase 3 #9): combo.ts 3180->3190 (+10 = one new `else if (strategy === \"quota-share\")` dispatch branch in handleComboChat that delegates 100% to selectQuotaShareTarget + its log line, plus the import). All the new logic lives OUT of the god-file in two new leaves under open-sse/services/combo/: quotaShareInflight.ts (in-flight counter with TTL/lease, ~150 LOC <cap) and quotaShareStrategy.ts (per-model bucket gating via isBucketSaturated + DRR proportional to weight + P2C over in-flight, ~240 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the headroom/reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. ZERO existing strategy cases were modified — only this branch was added, and the qtSd/ combos switched from fill-first to quota-share in src/lib/quota/quotaCombos.ts. Covered by tests/unit/quota-share-strategy.test.ts (gating, DRR fairness, P2C in-flight, fail-open, activation). Structural shrink of combo.ts tracked in #3501.",
+    "_rebaseline_2026_06_24_task_aware_routing": "Task-aware routing strategy (port PR #2045, OmniRoute #4945): combo.ts 3190->3225 (+35) = one new `else if (strategy === \"task-aware\")` dispatch branch delegating 100% to selectTaskAwareTarget + its imports/log lines. All scoring/classification logic lives OUT of the god-file in the new leaf open-sse/services/taskAwareRouting.ts (553 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors quota-share/headroom/reset-aware branches). ZERO existing strategy cases modified. Covered by tests/unit/combo-task-aware.test.ts (35 tests). Structural shrink of combo.ts tracked in #3501.",
+    "_rebaseline_2026_06_26_fidelity_gate_extraction": "Milestone-B fidelity-gate wiring residual: bodyToText+gateAdvance extracted to fidelityGateStep.ts (889->854, -35), but the StackOptions.fidelityGate field, the `const fidelityGate` reads at the two stacked-loop dispatch chokepoints, and the import of FidelityGateConfig are irreducible wiring that cannot leave strategySelector without an architectural refactor of the pre-existing stacked pipeline. Net: 889->854 (+6 vs the pre-Milestone-B frozen 848). Covered by tests/unit/compression/*.test.ts (940 pass).",
+    "_rebaseline_2026_06_27_5193_5203_antigravity_oauthmodal": "Antigravity remote-login own growth: OAuthModal.tsx 960->969 (gate units). #5193 (+~4: remote paste instruction shown for all remote incl. Google + its rationale comment) and #5203 (+~5: handleManualSubmit credential-blob branch + button guard; submit logic extracted to oauthBlobSubmit.ts to minimize). Frozen set to the SUM so either merge order passes. Cohesive at the existing manual-submit chokepoint.",
+    "_rebaseline_2026_06_27_5193_antigravity_basered": "Base-red (pre-existing release drift, fast-gate PR->release skips check:file-size): accountFallback.ts 1773->1777 and src/app/api/providers/[id]/test/route.ts 924->940 were already over their frozen caps on release/v3.8.39 independent of any antigravity change. Owner chose to rebaseline (keep the documented issue-reference comments #1846/#1449/#347 etc.) rather than accept the contributor comment-stripping in #5200/#5198. Reverted #5200 to restore the comments; bumped these two frozen caps to the actual base sizes. No logic change.",
+    "_rebaseline_2026_06_28_5237_impersonation_ua_refresh": "PR #5237 (refresh impersonation UAs): grok-web.ts 1871->1873 (+2), muse-spark-web.ts 1284->1302 (+18), perplexity-web.ts 1013->1032 (+19). Net semantic change in each file is a single User-Agent constant (Chrome 147->149 for grok/muse; perplexity kept at Firefox 148 to stay matched with the firefox_148 TLS profile — the contributor's 152 bump was reverted to avoid a UA-vs-JA3 mismatch, #2459). The growth is Prettier reflow that lint-staged unavoidably applies to these grandfathered long-line files the moment they are touched; not extractable. src/sse/services/auth.ts 2336->2401 in the same reconcile is #5222's antigravity-LRU-retry growth that merged via --admin without a baseline bump.",
+    "_rebaseline_2026_06_28_5243_risk_gate_prepass": "PR #5243 (compression risk-gate pre-pass) own growth: open-sse/services/compression/strategySelector.ts 854->899 (+45). The three exported entry points (applyCompression/applyStackedCompression/applyStackedCompressionAsync) become thin wrappers over pure-extracted private bodies (runCompression/runStackedCompression/runStackedCompressionAsync) so the risk-gate mask->run->restore wrapper sits strictly OUTSIDE the per-step loop — a single universal integration point. The wrapper logic itself (resolveRiskGate/withRiskGate) lives in the new riskGate/strategyWrap.ts (<cap); the residual growth is the duplicated thin-wrapper signatures + the extracted bodies' dispatch boundary, guarded by a byte-identical parity test (riskGateIntegration). Default off (DEFAULT_COMPRESSION_CONFIG unchanged). Not extractable without hiding the dispatch boundary, mirroring prior compression rebaselines. Structural shrink tracked in #3501.",
+    "_rebaseline_2026_06_28_5275_correlation_id_extract": "Extraction of the safe CorrelationId subset of #5275 (hartmark) — request correlation id stored in call_logs (migration 109) and returned via the X-Correlation-Id response header, WITHOUT the combo/resilience or build/lazy-loading changes (those stay in #5275). Own growth: callLogs.ts 975->985 (correlation_id column on CallLogSummaryRow + read/map), usageHistory.ts 983->988 (correlationId metadata normalize), chat.ts 1575->1632 (withCorrelationId response wiring + combo-failure log carrying correlationId), chatHelpers.ts new 811 (withCorrelationId helper + reqId threading; was 791<cap pre-feature). Cohesive request/logging chokepoint wiring; structural shrink of chat.ts tracked in #3501.",
+    "_rebaseline_2026_06_29_4038_cas_guard": "PR (#4038) own growth: tokenRefresh.ts 2103->2181 (+78 = the compare-and-swap guard on the refresh persist — runWithCasGuard/getActiveCasGuard AsyncLocalStorage pair mirroring runWithOnPersist, casGuardShouldSkipPersist that rereads the row right before persisting and skips the write when a concurrent writer already rotated the refresh_token past the one presented, plus getCasGuardStats counters). Fixes the sibling-rotation-revert → token-family-revocation storm. Gated behind an active guard (opt-in; no guard => byte-identical). Wiring lives at the two persist chokepoints inside getAccessToken; the comparison reuses wasRefreshTokenRotated from refreshSerializer. Not extractable without splitting the refresh hot path.",
+    "_rebaseline_2026_06_29_5286_memoization": "PR #5286 own growth: strategySelector.ts 899->960 (+61 = the opt-in result-memoization branches in applyCompression/applyCompressionAsync — principal+determinism gate, makeMemoKey lookup/store with model+supportsVision folded into the key, recompute-with-memo-off). Default off (memoizeCompressionResults), so zero behavior change. The memo helpers live in the leaf resultMemo.ts (<cap); the chokepoint wiring here is not extractable. Structural shrink of this hot-path file tracked in #3501.",
+    "_rebaseline_2026_07_01_v3843_release_5609": "Rebaseline v3.8.43 (PR #5609 release reconciliation). DRIFT dos 109 commits do ciclo: 8 god-files existentes cresceram (ApiManagerPageClient 2983->3017, combos/page 4594->4608, AddApiKeyModal 868->869, providerPageHelpers 974->996, chat.ts 1635->1647, auth.ts 2401->2403, batchProcessor 828->915, combo.ts 3368->3387) + 2 novos acima do cap (huggingchat.ts 813, tests web-cookie-providers-new 827) + 4 test files cresceram. Modularizacao deferida (blast-radius mid-release); congelado no estado atual p/ o proximo ciclo ratchetar daqui.",
+    "_rebaseline_2026_07_02_5816_qoder": "PR #5816 (@AgentKiller45, qoder PAT via qodercli): qoderCli.ts 666->989, new-above-cap frozen (owner-approved baseline freeze). The growth is the legitimate PAT job-token exchange + quota parsing CLI transport (the pure-JS Cosy path 500'd on every PAT request); extracting the spawn/parse helpers now would just add indirection to a contributor PR mid-merge. Test frozen also raised for this PR's coverage growth: providers-page-utils.test.ts 1052->1092. Additionally clears an inherited base-red from the already-merged #5933 (codex json_schema->text.format): translator-openai-responses-req.test.ts 1097->1172 (+75 regression tests, no offending branch left). All remain frozen (cannot grow further); release captain's rebaseline-at-release supersedes.",
+    "_rebaseline_2026_07_09_6126_clinepass_dual_auth": "PR #6126 (@hajilok, dual-auth ClinePass) own growth: tokenRefresh.ts 2181->2182 (+1 = a single `case \"clinepass\":` fallthrough label added to the existing `case \"cline\":` in _getAccessTokenInternal's provider switch, so clinepass token refresh dispatches to the already-shared refreshClineToken() instead of silently falling through to the generic OAuth refresh). Irreducible 1-line switch-case wiring at the existing chokepoint; the header-building logic for the same feature was extracted to a new leaf src/shared/utils/clineAuth.ts::buildClinepassHeaders() (well under cap) to avoid growing open-sse/executors/default.ts. Covered by tests/unit/clinepass-provider.test.ts.",
+    "_rebaseline_2026_07_09_6363_kiro_external_idp": "PR #6363 (@artickc, Kiro external IdP) own growth: tokenRefresh.ts 2182->2249 (+67 = the external_idp refresh branch inside refreshKiroToken — standard public-client OAuth2 refresh_token grant against the org IdP tokenEndpoint via buildExternalIdpRefreshParams/isExternalIdpAuthMethod from the new leaf open-sse/services/kiroExternalIdp.ts, with invalid_grant/invalid_client -> unrecoverable_refresh_error mapping). Cohesive addition at the existing refreshKiroToken chokepoint. Covered by tests/unit/kiro-external-idp.test.ts.",
+    "_rebaseline_2026_07_09_6587_kiro_api_key_auth": "PR #6587 (@strangersp) own growth for Kiro long-lived API-key auth, merged onto v3.8.47 tip: openai-to-kiro.ts 890->912 (+22, auth-header selection for API-key-vs-OAuth-token connections), providerLimits.ts 998->1000 (+2, API-key auth-type branch), translator-openai-to-kiro.test.ts 1234->1257 (+23), providers-page-utils.test.ts 1109->1107 (net -2 after merging with parallel release drift; connectionMatchesProviderCard api_key coverage added), provider-validation-specialty.test.ts 2856->2980 (+124 net after merge with parallel release drift; this PR also removed the file's `@typescript-eslint/no-explicit-any` eslint-suppression entry by fixing all `any` usages, adding typed replacements). Cohesive additive feature growth, well tested; not extractable without splitting the existing chokepoints mid-merge.",
+    "_rebaseline_2026_07_09_6678_routing_strategy_9router": "#6678 (SeaXen) — 9router-parity Routing Strategy settings card + per-provider/combo sticky-round-robin override. Own growth: ProviderDetailPageClient.tsx 784->786 (single ProviderAccountRoutingCard mount + import), auth.ts 2448->2458 (providerStrategies override resolution: fallbackStrategy/stickyRoundRobinLimit per-provider cascade in getProviderCredentials). Both additive, zero unrelated refactor; new UI/logic lives in new files (ProviderAccountRoutingCard.tsx, RoutingStrategyCard.tsx, rrState.ts::resolveComboStickyRoundRobinLimit). chat.ts value below reflects the current release tip (grown by other concurrent PRs, e.g. #6640), not this PR own change.",
+    "_rebaseline_2026_07_10_6318_omp_letta": "PR #6318 (@hamsa0x7, omp+letta CLI integrations) own growth: cliTools.ts (+53 = 2 registry entries incl. omp docsUrl) and cliRuntime.ts (+18 = runtime-detection wiring for the 2 new tools). Cohesive registry/wiring growth at the existing chokepoints; scope reduced from the original 5 tools (pi/codewhale/jcode shipped separately).",
+    "_rebaseline_2026_07_10_gcf_v3_2_decode": "PR #6838 own growth: new vendored file open-sse/services/compression/engines/headroom/gcf/decode_generic.ts frozen at 880 (> 800 cap). It is the vendored GCF generic-profile decoder (spec v3.2 nested flattening plus the prototype-pollution / hasOwnProperty hardening added in this PR's Gemini review). Kept as one file faithful to upstream gcf-typescript so re-vendoring stays a clean copy rather than a re-split each cycle (sibling generic.ts/scalar.ts stay < cap; extraction would also fragment the file's frozen eslint no-explicit-any suppressions). Round-trip + prototype-pollution regression coverage in tests/unit/compression/headroom-smartcrusher.test.ts. Frozen: only shrinks from here.",
+    "_rebaseline_2026_07_12_v3847_mergeprs_tail": "v3.8.47 /merge-prs tail (owner-approved): src/lib/localDb.ts NEW>800 (799->805, +6 re-exports countFreeProxies + recordFreeProxySyncErrors/clearFreeProxySyncErrors/getFreeProxySyncErrors + FreeProxySyncErrors type for #6909 free-pool relay-repair; re-export-only per Hard Rule #2, not extractable).",
+    "_rebaseline_2026_07_15_7070_combos_memo": "PR #7070 (perf/p1-memo) own growth: src/app/(dashboard)/dashboard/combos/page.tsx 4655->4656 (+1 = React.memo wrapping of ComboCard). Covered by tests/unit/ui/combos-page-smoke.test.tsx.",
+    "_rebaseline_2026_07_18_7399_xai_oauth_modal": "PR #7399 (xAI OAuth PKCE) own growth: OAuthModal.tsx 993->998 (+5 = provider entry + PKCE flow branch wiring at the existing provider-switch chokepoint; the provider logic itself lives in src/lib/oauth/providers/xai-oauth.ts, new leaf). Third irreducible wiring bump on this modal (969->989->993->998); structural shrink tracked in #3501.",
+    "_rebaseline_2026_07_19_6636_codex_session_json": "#6636 own growth: OAuthModal.tsx 998->1030 (gate units, split(\"\\n\").length incl. trailing newline; +32 = session-JSON paste branch for handleManualSubmit plus a shared submitCodexAccessToken() helper extracted from the pre-existing bare-JWT branch, mirroring the #5203 oauthBlobSubmit.ts extraction precedent; the normalizer logic itself lives in the new src/lib/oauth/utils/codexSessionImport.ts leaf module, not here). Fourth irreducible wiring bump on this modal (969->989->993->998->1030); structural shrink tracked in #3501.",
+    "_rebaseline_2026_07_19_7546_ghe_copilot_modal": "PR #7546 (GHE Copilot OAuth provider) own growth: OAuthModal.tsx 1030->1056 (gate units). Adds a gheUrl input state, routes ghe-copilot through the existing device-code branch, and threads gheUrl into the device-code request/poll extraData at the existing provider-switch chokepoints (+~24 lines, cohesive with the same pattern as #7399/#6636). The standalone GHE enterprise-URL config step JSX (originally +31 lines inline) was extracted to the new src/shared/components/oauthModal/GheConfigStep.tsx leaf component to minimize the bump; what remains is the irreducible provider-branch wiring. Fifth bump on this modal (969->989->993->998->1030->1056); structural shrink tracked in #3501.",
+    "_rebaseline_2026_07_19_7787_ic2_localdb_reexports": "PR #7787 (IC2 raw connections cache + lazy-decrypt) own growth: localDb.ts 805->807 (gate units, +2). localDb.ts is the re-export-only layer (hard rule #2 — no logic); the PR adds 4 new db/readCache re-exports (touchConnectionLastUsed, getCachedRawProviderConnections, getCachedProviderConnectionById, getCachedProviderNodes) required by existing barrel importers. Irreducible for a re-export list; frozen so it can only shrink.",
+    "_rebaseline_2026_07_20_7779_routingcombo_thread": "PR #7779 own growth: chatHelpers.ts 876->877 (+1, thread routingComboId into executeChatWithBreaker for compression-combo assignment). Frozen so it can only shrink.",
+    "_rebaseline_2026_07_20_7819_autocandidateoverrides_reexport": "PR for #7819 (Level 1+2: read-only auto/* candidate transparency + per-API-key exclusions) own growth: localDb.ts 807->808 (+1). Adds a single `export * from \"./db/autoCandidateOverrides\"` barrel re-export (hard rule #2 — no logic) for the new DB module backing per-apiKey candidate exclusions. Irreducible for a re-export list; frozen so it can only shrink.",
+    "_rebaseline_2026_07_21_8027_grok_cli_auth_json_paste": "PR #8027 (RaviTharuma, fix(grok-cli) #7610) own growth: OAuthModal.tsx 1080->1100 (gate units). Requires the full ~/.grok/auth.json (with refresh_token) on the paste-import path instead of a bare JWT, at the existing paste-token chokepoint (renamed tab label, updated instructions/placeholder, textarea for the auth.json blob, inline error surface). The validation logic itself (parseGrokCliPasteToken, previously an inline ~75-line function) was extracted to the new src/lib/oauth/utils/grokCliAuthJson.ts leaf module — mirroring the #6636/#7546 extraction precedent — so only the irreducible UI wiring remains here. Sixth bump on this modal (969->989->993->998->1030->1056->1100); structural shrink tracked in #3501.",
+    "_rebaseline_2026_07_21_8034_compression_exclusions_sidebar": "#8034 (compression exclusions dashboard tab) own growth: sections.ts 796->806 (+10, one new COMPRESSION_CONTEXT_GROUP sidebar item linking /dashboard/compression/exclusions). The file was already 796/800 before this PR (organic growth from prior sidebar entries), so a single new nav item pushed it 6 lines over cap. Freezing at 806 (cannot grow further); the sidebar item array is data, not extractable logic.",
+    "_rebaseline_2026_07_22_7936_namespace_roundtrip": "#7936 (@RCrushMe, Responses-Chat namespace round-trip identity seam) own growth: open-sse/translator/response/openai-responses.ts 1092->1125 (+33) and open-sse/utils/stream.ts 2814->2869 (+55) — threading the namespace-identity seam through the Responses↔Chat translation + stream paths so tool-call namespaces survive the round-trip. Cohesive translation/stream wiring at existing chokepoints, frozen at new size.",
+    "_rebaseline_2026_07_22_8010_codex_responses_engine": "PR #8010 (@JxnLexn) own growth: open-sse/mcp-server/schemas/tools.ts 1497->1505 (+8 = threading the new \"codex-responses\" literal into the compressionConfigureInput strategy/autoTriggerMode Zod enums and setCompressionEngineInput engine enum, mirroring the existing rtk/omniglyph enum entries; no new tool). open-sse/services/compression/strategySelector.ts 1043->1054 (+11 = one new `if (mode === \"codex-responses\")` dispatch branch in runCompression that delegates 100% to the new codexResponsesEngine.apply, mirroring the existing rtk single-mode dispatch, plus threading config.codexResponsesConfig.preserveToolNames into the shared adaptBodyForCompression call at the 3 existing call sites). src/lib/db/compression.ts (untracked, new-file cap 800) 794->845 (+51 = normalizeCodexResponsesConfig, mirroring the existing normalizeRtkConfig normalizer, plus registering \"codex-responses\" in the COMPRESSION_MODES/STACKED_PIPELINE_ENGINE_IDS/SINGLE_MODE_ENGINE sets and the getCompressionSettings load/save switch) — added to the baseline at its current size. All three are cohesive dispatch/normalizer wiring at existing chokepoints (mirroring the prior compression-mode rebaselines #6534/#6556), not extractable without hiding the mode-dispatch boundary. Covered by tests/unit/compression/codex-responses.test.ts (6) + omniglyph-registries.test.ts/types.test.ts (22, updated for the new mode).",
+    "_rebaseline_2026_07_22_8034_compression_exclusions_persistence": "#8034 (compression exclusions) own growth: src/lib/db/compression.ts 845->850 (+5 = threading the new compressionExclusions field through the existing getCompressionSettings/saveCompressionSettings load/save switch over the shared key_value compression namespace — no new table, no raw SQL). Mirrors the prior compression-field rebaselines (#8010 codex-responses normalizer at the same chokepoint); the load/save switch is a single dispatch boundary, not extractable without hiding it. Covered by the PR's 8 node:test + 3 vitest cases.",
+    "_rebaseline_2026_07_22_8050_model_lockout_exact_family": "#8050 (@AndrianBalanescu) own growth: accountFallback.ts 1864->1892 (+28) — exact-vs-family model-lockout scoping (getModelLockKey/isModelLocked/clearModelLock/getModelLockoutInfo) so an Antigravity 404 for one bare model no longer hijacks the whole family cooldown. Cohesive lockout logic; frozen at new size.",
+    "_rebaseline_2026_07_22_8056_headroom_minrows": "#8056 (@RaviTharuma, persist Headroom minRows) own growth: src/lib/db/compression.ts 850->866 (+16 HeadroomConfig+DEFAULT_HEADROOM_CONFIG+normalize/store in get/updateCompressionSettings) and open-sse/services/compression/strategySelector.ts 1054->1060 (+6 merge settings.headroom into stacked stepConfig). Cohesive settings-persistence + stacked-merge wiring at existing chokepoints, frozen at new size.",
+    "_rebaseline_2026_07_22_8081_reasoning_placeholder_guard": "#8081 (@Dingding-leo) own growth: openai-responses.ts 1125->1137 (+12) restructuring the reasoning-placeholder guard so it skips only the empty content block and still emits finish_reason/tool_calls in the same chunk. Cohesive translator wiring; frozen at new size.",
+    "_rebaseline_2026_07_22_8210_openrouter_midstream_error": "PR #8210 (hartmark, fix/openrouter-midstream-error-surfacing) own growth: open-sse/translator/response/openai-responses.ts 1137->1163 (+26) measured on the merged tip (release 1137 + this PR own growth). Adds a single new branch inside openaiToOpenAIResponsesResponse() that detects an OpenRouter-style mid-stream aggregator error (HTTP 200 SSE chunk with empty choices + a top-level error object) and surfaces it as state.upstreamError instead of silently falling through to the no-op/awaitingTrailingUsage path, which previously masked the failure as a false empty-success completion and skipped combo fallback. Irreducible call-site addition at the existing chunk-dispatch chokepoint (mirrors the Gemini-to-OpenAI translator's #4177 precedent for the same class of upstream error surfacing). Note: this baseline entry does NOT cover the separate pre-existing +11 drift already on the release tip from #8081/#8162 (1125->1136, unrelated reasoning-placeholder-stripping fix merged after this PR branched) — that drift belongs to the maintainer's rebaseline, not this PR.",
+    "_rebaseline_2026_07_22_8211_gemini_malformed_tool_choice": "PR #8211 (hartmark, fix/gemini-malformed-function-call-tool-choice) own growth: open-sse/translator/response/gemini-to-openai.ts 771->821 (+50, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds MALFORMED_FUNCTION_CALL/UNEXPECTED_TOOL_CALL handling inside geminiToOpenAIResponse(): synthesizes a `malformed_tool_call` tool_calls entry so finish_reason normalizes to the standard \"tool_calls\" instead of an unrecognized raw enum value that OpenAI-compatible clients (e.g. OpenClaw) silently ignore, and always synthesizes (rather than skipping when a real tool call already exists) so a malformed attempt alongside a real one in the same turn is not silently discarded. Irreducible cohesive addition at the existing candidate/finishReason translation chokepoint (mirrors the 9router#2462 raw-finish-reason precedent immediately below it in the same function). Covered by the PR's own tests/unit test additions for both the malformed-only and malformed-plus-real-call cases.",
+    "_rebaseline_2026_07_22_8213_chat_abandoned_target_abort": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/sse/handlers/chat.ts 1794->1860 (+66, measured against the PR's own merge-base — the release tip separately carries an unrelated -5 net shrink from #8013's antigravity callable-catalog alignment, which this PR's branch does not include and this entry does not cover). Adds resolveDispatchClientRawRequest(): merges a per-target modelAbortSignal into clientRawRequest.signal (via mergeAbortSignals) so a combo target abandoned by comboTargetTimeoutMs actually observes its own abort and reaches its cleanup path, instead of hanging forever inside withRateLimit/acquireAccountSemaphore and leaking a permanent 'pending' dashboard entry (live incident, log id 1784418258231-14961a). Also wires combo-exhausted rejection logging to capture request body + attempted models via the new rejectedRequestUsage helper. Irreducible additions at the existing chat dispatch chokepoint. Covered by the PR's own combo-config + integration test additions.",
+    "_rebaseline_2026_07_22_8213_combo_cooldown_wait_recording": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/combo.ts 3548->3604 (+56, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Fixes combo cooldown-wait state recording so a bogus 503 is no longer crystallized when the cooldown-wait vars reset every setTry, adds an OpenAI-format SSE error frame path for combo-exhausted rejections (capturing request body + attempted models), and gives an abandoned per-target dispatch its own timeout instead of leaking a permanent 'pending' dashboard entry. Irreducible additions at the existing handleComboChat dispatch/retry chokepoint (mirrors the prior quota-share/headroom/task-aware strategy-branch precedents already frozen in this file). Covered by the PR's own combo-config + Gemini TPM-ceiling benchmark test additions.",
+    "_rebaseline_2026_07_22_8213_gemini_tpm_quota_cooldown_wait": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/accountFallback.ts 1857->1932 on the merged tip (release 1892 incl #8050 +35, plus this PR own growth +40); measured against the PR own merge-base was 1857->1898 (+41 — the release tip separately carries an unrelated +34 from #8050's antigravity 404 model-not-found lockout scoping, which this PR's branch does not include and this entry does not cover). Own growth is the Gemini TPM-ceiling classification + cooldown-wait wiring feeding into the combo cooldown-wait state machine (rate-limit wedge recovery) introduced by this PR's commit series. Irreducible additions at the existing account-fallback/model-lockout chokepoint. Covered by the PR's own gemini-rate-limit-tracker and TPM-ceiling benchmark test additions.",
+    "_rebaseline_2026_07_22_8213_health_unblock_model_cooldowns": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/app/(dashboard)/dashboard/health/page.tsx 1094->1165 (+71, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds handleUnblockAll/handleUnblockOne dashboard actions (DELETE /api/resilience/model-cooldowns) so an operator can manually clear a Gemini TPM-wedge model lockout surfaced by this PR's cooldown-wait fixes, instead of waiting out the ceiling. Irreducible UI wiring at the existing health-page action chokepoint. Covered by the PR's own dashboard/resilience test additions.",
+    "_rebaseline_2026_07_22_8213_requestloggerdetail_unblock_ui": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/shared/components/RequestLoggerDetail.tsx 799->941 (+142, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip; crosses the general 800-line new-file cap so is frozen here for the first time). Adds a collapsible section header (open/expand-less toggle) plus per-log-entry unblock (`unblocking`/`cleared` state, isCombo503 detection) so the request-logger detail panel surfaces the same Gemini TPM cooldown-wait / model-lockout unblock action introduced by this PR at the individual-request level (mirrors the health-page bulk unblock action added in the same PR). Covered by the PR's own dashboard/resilience test additions.",
+    "_rebaseline_2026_07_22_fusion_8013_8098_antigravity": "Fusion of #8013 (backryun, catalog/IDE-CLI-split rewrite) + #8098 (nguyenha935, protocol-fidelity/fail-closed/credits/tool-cloaking): open-sse/services/usage/antigravity.ts NEW 802 (>cap 800, +2 — #8098 credits/tier usage service on #8013's profile-aware headers). Test growth (models-catalog-route 1605->1608, provider-models-route 1752->1757 from #8013 Gemini 3.6 catalog) tracked in testFrozen.",
+    "_rebaseline_2026_07_23_8127_grok_weekly_quota": "#8127 (@apoapostolov) own growth: src/sse/handlers/chat.ts 1861->1865 (+4) — weekly quota tracking for grok-web wires a quota-fetch hook at the existing dispatch chokepoint. Thin wiring mirroring adjacent provider-quota branches; not extractable. Covered by tests/unit/grok-quota-fetcher.test.ts.",
+    "_rebaseline_2026_07_23_8143_empty_catch_logging": "#8143 (@chirag127) own growth: open-sse/utils/stream.ts 2869->2887 (+18) — replacing empty catch blocks in the SSE stream subsystem with console.debug logging (Rule #6 silent-swallow fix, issues #8138-#8142). Cohesive logging additions at the existing catch chokepoints, not extractable; frozen at new size. Covered by tests/unit/stream-handler-catch-logging-8143.test.ts.",
+    "_rebaseline_2026_07_23_8219_cache_ttl_settings_sidebar": "#8219 (@oyi77) own growth: sections.ts 806->813 (+7) — configurable model-catalog cache-TTL settings adds a new sidebar nav entry + its visibility wiring. Sidebar item array is data, not extractable logic; frozen at new size.",
+    "_rebaseline_2026_07_23_8247_8248_model_unhealthy": "#8247+#8248 own growth: accountFallback.ts 1940->1941 (+1, irreducible import statement only — the substantive #8248 DEGRADED-pattern classifier was extracted into open-sse/config/errorConfig.ts, which has ample headroom, instead of growing this frozen file; #8247's fix is a single existing-line condition change, net zero lines). Scoping the credits-exhausted 403/429 branch to isCompatibleProvider() (per-model-quota openai/anthropic-compatible-* nicknames) so it stays model-scoped instead of terminalling the whole connection, and classifying NVIDIA NIM 'Function ... DEGRADED' 400 bodies as model-access-denied instead of a raw passthrough 400. Covered by tests/unit/8247-accountfallback-model-unhealthy.test.ts and tests/unit/8248-accountfallback-nvidia-degraded.test.ts.",
+    "_rebaseline_2026_07_23_8252_combo_400_advance": "#8252 (@RaviTharuma) own growth: accountFallback.ts 1932->1940 (+8) + combo.ts 3604->3630 (+26) — advance combo on model-scoped 400s wrapped as invalid/Bad-Request. Irreducible wiring at existing account-fallback + combo dispatch chokepoints. Covered by combo-model-scoped-400-advance.test.ts.",
+    "_rebaseline_2026_07_23_8266_alibaba_media": "#8266 (@backryun) own growth: imageRegistry.ts 821->979 (+158) — Alibaba-family media models (Qwen image/video, Bailian, Wan) added to the image/video registry. Registry model data, not extractable logic; frozen at new size.",
+    "_rebaseline_2026_07_24_8388_compression_detail_persist": "#8388 (compression engine DETAIL settings — Headroom/session-dedup/CCR — dropped on save) own growth: src/lib/db/compression.ts 866->872 (+6 = irreducible call-site wiring at the existing getCompressionSettings/updateCompressionSettings chokepoint: one import line, one `...buildDetailConfigDefaults()` spread in the seed config, and one `case \"sessionDedup\": case \"ccr\": applyDetailConfigUpdate(config, key, parsed); break;` load-switch case, mirroring the existing headroom/#8056 case immediately above it). The actual normalizer logic (normalizeSessionDedupConfig/normalizeCcrConfig, matching SESSION_DEDUP_SCHEMA/CCR_SCHEMA bounds) was EXTRACTED into a new leaf src/lib/db/compressionDetailNormalizers.ts (well under cap) so this frozen file only carries the minimal dispatch wiring. Covered by tests/unit/8388-compression-detail-persist.test.ts (schema-accept + full DB save->reload round-trip for both new sub-objects, plus a no-regression assertion on the existing headroom round-trip).",
+    "_rebaseline_2026_07_24_responses_toolcalls_log_summary": "hartmark, fix/responses-tool-calls-log-summary own growth: open-sse/translator/response/openai-responses.ts 1163->1174 (+11). closeToolCall() now also writes the completed tool call into the shared state.toolCalls Map (already populated by the openai-to-claude / claude-to-openai / gemini-to-openai response translators) so stream.ts's completion-log summary builder (which reads state.toolCalls, not this translator's own funcCallIds/funcNames/funcArgsBuf bookkeeping) reports finish_reason \"tool_calls\" and message.tool_calls for openai->openai-responses translated streams instead of always logging \"stop\" with no tool_calls — the actual client-facing SSE events were already correct; only the persisted call-log summary was wrong. Irreducible call-site addition at the existing tool-call-close chokepoint. Covered by the new regression test in tests/unit/translator-resp-openai-responses.test.ts.",
+    "_rebaseline_2026_07_25_8476_combo_input_bound_homogeneous_scope": "PR #8476 (herjarsa, fix/8375-8459-combo-image-fixes, #8375) own growth: open-sse/services/combo.ts 3642->3679 (+37 net: +29 the PR's own isInputBoundFailure short-circuit for deterministic context_length_exceeded/context_window_exceeded failures, +8 a /green-prs pre-merge fix scoping that short-circuit to homogeneous remainders only — the shipped code fired unconditionally on ANY target, regressing the intentional heterogeneous-combo fallback #6637/isContextOverflow400 protects, exactly as flagged by this PR's own review evidence but never actually implemented in the branch). The fix compares orderedTargets[i+1..] modelStr against the failing target's modelStr at the existing executeTarget dispatch chokepoint (mirrors the sameProviderNext precedent a few lines below) — irreducible call-site wiring, not extractable without hiding the dispatch boundary. Covered by tests/unit/combo-input-bound-failure-8375.test.ts (homogeneous pool still short-circuits) and the new tests/unit/combo-input-bound-heterogeneous-8375.test.ts (heterogeneous combo now correctly falls through to the larger-context target).",
+    "_rebaseline_2026_07_25_adobe_firefly_reference_images": "Follow-up to #8006: storage upload + referenceBlobs for image/video and /v1/images/edits dispatch. adobeFireflyClient.ts 1958->2317 (+upload helpers, extract sources, resolve blob ids). Note: 2317 not 2316 — check-file-size.mjs counts LOC via split(\"\\n\").length (counts the trailing-newline empty element), which is 1 higher than `wc -l` on a file ending in \\n; the PR's original entry (2316) was measured with wc -l and undercounted by 1 against the actual gate.",
+    "_rebaseline_pr1043_minimax_tts": "Upstream port decolua/9router#1043 (toanalien) own growth: audioSpeech.ts 965->1061 (+96). Adds MiniMax T2A v2 TTS dispatch (handleMinimaxSpeech + hexToBytes helper) — provider entry was already in audioRegistry (format: minimax-tts) but no handler existed, falling through to the OpenAI-compatible default that fails (T2A has custom shape + hex-encoded audio + base_resp envelope). New branch sits next to the other inline provider branches (xiaomi-mimo, coqui, tortoise, aws-polly) — extracting would just create indirection. Covered by tests/unit/minimax-tts-1043.test.ts (3 tests, GREEN: success, base_resp error, invalid-hex).",
+    "_rebaseline_pr4592_exclude_exhausted_auto": "Reconcile #4592 already-merged growth: combo.ts 2991->3036 (+45, terminal-status quota-cutoff exclusion in buildAutoCandidates + opt-in gate). Fast-gate PR->release does not run check:file-size.",
+    "open-sse/executors/antigravity.ts": 1528,
+    "open-sse/executors/base.ts": 1640,
+    "open-sse/executors/chatgpt-web.ts": 3241,
+    "open-sse/executors/codex.ts": 1562,
+    "open-sse/executors/cursor.ts": 1563,
+    "open-sse/executors/deepseek-web.ts": 1148,
+    "open-sse/executors/grok-web.ts": 1044,
+    "open-sse/executors/muse-spark-web.ts": 1405,
+    "open-sse/handlers/chatCore.ts": 5034,
+    "open-sse/handlers/imageGeneration.ts": 3101,
+    "open-sse/handlers/responseSanitizer.ts": 1128,
+    "open-sse/handlers/search.ts": 1536,
+    "open-sse/handlers/videoGeneration.ts": 1063,
+    "open-sse/mcp-server/schemas/tools.ts": 1553,
+    "open-sse/mcp-server/server.ts": 1448,
+    "open-sse/mcp-server/tools/advancedTools.ts": 1120,
+    "open-sse/services/accountFallback.ts": 1978,
+    "open-sse/services/adobeFireflyClient.ts": 2385,
+    "open-sse/services/claudeCodeCompatible.ts": 1202,
+    "open-sse/services/combo.ts": 3648,
+    "open-sse/services/compression/strategySelector.ts": 1060,
+    "open-sse/services/rateLimitManager.ts": 1167,
+    "open-sse/translator/response/openai-responses.ts": 1204,
+    "open-sse/utils/cursorAgentProtobuf.ts": 1505,
+    "open-sse/utils/stream.ts": 2889,
+    "src/app/(dashboard)/dashboard/HomePageClient.tsx": 1388,
+    "src/app/(dashboard)/dashboard/analytics/ComboHealthTab.tsx": 1031,
+    "src/app/(dashboard)/dashboard/api-manager/ApiManagerPageClient.tsx": 3117,
+    "src/app/(dashboard)/dashboard/cache/media/MediaPageClient.tsx": 1067,
+    "src/app/(dashboard)/dashboard/combos/page.tsx": 4703,
+    "src/app/(dashboard)/dashboard/costs/CostOverviewTab.tsx": 1283,
+    "src/app/(dashboard)/dashboard/costs/quota-share/components/PoolWizard.tsx": 1022,
+    "src/app/(dashboard)/dashboard/endpoint/EndpointPageClient.tsx": 2615,
+    "src/app/(dashboard)/dashboard/health/page.tsx": 1165,
+    "src/app/(dashboard)/dashboard/providers/[id]/components/modals/EditConnectionModal.tsx": 1324,
+    "src/app/(dashboard)/dashboard/providers/page.tsx": 1944,
+    "src/app/(dashboard)/dashboard/runtime/RuntimePageClient.tsx": 1201,
+    "src/app/(dashboard)/dashboard/settings/components/PricingTab.tsx": 1019,
+    "src/app/(dashboard)/dashboard/settings/components/ProxyRegistryManager.tsx": 1470,
+    "src/app/(dashboard)/dashboard/settings/components/ResilienceTab.tsx": 1123,
+    "src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx": 1629,
+    "src/app/(dashboard)/dashboard/settings/components/SystemStorageTab.tsx": 1573,
+    "src/app/(dashboard)/dashboard/usage/components/BudgetTab.tsx": 1028,
+    "src/app/(dashboard)/dashboard/usage/components/EvalsTab.tsx": 2148,
+    "src/app/(dashboard)/dashboard/usage/components/ProviderLimits/index.tsx": 1119,
+    "src/app/api/providers/[id]/models/route.ts": 2361,
+    "src/app/api/v1/models/catalog.ts": 1590,
+    "src/lib/tokenHealthCheck.ts": 1053,
+    "src/lib/db/apiKeys.ts": 1529,
+    "src/lib/db/core.ts": 1639,
+    "src/lib/db/migrationRunner.ts": 1094,
+    "src/lib/db/models.ts": 1097,
+    "src/lib/db/providers.ts": 1034,
+    "src/lib/memory/retrieval.ts": 1073,
+    "src/lib/tailscaleTunnel.ts": 1202,
+    "src/lib/usage/providerLimits.ts": 1013,
+    "src/shared/components/OAuthModal.tsx": 1134,
+    "src/shared/components/RequestLoggerV2.tsx": 1629,
+    "src/shared/components/analytics/charts.tsx": 1035,
+    "src/shared/services/cliRuntime.ts": 1122,
+    "src/sse/handlers/chat.ts": 1904,
+    "src/sse/services/auth.ts": 2508,
+    "tests/unit/account-fallback-service.test.ts": 1572,
+    "tests/unit/provider-validation-specialty.test.ts": 2985,
+    "open-sse/executors/hyperagent.ts": 1026,
+    "open-sse/executors/default.ts": 1042,
+    "open-sse/executors/kiro.ts": 1069,
+    "open-sse/translator/request/openai-to-kiro.ts": 1057
+  },
   "testCap": 1000,
   "testFrozen": {
     "_rebaseline_2026_06_27_5193_antigravity_test": "#5193 own test growth: oauth-providers-config.test.ts 870->873 (+3: antigravity projectId assertion + 50ms tick for the now fire-and-forget onboarding, matching the no-PKCE/no-openid flow).",
@@ -283,134 +413,7 @@
   "_rebaseline_2026_07_27_3850_relax_filesize_cap_v2_20pct": "OWNER-APPROVED TEMPORARY relax for v3.8.50-3.8.54 PREPARE phase (docs/ROADMAP.md). v1 was cap 800->900 / testCap 800->900 on 2026-07-27; v2 = v1 +20% buffer = cap 900->1000 (+100), testCap 900->1000 (+100). Justification: same as complexity v2 — the v3.8.50 release cut coincides with high-merge activity; owner accepted enlarging the headroom to cover the entire PREPARE phase (5 minor cycles .50-.54) without per-PR rebaseline noise. Targets: decompose-existing-frozen unchanged (frozen still only-shrink — see frozen[] entries and the 105 files >900 that still need structural decomposition regardless of cap); this only relaxes the cap for NEW files in the decompose/extract-while-PREPARE phase (.51='executor registry in-place' and .52='combo.ts decomposition' create new leaf modules above 800). RE-TIGHTENING MANDATORY in v3.8.51: cap target 850 = 850 once decomposition wave stabilizes (gives 150 units of post-tighten headroom vs the new 1000 ceiling). Tracked via same roadmap issue as complexity v2. Window: v3.8.50 (release cut) → v3.8.54 close (RE-TIGHTEN at v3.8.51 prep merge per ROADMAP.md). Last entry unless measured regression. v1 entry retained below for audit trail.",
   "_rebaseline_2026_07_27_3850_relax_filesize_cap": "OWNER-APPROVED TEMPORARY relax for v3.8.50-3.8.54 PREPARE phase (docs/ROADMAP.md). cap 800->900 (+100), testCap 800->900 (+100). Targets: decompose-existing-frozen unchanged (frozen still only-shrink); this only relaxes the cap for NEW files in the decompose/extract-while-PREPARE phase (.51='executor registry in-place' and .52='combo.ts decomposition' create new leaf modules above 800). RE-TIGHTENING MANDATORY in v3.8.51: cap target 850 = 850 once decomposition wave stabilizes. SUPERSEDED by _rebaseline_2026_07_27_3850_relax_filesize_cap_v2_20pct (v1 +20% buffer) — retained for audit. Tracked via same roadmap issue.",
   "_rebaseline_2026_07_27_v3849_train1h": "Merge-train 1H (31 PRs) — owner-approved 2026-07-27. Two distinct causes, kept separate on purpose: (1) GENUINE irreducible growth at existing chokepoints — providerLimits/auth (#8632 Kimi quota-reset recovery), rateLimitManager (#8616 idle wedged limiters), models-catalog-route.test (#8610 OpenCode Go effort aliases); (2) COLLISION with #8585, which banked shrinks measured on the pre-train release tip while 30 sibling PRs in the SAME train grew those files again — chat/accountFallback (#8628), chatCore (#8613), videoGeneration (#8581), imageGeneration. The zero-headroom frozen entries cannot absorb either. Ceilings re-pinned to the post-merge tip; #8612 (also in this train) automates shrink-banking so this self-inflicted drift stops recurring. Detail: src/lib/usage/providerLimits.ts 1006->1013 (#8632); src/sse/services/auth.ts 2492->2508 (#8632); open-sse/services/rateLimitManager.ts 1014->1060 (#8616); src/sse/handlers/chat.ts 1842->1845 (#8628); open-sse/handlers/chatCore.ts 4939->4955 (#8613); open-sse/handlers/imageGeneration.ts 3100->3101 ((sem PR — teto do #8585)); open-sse/handlers/videoGeneration.ts 1038->1063 (#8581); open-sse/services/accountFallback.ts 1965->1966 (#8628); tests/unit/models-catalog-route.test.ts 1608->1636 (#8610)",
-  "frozen": {
-    "_rebaseline_2026_06_22_4644_deepseek_web_tools": "PR #4644 (BugsBag/robust deepseek-web tool-call parsing): open-sse/executors/deepseek-web.ts 1117->1125 (+8). The new agentic tool-call path emits surrounding text + reasoning before tool_calls and swaps to the dedicated deepseekWebTools.ts parser; the +8 lines are cohesive wiring at the existing transformSSE chokepoint (the parser itself lives in the new deepseekWebTools.ts file, already under cap). The PR's own fast-gate (PR->release) does not run check:file-size, so this surfaced only at release reconcile. Covered by tests/unit/deepseek-web-tools-variants.test.ts + deepseek-web-tools-execute.test.ts.",
-    "_rebaseline_2026_06_23_4712_deepseek_web_tool_results": "PR for #4712 (deepseek-web drops role:tool): open-sse/executors/deepseek-web.ts 1125->1148 (+23). messagesToPrompt() now folds role:\"tool\" results into the single-prompt transcript (recovering the tool name from the preceding assistant tool_calls by tool_call_id) instead of silently dropping them; the lines are cohesive wiring inside the existing function. Covered by tests/unit/deepseek-web-tool-result-prompt-4712.test.ts.",
-    "_rebaseline_2026_06_24_headroom_strategy": "Headroom-aware connection selection (dario technique): combo.ts 3168->3180 (+12 = a new `else if (strategy === \"headroom\")` dispatch branch in handleComboChat that delegates to orderTargetsByHeadroom + its log line, plus the import). The actual logic lives OUT of the god-file: the pure ranker rankByHeadroom/computeHeadroom is the new leaf open-sse/services/combo/headroomRanking.ts (91 LOC, <cap) and the async orderer orderTargetsByHeadroom is appended to the existing open-sse/services/combo/quotaStrategies.ts (<cap) next to its sibling reset-aware/reset-window orderers (reuses their connection-expansion machinery). headroom = 1 - max(util_5h, util_7d) from getSaturation (src/lib/quota/saturationSignals.ts), prefers the connection with the most free capacity. Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. fill-first stays default; all existing strategies untouched. Covered by tests/unit/combo-headroom-ranking.test.ts (pure helper) + tests/unit/combo-headroom-strategy.test.ts (orderer, saturation injected). Structural shrink of combo.ts tracked in #3501.",
-    "_rebaseline_2026_06_24_quota_share_strategy": "Dedicated quota-share strategy (Phase 3 #9): combo.ts 3180->3190 (+10 = one new `else if (strategy === \"quota-share\")` dispatch branch in handleComboChat that delegates 100% to selectQuotaShareTarget + its log line, plus the import). All the new logic lives OUT of the god-file in two new leaves under open-sse/services/combo/: quotaShareInflight.ts (in-flight counter with TTL/lease, ~150 LOC <cap) and quotaShareStrategy.ts (per-model bucket gating via isBucketSaturated + DRR proportional to weight + P2C over in-flight, ~240 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the headroom/reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. ZERO existing strategy cases were modified — only this branch was added, and the qtSd/ combos switched from fill-first to quota-share in src/lib/quota/quotaCombos.ts. Covered by tests/unit/quota-share-strategy.test.ts (gating, DRR fairness, P2C in-flight, fail-open, activation). Structural shrink of combo.ts tracked in #3501.",
-    "_rebaseline_2026_06_24_task_aware_routing": "Task-aware routing strategy (port PR #2045, OmniRoute #4945): combo.ts 3190->3225 (+35) = one new `else if (strategy === \"task-aware\")` dispatch branch delegating 100% to selectTaskAwareTarget + its imports/log lines. All scoring/classification logic lives OUT of the god-file in the new leaf open-sse/services/taskAwareRouting.ts (553 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors quota-share/headroom/reset-aware branches). ZERO existing strategy cases modified. Covered by tests/unit/combo-task-aware.test.ts (35 tests). Structural shrink of combo.ts tracked in #3501.",
-    "_rebaseline_2026_06_26_fidelity_gate_extraction": "Milestone-B fidelity-gate wiring residual: bodyToText+gateAdvance extracted to fidelityGateStep.ts (889->854, -35), but the StackOptions.fidelityGate field, the `const fidelityGate` reads at the two stacked-loop dispatch chokepoints, and the import of FidelityGateConfig are irreducible wiring that cannot leave strategySelector without an architectural refactor of the pre-existing stacked pipeline. Net: 889->854 (+6 vs the pre-Milestone-B frozen 848). Covered by tests/unit/compression/*.test.ts (940 pass).",
-    "_rebaseline_2026_06_27_5193_5203_antigravity_oauthmodal": "Antigravity remote-login own growth: OAuthModal.tsx 960->969 (gate units). #5193 (+~4: remote paste instruction shown for all remote incl. Google + its rationale comment) and #5203 (+~5: handleManualSubmit credential-blob branch + button guard; submit logic extracted to oauthBlobSubmit.ts to minimize). Frozen set to the SUM so either merge order passes. Cohesive at the existing manual-submit chokepoint.",
-    "_rebaseline_2026_06_27_5193_antigravity_basered": "Base-red (pre-existing release drift, fast-gate PR->release skips check:file-size): accountFallback.ts 1773->1777 and src/app/api/providers/[id]/test/route.ts 924->940 were already over their frozen caps on release/v3.8.39 independent of any antigravity change. Owner chose to rebaseline (keep the documented issue-reference comments #1846/#1449/#347 etc.) rather than accept the contributor comment-stripping in #5200/#5198. Reverted #5200 to restore the comments; bumped these two frozen caps to the actual base sizes. No logic change.",
-    "_rebaseline_2026_06_28_5237_impersonation_ua_refresh": "PR #5237 (refresh impersonation UAs): grok-web.ts 1871->1873 (+2), muse-spark-web.ts 1284->1302 (+18), perplexity-web.ts 1013->1032 (+19). Net semantic change in each file is a single User-Agent constant (Chrome 147->149 for grok/muse; perplexity kept at Firefox 148 to stay matched with the firefox_148 TLS profile — the contributor's 152 bump was reverted to avoid a UA-vs-JA3 mismatch, #2459). The growth is Prettier reflow that lint-staged unavoidably applies to these grandfathered long-line files the moment they are touched; not extractable. src/sse/services/auth.ts 2336->2401 in the same reconcile is #5222's antigravity-LRU-retry growth that merged via --admin without a baseline bump.",
-    "_rebaseline_2026_06_28_5243_risk_gate_prepass": "PR #5243 (compression risk-gate pre-pass) own growth: open-sse/services/compression/strategySelector.ts 854->899 (+45). The three exported entry points (applyCompression/applyStackedCompression/applyStackedCompressionAsync) become thin wrappers over pure-extracted private bodies (runCompression/runStackedCompression/runStackedCompressionAsync) so the risk-gate mask->run->restore wrapper sits strictly OUTSIDE the per-step loop — a single universal integration point. The wrapper logic itself (resolveRiskGate/withRiskGate) lives in the new riskGate/strategyWrap.ts (<cap); the residual growth is the duplicated thin-wrapper signatures + the extracted bodies' dispatch boundary, guarded by a byte-identical parity test (riskGateIntegration). Default off (DEFAULT_COMPRESSION_CONFIG unchanged). Not extractable without hiding the dispatch boundary, mirroring prior compression rebaselines. Structural shrink tracked in #3501.",
-    "_rebaseline_2026_06_28_5275_correlation_id_extract": "Extraction of the safe CorrelationId subset of #5275 (hartmark) — request correlation id stored in call_logs (migration 109) and returned via the X-Correlation-Id response header, WITHOUT the combo/resilience or build/lazy-loading changes (those stay in #5275). Own growth: callLogs.ts 975->985 (correlation_id column on CallLogSummaryRow + read/map), usageHistory.ts 983->988 (correlationId metadata normalize), chat.ts 1575->1632 (withCorrelationId response wiring + combo-failure log carrying correlationId), chatHelpers.ts new 811 (withCorrelationId helper + reqId threading; was 791<cap pre-feature). Cohesive request/logging chokepoint wiring; structural shrink of chat.ts tracked in #3501.",
-    "_rebaseline_2026_06_29_4038_cas_guard": "PR (#4038) own growth: tokenRefresh.ts 2103->2181 (+78 = the compare-and-swap guard on the refresh persist — runWithCasGuard/getActiveCasGuard AsyncLocalStorage pair mirroring runWithOnPersist, casGuardShouldSkipPersist that rereads the row right before persisting and skips the write when a concurrent writer already rotated the refresh_token past the one presented, plus getCasGuardStats counters). Fixes the sibling-rotation-revert → token-family-revocation storm. Gated behind an active guard (opt-in; no guard => byte-identical). Wiring lives at the two persist chokepoints inside getAccessToken; the comparison reuses wasRefreshTokenRotated from refreshSerializer. Not extractable without splitting the refresh hot path.",
-    "_rebaseline_2026_06_29_5286_memoization": "PR #5286 own growth: strategySelector.ts 899->960 (+61 = the opt-in result-memoization branches in applyCompression/applyCompressionAsync — principal+determinism gate, makeMemoKey lookup/store with model+supportsVision folded into the key, recompute-with-memo-off). Default off (memoizeCompressionResults), so zero behavior change. The memo helpers live in the leaf resultMemo.ts (<cap); the chokepoint wiring here is not extractable. Structural shrink of this hot-path file tracked in #3501.",
-    "_rebaseline_2026_07_01_v3843_release_5609": "Rebaseline v3.8.43 (PR #5609 release reconciliation). DRIFT dos 109 commits do ciclo: 8 god-files existentes cresceram (ApiManagerPageClient 2983->3017, combos/page 4594->4608, AddApiKeyModal 868->869, providerPageHelpers 974->996, chat.ts 1635->1647, auth.ts 2401->2403, batchProcessor 828->915, combo.ts 3368->3387) + 2 novos acima do cap (huggingchat.ts 813, tests web-cookie-providers-new 827) + 4 test files cresceram. Modularizacao deferida (blast-radius mid-release); congelado no estado atual p/ o proximo ciclo ratchetar daqui.",
-    "_rebaseline_2026_07_02_5816_qoder": "PR #5816 (@AgentKiller45, qoder PAT via qodercli): qoderCli.ts 666->989, new-above-cap frozen (owner-approved baseline freeze). The growth is the legitimate PAT job-token exchange + quota parsing CLI transport (the pure-JS Cosy path 500'd on every PAT request); extracting the spawn/parse helpers now would just add indirection to a contributor PR mid-merge. Test frozen also raised for this PR's coverage growth: providers-page-utils.test.ts 1052->1092. Additionally clears an inherited base-red from the already-merged #5933 (codex json_schema->text.format): translator-openai-responses-req.test.ts 1097->1172 (+75 regression tests, no offending branch left). All remain frozen (cannot grow further); release captain's rebaseline-at-release supersedes.",
-    "_rebaseline_2026_07_09_6126_clinepass_dual_auth": "PR #6126 (@hajilok, dual-auth ClinePass) own growth: tokenRefresh.ts 2181->2182 (+1 = a single `case \"clinepass\":` fallthrough label added to the existing `case \"cline\":` in _getAccessTokenInternal's provider switch, so clinepass token refresh dispatches to the already-shared refreshClineToken() instead of silently falling through to the generic OAuth refresh). Irreducible 1-line switch-case wiring at the existing chokepoint; the header-building logic for the same feature was extracted to a new leaf src/shared/utils/clineAuth.ts::buildClinepassHeaders() (well under cap) to avoid growing open-sse/executors/default.ts. Covered by tests/unit/clinepass-provider.test.ts.",
-    "_rebaseline_2026_07_09_6363_kiro_external_idp": "PR #6363 (@artickc, Kiro external IdP) own growth: tokenRefresh.ts 2182->2249 (+67 = the external_idp refresh branch inside refreshKiroToken — standard public-client OAuth2 refresh_token grant against the org IdP tokenEndpoint via buildExternalIdpRefreshParams/isExternalIdpAuthMethod from the new leaf open-sse/services/kiroExternalIdp.ts, with invalid_grant/invalid_client -> unrecoverable_refresh_error mapping). Cohesive addition at the existing refreshKiroToken chokepoint. Covered by tests/unit/kiro-external-idp.test.ts.",
-    "_rebaseline_2026_07_09_6587_kiro_api_key_auth": "PR #6587 (@strangersp) own growth for Kiro long-lived API-key auth, merged onto v3.8.47 tip: openai-to-kiro.ts 890->912 (+22, auth-header selection for API-key-vs-OAuth-token connections), providerLimits.ts 998->1000 (+2, API-key auth-type branch), translator-openai-to-kiro.test.ts 1234->1257 (+23), providers-page-utils.test.ts 1109->1107 (net -2 after merging with parallel release drift; connectionMatchesProviderCard api_key coverage added), provider-validation-specialty.test.ts 2856->2980 (+124 net after merge with parallel release drift; this PR also removed the file's `@typescript-eslint/no-explicit-any` eslint-suppression entry by fixing all `any` usages, adding typed replacements). Cohesive additive feature growth, well tested; not extractable without splitting the existing chokepoints mid-merge.",
-    "_rebaseline_2026_07_09_6678_routing_strategy_9router": "#6678 (SeaXen) — 9router-parity Routing Strategy settings card + per-provider/combo sticky-round-robin override. Own growth: ProviderDetailPageClient.tsx 784->786 (single ProviderAccountRoutingCard mount + import), auth.ts 2448->2458 (providerStrategies override resolution: fallbackStrategy/stickyRoundRobinLimit per-provider cascade in getProviderCredentials). Both additive, zero unrelated refactor; new UI/logic lives in new files (ProviderAccountRoutingCard.tsx, RoutingStrategyCard.tsx, rrState.ts::resolveComboStickyRoundRobinLimit). chat.ts value below reflects the current release tip (grown by other concurrent PRs, e.g. #6640), not this PR own change.",
-    "_rebaseline_2026_07_10_6318_omp_letta": "PR #6318 (@hamsa0x7, omp+letta CLI integrations) own growth: cliTools.ts (+53 = 2 registry entries incl. omp docsUrl) and cliRuntime.ts (+18 = runtime-detection wiring for the 2 new tools). Cohesive registry/wiring growth at the existing chokepoints; scope reduced from the original 5 tools (pi/codewhale/jcode shipped separately).",
-    "_rebaseline_2026_07_10_gcf_v3_2_decode": "PR #6838 own growth: new vendored file open-sse/services/compression/engines/headroom/gcf/decode_generic.ts frozen at 880 (> 800 cap). It is the vendored GCF generic-profile decoder (spec v3.2 nested flattening plus the prototype-pollution / hasOwnProperty hardening added in this PR's Gemini review). Kept as one file faithful to upstream gcf-typescript so re-vendoring stays a clean copy rather than a re-split each cycle (sibling generic.ts/scalar.ts stay < cap; extraction would also fragment the file's frozen eslint no-explicit-any suppressions). Round-trip + prototype-pollution regression coverage in tests/unit/compression/headroom-smartcrusher.test.ts. Frozen: only shrinks from here.",
-    "_rebaseline_2026_07_12_v3847_mergeprs_tail": "v3.8.47 /merge-prs tail (owner-approved): src/lib/localDb.ts NEW>800 (799->805, +6 re-exports countFreeProxies + recordFreeProxySyncErrors/clearFreeProxySyncErrors/getFreeProxySyncErrors + FreeProxySyncErrors type for #6909 free-pool relay-repair; re-export-only per Hard Rule #2, not extractable).",
-    "_rebaseline_2026_07_15_7070_combos_memo": "PR #7070 (perf/p1-memo) own growth: src/app/(dashboard)/dashboard/combos/page.tsx 4655->4656 (+1 = React.memo wrapping of ComboCard). Covered by tests/unit/ui/combos-page-smoke.test.tsx.",
-    "_rebaseline_2026_07_18_7399_xai_oauth_modal": "PR #7399 (xAI OAuth PKCE) own growth: OAuthModal.tsx 993->998 (+5 = provider entry + PKCE flow branch wiring at the existing provider-switch chokepoint; the provider logic itself lives in src/lib/oauth/providers/xai-oauth.ts, new leaf). Third irreducible wiring bump on this modal (969->989->993->998); structural shrink tracked in #3501.",
-    "_rebaseline_2026_07_19_6636_codex_session_json": "#6636 own growth: OAuthModal.tsx 998->1030 (gate units, split(\"\\n\").length incl. trailing newline; +32 = session-JSON paste branch for handleManualSubmit plus a shared submitCodexAccessToken() helper extracted from the pre-existing bare-JWT branch, mirroring the #5203 oauthBlobSubmit.ts extraction precedent; the normalizer logic itself lives in the new src/lib/oauth/utils/codexSessionImport.ts leaf module, not here). Fourth irreducible wiring bump on this modal (969->989->993->998->1030); structural shrink tracked in #3501.",
-    "_rebaseline_2026_07_19_7546_ghe_copilot_modal": "PR #7546 (GHE Copilot OAuth provider) own growth: OAuthModal.tsx 1030->1056 (gate units). Adds a gheUrl input state, routes ghe-copilot through the existing device-code branch, and threads gheUrl into the device-code request/poll extraData at the existing provider-switch chokepoints (+~24 lines, cohesive with the same pattern as #7399/#6636). The standalone GHE enterprise-URL config step JSX (originally +31 lines inline) was extracted to the new src/shared/components/oauthModal/GheConfigStep.tsx leaf component to minimize the bump; what remains is the irreducible provider-branch wiring. Fifth bump on this modal (969->989->993->998->1030->1056); structural shrink tracked in #3501.",
-    "_rebaseline_2026_07_19_7787_ic2_localdb_reexports": "PR #7787 (IC2 raw connections cache + lazy-decrypt) own growth: localDb.ts 805->807 (gate units, +2). localDb.ts is the re-export-only layer (hard rule #2 — no logic); the PR adds 4 new db/readCache re-exports (touchConnectionLastUsed, getCachedRawProviderConnections, getCachedProviderConnectionById, getCachedProviderNodes) required by existing barrel importers. Irreducible for a re-export list; frozen so it can only shrink.",
-    "_rebaseline_2026_07_20_7779_routingcombo_thread": "PR #7779 own growth: chatHelpers.ts 876->877 (+1, thread routingComboId into executeChatWithBreaker for compression-combo assignment). Frozen so it can only shrink.",
-    "_rebaseline_2026_07_20_7819_autocandidateoverrides_reexport": "PR for #7819 (Level 1+2: read-only auto/* candidate transparency + per-API-key exclusions) own growth: localDb.ts 807->808 (+1). Adds a single `export * from \"./db/autoCandidateOverrides\"` barrel re-export (hard rule #2 — no logic) for the new DB module backing per-apiKey candidate exclusions. Irreducible for a re-export list; frozen so it can only shrink.",
-    "_rebaseline_2026_07_21_8027_grok_cli_auth_json_paste": "PR #8027 (RaviTharuma, fix(grok-cli) #7610) own growth: OAuthModal.tsx 1080->1100 (gate units). Requires the full ~/.grok/auth.json (with refresh_token) on the paste-import path instead of a bare JWT, at the existing paste-token chokepoint (renamed tab label, updated instructions/placeholder, textarea for the auth.json blob, inline error surface). The validation logic itself (parseGrokCliPasteToken, previously an inline ~75-line function) was extracted to the new src/lib/oauth/utils/grokCliAuthJson.ts leaf module — mirroring the #6636/#7546 extraction precedent — so only the irreducible UI wiring remains here. Sixth bump on this modal (969->989->993->998->1030->1056->1100); structural shrink tracked in #3501.",
-    "_rebaseline_2026_07_21_8034_compression_exclusions_sidebar": "#8034 (compression exclusions dashboard tab) own growth: sections.ts 796->806 (+10, one new COMPRESSION_CONTEXT_GROUP sidebar item linking /dashboard/compression/exclusions). The file was already 796/800 before this PR (organic growth from prior sidebar entries), so a single new nav item pushed it 6 lines over cap. Freezing at 806 (cannot grow further); the sidebar item array is data, not extractable logic.",
-    "_rebaseline_2026_07_22_7936_namespace_roundtrip": "#7936 (@RCrushMe, Responses-Chat namespace round-trip identity seam) own growth: open-sse/translator/response/openai-responses.ts 1092->1125 (+33) and open-sse/utils/stream.ts 2814->2869 (+55) — threading the namespace-identity seam through the Responses↔Chat translation + stream paths so tool-call namespaces survive the round-trip. Cohesive translation/stream wiring at existing chokepoints, frozen at new size.",
-    "_rebaseline_2026_07_22_8010_codex_responses_engine": "PR #8010 (@JxnLexn) own growth: open-sse/mcp-server/schemas/tools.ts 1497->1505 (+8 = threading the new \"codex-responses\" literal into the compressionConfigureInput strategy/autoTriggerMode Zod enums and setCompressionEngineInput engine enum, mirroring the existing rtk/omniglyph enum entries; no new tool). open-sse/services/compression/strategySelector.ts 1043->1054 (+11 = one new `if (mode === \"codex-responses\")` dispatch branch in runCompression that delegates 100% to the new codexResponsesEngine.apply, mirroring the existing rtk single-mode dispatch, plus threading config.codexResponsesConfig.preserveToolNames into the shared adaptBodyForCompression call at the 3 existing call sites). src/lib/db/compression.ts (untracked, new-file cap 800) 794->845 (+51 = normalizeCodexResponsesConfig, mirroring the existing normalizeRtkConfig normalizer, plus registering \"codex-responses\" in the COMPRESSION_MODES/STACKED_PIPELINE_ENGINE_IDS/SINGLE_MODE_ENGINE sets and the getCompressionSettings load/save switch) — added to the baseline at its current size. All three are cohesive dispatch/normalizer wiring at existing chokepoints (mirroring the prior compression-mode rebaselines #6534/#6556), not extractable without hiding the mode-dispatch boundary. Covered by tests/unit/compression/codex-responses.test.ts (6) + omniglyph-registries.test.ts/types.test.ts (22, updated for the new mode).",
-    "_rebaseline_2026_07_22_8034_compression_exclusions_persistence": "#8034 (compression exclusions) own growth: src/lib/db/compression.ts 845->850 (+5 = threading the new compressionExclusions field through the existing getCompressionSettings/saveCompressionSettings load/save switch over the shared key_value compression namespace — no new table, no raw SQL). Mirrors the prior compression-field rebaselines (#8010 codex-responses normalizer at the same chokepoint); the load/save switch is a single dispatch boundary, not extractable without hiding it. Covered by the PR's 8 node:test + 3 vitest cases.",
-    "_rebaseline_2026_07_22_8050_model_lockout_exact_family": "#8050 (@AndrianBalanescu) own growth: accountFallback.ts 1864->1892 (+28) — exact-vs-family model-lockout scoping (getModelLockKey/isModelLocked/clearModelLock/getModelLockoutInfo) so an Antigravity 404 for one bare model no longer hijacks the whole family cooldown. Cohesive lockout logic; frozen at new size.",
-    "_rebaseline_2026_07_22_8056_headroom_minrows": "#8056 (@RaviTharuma, persist Headroom minRows) own growth: src/lib/db/compression.ts 850->866 (+16 HeadroomConfig+DEFAULT_HEADROOM_CONFIG+normalize/store in get/updateCompressionSettings) and open-sse/services/compression/strategySelector.ts 1054->1060 (+6 merge settings.headroom into stacked stepConfig). Cohesive settings-persistence + stacked-merge wiring at existing chokepoints, frozen at new size.",
-    "_rebaseline_2026_07_22_8081_reasoning_placeholder_guard": "#8081 (@Dingding-leo) own growth: openai-responses.ts 1125->1137 (+12) restructuring the reasoning-placeholder guard so it skips only the empty content block and still emits finish_reason/tool_calls in the same chunk. Cohesive translator wiring; frozen at new size.",
-    "_rebaseline_2026_07_22_8210_openrouter_midstream_error": "PR #8210 (hartmark, fix/openrouter-midstream-error-surfacing) own growth: open-sse/translator/response/openai-responses.ts 1137->1163 (+26) measured on the merged tip (release 1137 + this PR own growth). Adds a single new branch inside openaiToOpenAIResponsesResponse() that detects an OpenRouter-style mid-stream aggregator error (HTTP 200 SSE chunk with empty choices + a top-level error object) and surfaces it as state.upstreamError instead of silently falling through to the no-op/awaitingTrailingUsage path, which previously masked the failure as a false empty-success completion and skipped combo fallback. Irreducible call-site addition at the existing chunk-dispatch chokepoint (mirrors the Gemini-to-OpenAI translator's #4177 precedent for the same class of upstream error surfacing). Note: this baseline entry does NOT cover the separate pre-existing +11 drift already on the release tip from #8081/#8162 (1125->1136, unrelated reasoning-placeholder-stripping fix merged after this PR branched) — that drift belongs to the maintainer's rebaseline, not this PR.",
-    "_rebaseline_2026_07_22_8211_gemini_malformed_tool_choice": "PR #8211 (hartmark, fix/gemini-malformed-function-call-tool-choice) own growth: open-sse/translator/response/gemini-to-openai.ts 771->821 (+50, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds MALFORMED_FUNCTION_CALL/UNEXPECTED_TOOL_CALL handling inside geminiToOpenAIResponse(): synthesizes a `malformed_tool_call` tool_calls entry so finish_reason normalizes to the standard \"tool_calls\" instead of an unrecognized raw enum value that OpenAI-compatible clients (e.g. OpenClaw) silently ignore, and always synthesizes (rather than skipping when a real tool call already exists) so a malformed attempt alongside a real one in the same turn is not silently discarded. Irreducible cohesive addition at the existing candidate/finishReason translation chokepoint (mirrors the 9router#2462 raw-finish-reason precedent immediately below it in the same function). Covered by the PR's own tests/unit test additions for both the malformed-only and malformed-plus-real-call cases.",
-    "_rebaseline_2026_07_22_8213_chat_abandoned_target_abort": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/sse/handlers/chat.ts 1794->1860 (+66, measured against the PR's own merge-base — the release tip separately carries an unrelated -5 net shrink from #8013's antigravity callable-catalog alignment, which this PR's branch does not include and this entry does not cover). Adds resolveDispatchClientRawRequest(): merges a per-target modelAbortSignal into clientRawRequest.signal (via mergeAbortSignals) so a combo target abandoned by comboTargetTimeoutMs actually observes its own abort and reaches its cleanup path, instead of hanging forever inside withRateLimit/acquireAccountSemaphore and leaking a permanent 'pending' dashboard entry (live incident, log id 1784418258231-14961a). Also wires combo-exhausted rejection logging to capture request body + attempted models via the new rejectedRequestUsage helper. Irreducible additions at the existing chat dispatch chokepoint. Covered by the PR's own combo-config + integration test additions.",
-    "_rebaseline_2026_07_22_8213_combo_cooldown_wait_recording": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/combo.ts 3548->3604 (+56, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Fixes combo cooldown-wait state recording so a bogus 503 is no longer crystallized when the cooldown-wait vars reset every setTry, adds an OpenAI-format SSE error frame path for combo-exhausted rejections (capturing request body + attempted models), and gives an abandoned per-target dispatch its own timeout instead of leaking a permanent 'pending' dashboard entry. Irreducible additions at the existing handleComboChat dispatch/retry chokepoint (mirrors the prior quota-share/headroom/task-aware strategy-branch precedents already frozen in this file). Covered by the PR's own combo-config + Gemini TPM-ceiling benchmark test additions.",
-    "_rebaseline_2026_07_22_8213_gemini_tpm_quota_cooldown_wait": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/accountFallback.ts 1857->1932 on the merged tip (release 1892 incl #8050 +35, plus this PR own growth +40); measured against the PR own merge-base was 1857->1898 (+41 — the release tip separately carries an unrelated +34 from #8050's antigravity 404 model-not-found lockout scoping, which this PR's branch does not include and this entry does not cover). Own growth is the Gemini TPM-ceiling classification + cooldown-wait wiring feeding into the combo cooldown-wait state machine (rate-limit wedge recovery) introduced by this PR's commit series. Irreducible additions at the existing account-fallback/model-lockout chokepoint. Covered by the PR's own gemini-rate-limit-tracker and TPM-ceiling benchmark test additions.",
-    "_rebaseline_2026_07_22_8213_health_unblock_model_cooldowns": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/app/(dashboard)/dashboard/health/page.tsx 1094->1165 (+71, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds handleUnblockAll/handleUnblockOne dashboard actions (DELETE /api/resilience/model-cooldowns) so an operator can manually clear a Gemini TPM-wedge model lockout surfaced by this PR's cooldown-wait fixes, instead of waiting out the ceiling. Irreducible UI wiring at the existing health-page action chokepoint. Covered by the PR's own dashboard/resilience test additions.",
-    "_rebaseline_2026_07_22_8213_requestloggerdetail_unblock_ui": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/shared/components/RequestLoggerDetail.tsx 799->941 (+142, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip; crosses the general 800-line new-file cap so is frozen here for the first time). Adds a collapsible section header (open/expand-less toggle) plus per-log-entry unblock (`unblocking`/`cleared` state, isCombo503 detection) so the request-logger detail panel surfaces the same Gemini TPM cooldown-wait / model-lockout unblock action introduced by this PR at the individual-request level (mirrors the health-page bulk unblock action added in the same PR). Covered by the PR's own dashboard/resilience test additions.",
-    "_rebaseline_2026_07_22_fusion_8013_8098_antigravity": "Fusion of #8013 (backryun, catalog/IDE-CLI-split rewrite) + #8098 (nguyenha935, protocol-fidelity/fail-closed/credits/tool-cloaking): open-sse/services/usage/antigravity.ts NEW 802 (>cap 800, +2 — #8098 credits/tier usage service on #8013's profile-aware headers). Test growth (models-catalog-route 1605->1608, provider-models-route 1752->1757 from #8013 Gemini 3.6 catalog) tracked in testFrozen.",
-    "_rebaseline_2026_07_23_8127_grok_weekly_quota": "#8127 (@apoapostolov) own growth: src/sse/handlers/chat.ts 1861->1865 (+4) — weekly quota tracking for grok-web wires a quota-fetch hook at the existing dispatch chokepoint. Thin wiring mirroring adjacent provider-quota branches; not extractable. Covered by tests/unit/grok-quota-fetcher.test.ts.",
-    "_rebaseline_2026_07_23_8143_empty_catch_logging": "#8143 (@chirag127) own growth: open-sse/utils/stream.ts 2869->2887 (+18) — replacing empty catch blocks in the SSE stream subsystem with console.debug logging (Rule #6 silent-swallow fix, issues #8138-#8142). Cohesive logging additions at the existing catch chokepoints, not extractable; frozen at new size. Covered by tests/unit/stream-handler-catch-logging-8143.test.ts.",
-    "_rebaseline_2026_07_23_8219_cache_ttl_settings_sidebar": "#8219 (@oyi77) own growth: sections.ts 806->813 (+7) — configurable model-catalog cache-TTL settings adds a new sidebar nav entry + its visibility wiring. Sidebar item array is data, not extractable logic; frozen at new size.",
-    "_rebaseline_2026_07_23_8247_8248_model_unhealthy": "#8247+#8248 own growth: accountFallback.ts 1940->1941 (+1, irreducible import statement only — the substantive #8248 DEGRADED-pattern classifier was extracted into open-sse/config/errorConfig.ts, which has ample headroom, instead of growing this frozen file; #8247's fix is a single existing-line condition change, net zero lines). Scoping the credits-exhausted 403/429 branch to isCompatibleProvider() (per-model-quota openai/anthropic-compatible-* nicknames) so it stays model-scoped instead of terminalling the whole connection, and classifying NVIDIA NIM 'Function ... DEGRADED' 400 bodies as model-access-denied instead of a raw passthrough 400. Covered by tests/unit/8247-accountfallback-model-unhealthy.test.ts and tests/unit/8248-accountfallback-nvidia-degraded.test.ts.",
-    "_rebaseline_2026_07_23_8252_combo_400_advance": "#8252 (@RaviTharuma) own growth: accountFallback.ts 1932->1940 (+8) + combo.ts 3604->3630 (+26) — advance combo on model-scoped 400s wrapped as invalid/Bad-Request. Irreducible wiring at existing account-fallback + combo dispatch chokepoints. Covered by combo-model-scoped-400-advance.test.ts.",
-    "_rebaseline_2026_07_23_8266_alibaba_media": "#8266 (@backryun) own growth: imageRegistry.ts 821->979 (+158) — Alibaba-family media models (Qwen image/video, Bailian, Wan) added to the image/video registry. Registry model data, not extractable logic; frozen at new size.",
-    "_rebaseline_2026_07_24_8388_compression_detail_persist": "#8388 (compression engine DETAIL settings — Headroom/session-dedup/CCR — dropped on save) own growth: src/lib/db/compression.ts 866->872 (+6 = irreducible call-site wiring at the existing getCompressionSettings/updateCompressionSettings chokepoint: one import line, one `...buildDetailConfigDefaults()` spread in the seed config, and one `case \"sessionDedup\": case \"ccr\": applyDetailConfigUpdate(config, key, parsed); break;` load-switch case, mirroring the existing headroom/#8056 case immediately above it). The actual normalizer logic (normalizeSessionDedupConfig/normalizeCcrConfig, matching SESSION_DEDUP_SCHEMA/CCR_SCHEMA bounds) was EXTRACTED into a new leaf src/lib/db/compressionDetailNormalizers.ts (well under cap) so this frozen file only carries the minimal dispatch wiring. Covered by tests/unit/8388-compression-detail-persist.test.ts (schema-accept + full DB save->reload round-trip for both new sub-objects, plus a no-regression assertion on the existing headroom round-trip).",
-    "_rebaseline_2026_07_24_responses_toolcalls_log_summary": "hartmark, fix/responses-tool-calls-log-summary own growth: open-sse/translator/response/openai-responses.ts 1163->1174 (+11). closeToolCall() now also writes the completed tool call into the shared state.toolCalls Map (already populated by the openai-to-claude / claude-to-openai / gemini-to-openai response translators) so stream.ts's completion-log summary builder (which reads state.toolCalls, not this translator's own funcCallIds/funcNames/funcArgsBuf bookkeeping) reports finish_reason \"tool_calls\" and message.tool_calls for openai->openai-responses translated streams instead of always logging \"stop\" with no tool_calls — the actual client-facing SSE events were already correct; only the persisted call-log summary was wrong. Irreducible call-site addition at the existing tool-call-close chokepoint. Covered by the new regression test in tests/unit/translator-resp-openai-responses.test.ts.",
-    "_rebaseline_2026_07_25_8476_combo_input_bound_homogeneous_scope": "PR #8476 (herjarsa, fix/8375-8459-combo-image-fixes, #8375) own growth: open-sse/services/combo.ts 3642->3679 (+37 net: +29 the PR's own isInputBoundFailure short-circuit for deterministic context_length_exceeded/context_window_exceeded failures, +8 a /green-prs pre-merge fix scoping that short-circuit to homogeneous remainders only — the shipped code fired unconditionally on ANY target, regressing the intentional heterogeneous-combo fallback #6637/isContextOverflow400 protects, exactly as flagged by this PR's own review evidence but never actually implemented in the branch). The fix compares orderedTargets[i+1..] modelStr against the failing target's modelStr at the existing executeTarget dispatch chokepoint (mirrors the sameProviderNext precedent a few lines below) — irreducible call-site wiring, not extractable without hiding the dispatch boundary. Covered by tests/unit/combo-input-bound-failure-8375.test.ts (homogeneous pool still short-circuits) and the new tests/unit/combo-input-bound-heterogeneous-8375.test.ts (heterogeneous combo now correctly falls through to the larger-context target).",
-    "_rebaseline_2026_07_25_adobe_firefly_reference_images": "Follow-up to #8006: storage upload + referenceBlobs for image/video and /v1/images/edits dispatch. adobeFireflyClient.ts 1958->2317 (+upload helpers, extract sources, resolve blob ids). Note: 2317 not 2316 — check-file-size.mjs counts LOC via split(\"\\n\").length (counts the trailing-newline empty element), which is 1 higher than `wc -l` on a file ending in \\n; the PR's original entry (2316) was measured with wc -l and undercounted by 1 against the actual gate.",
-    "_rebaseline_pr1043_minimax_tts": "Upstream port decolua/9router#1043 (toanalien) own growth: audioSpeech.ts 965->1061 (+96). Adds MiniMax T2A v2 TTS dispatch (handleMinimaxSpeech + hexToBytes helper) — provider entry was already in audioRegistry (format: minimax-tts) but no handler existed, falling through to the OpenAI-compatible default that fails (T2A has custom shape + hex-encoded audio + base_resp envelope). New branch sits next to the other inline provider branches (xiaomi-mimo, coqui, tortoise, aws-polly) — extracting would just create indirection. Covered by tests/unit/minimax-tts-1043.test.ts (3 tests, GREEN: success, base_resp error, invalid-hex).",
-    "_rebaseline_pr4592_exclude_exhausted_auto": "Reconcile #4592 already-merged growth: combo.ts 2991->3036 (+45, terminal-status quota-cutoff exclusion in buildAutoCandidates + opt-in gate). Fast-gate PR->release does not run check:file-size.",
-    "open-sse/executors/antigravity.ts": 1528,
-    "open-sse/executors/base.ts": 1640,
-    "open-sse/executors/chatgpt-web.ts": 3241,
-    "open-sse/executors/codex.ts": 1562,
-    "open-sse/executors/cursor.ts": 1563,
-    "open-sse/executors/deepseek-web.ts": 1148,
-    "open-sse/executors/grok-web.ts": 1044,
-    "open-sse/executors/muse-spark-web.ts": 1405,
-    "open-sse/handlers/chatCore.ts": 5034,
-    "open-sse/handlers/imageGeneration.ts": 3101,
-    "open-sse/handlers/responseSanitizer.ts": 1128,
-    "open-sse/handlers/search.ts": 1536,
-    "open-sse/handlers/videoGeneration.ts": 1063,
-    "open-sse/mcp-server/schemas/tools.ts": 1553,
-    "open-sse/mcp-server/server.ts": 1448,
-    "open-sse/mcp-server/tools/advancedTools.ts": 1120,
-    "open-sse/services/accountFallback.ts": 1978,
-    "open-sse/services/adobeFireflyClient.ts": 2385,
-    "open-sse/services/claudeCodeCompatible.ts": 1202,
-    "open-sse/services/combo.ts": 3648,
-    "open-sse/services/compression/strategySelector.ts": 1060,
-    "open-sse/services/rateLimitManager.ts": 1167,
-    "open-sse/translator/response/openai-responses.ts": 1204,
-    "open-sse/utils/cursorAgentProtobuf.ts": 1505,
-    "open-sse/utils/stream.ts": 2889,
-    "src/app/(dashboard)/dashboard/HomePageClient.tsx": 1388,
-    "src/app/(dashboard)/dashboard/analytics/ComboHealthTab.tsx": 1031,
-    "src/app/(dashboard)/dashboard/api-manager/ApiManagerPageClient.tsx": 3117,
-    "src/app/(dashboard)/dashboard/cache/media/MediaPageClient.tsx": 1067,
-    "src/app/(dashboard)/dashboard/combos/page.tsx": 4703,
-    "src/app/(dashboard)/dashboard/costs/CostOverviewTab.tsx": 1283,
-    "src/app/(dashboard)/dashboard/costs/quota-share/components/PoolWizard.tsx": 1022,
-    "src/app/(dashboard)/dashboard/endpoint/EndpointPageClient.tsx": 2615,
-    "src/app/(dashboard)/dashboard/health/page.tsx": 1165,
-    "src/app/(dashboard)/dashboard/providers/[id]/components/modals/EditConnectionModal.tsx": 1324,
-    "src/app/(dashboard)/dashboard/providers/page.tsx": 1944,
-    "src/app/(dashboard)/dashboard/runtime/RuntimePageClient.tsx": 1201,
-    "src/app/(dashboard)/dashboard/settings/components/PricingTab.tsx": 1019,
-    "src/app/(dashboard)/dashboard/settings/components/ProxyRegistryManager.tsx": 1470,
-    "src/app/(dashboard)/dashboard/settings/components/ResilienceTab.tsx": 1123,
-    "src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx": 1629,
-    "src/app/(dashboard)/dashboard/settings/components/SystemStorageTab.tsx": 1573,
-    "src/app/(dashboard)/dashboard/usage/components/BudgetTab.tsx": 1028,
-    "src/app/(dashboard)/dashboard/usage/components/EvalsTab.tsx": 2148,
-    "src/app/(dashboard)/dashboard/usage/components/ProviderLimits/index.tsx": 1119,
-    "src/app/api/providers/[id]/models/route.ts": 2361,
-    "src/app/api/v1/models/catalog.ts": 1590,
-    "src/lib/db/apiKeys.ts": 1529,
-    "src/lib/db/core.ts": 1639,
-    "src/lib/db/migrationRunner.ts": 1094,
-    "src/lib/db/models.ts": 1097,
-    "src/lib/db/providers.ts": 1034,
-    "src/lib/memory/retrieval.ts": 1073,
-    "src/lib/tailscaleTunnel.ts": 1202,
-    "src/lib/usage/providerLimits.ts": 1013,
-    "src/shared/components/OAuthModal.tsx": 1134,
-    "src/shared/components/RequestLoggerV2.tsx": 1629,
-    "src/shared/components/analytics/charts.tsx": 1035,
-    "src/shared/services/cliRuntime.ts": 1122,
-    "src/sse/handlers/chat.ts": 1904,
-    "src/sse/services/auth.ts": 2520,
-    "tests/unit/account-fallback-service.test.ts": 1572,
-    "tests/unit/provider-validation-specialty.test.ts": 2985,
-    "open-sse/executors/hyperagent.ts": 1026,
-    "src/lib/tokenHealthCheck.ts": 1053,
-    "open-sse/executors/default.ts": 1042,
-    "open-sse/executors/kiro.ts": 1069,
-    "open-sse/translator/request/openai-to-kiro.ts": 1057,
-    "open-sse/utils/sseHeartbeat.ts": 149
-  },
+  "_rebaseline_2026_08_02_9242_token_health_transient": "PR #9242 (fix/refresh-circuit-transient): src/lib/tokenHealthCheck.ts 1021 (new file, above cap 1000). The file consolidates token-refresh health checking logic that was previously scattered across auth.ts and tokenRefresh.ts. Cohesive single-responsibility module for refresh circuit state management; not extractable without splitting the refresh state machine. Covered by tests/unit/tokenHealthCheck-transient.test.ts.",
   "_rebaseline_2026_07_27_v3849_train2": "Merge-train 2 (7 PRs) — owner-approved 2026-07-27. Single entry: chatCore.ts 4955->5006 (#8595, Responses multi-turn image compaction before the context hard-reject). Genuine irreducible growth at the existing compaction chokepoint in handleChatCore — the PR adds a last-resort retry against the concrete budget plus the estimateFinalInputTokens helper, both wired at the pre-existing call site rather than a new branch. Covered by tests/unit/8560-responses-image-compaction.test.ts (4 tests).",
   "_rebaseline_2026_07_27_v3849_train3": "Merge-train 3 (13 PRs) — owner-approved 2026-07-27. Both entries are genuine irreducible growth at existing chokepoints, not new branches: src/lib/db/apiKeys.ts 1518->1529 (#8805 cx/* ≡ codex/* API-key model permissions); open-sse/handlers/chatCore.ts 5006->5020 (#8806 real response payload into plugin onResponse hooks). Covered by tests/unit/db-apiKeys-crud.test.ts (4 new cases) and the two plugin-hook test files updated in #8806 respectively.",
   "_rebaseline_2026_07_28_8842_antigravity_projectid_refresh": "PR #8842 (fix/antigravity-projectid-refresh) own growth: open-sse/executors/antigravity.ts 1493->1528 (+35 = projectId discovery in refreshCredentials: import ensureAntigravityProjectAssigned + trim projectId + call ensureAntigravityProjectAssigned with 8s timeout + persistDiscoveredAntigravityProjectId + log success/failure). Irreducible wiring at the existing credential-refresh chokepoint. Covered by tests/unit/executor-antigravity.test.ts (4 new test cases).",
@@ -418,146 +421,15 @@
   "_rebaseline_2026_07_28_8860_tokenrefresh_projectid": "PR #8860 (fix/antigravity-projectid-centralized) own test growth: tests/unit/token-refresh-service.test.ts 1311->1378 (+67 = 4 cases covering projectId discovery on the tokenRefresh.ts path — the Dashboard/health-check refresh route, which #8842 did not reach since that fixed the executor path). Covered by the same file.",
   "_rebaseline_2026_07_28_8861_xiaomi_token_plan": "PR #8861 (feat/xiaomi-token-plan-protocol-selector) own growth: EditConnectionModal.tsx 1283->1316 (+33 = the per-connection API-protocol selector field) and open-sse/executors/base.ts 1540->1562 (+22 = alternate-format resolution at the existing buildUrl/headers chokepoint). Both are irreducible wiring at existing call sites.",
   "_rebaseline_2026_07_28_8863_firefly_detail_level": "PR #8863 (fix/adobe-firefly-gpt-detail-level-max) own growth: adobeFireflyClient.ts 2317->2322 (+5 = gpt-image detailLevel defaulting to maximal at the existing payload-build site). Covered by tests/unit/adobe-firefly.test.ts.",
-  "_rebaseline_2026_07_29_8281_home_quickstart_prefetch": "Release v3.8.49 base-red fix (no PR — captain sweep): src/app/(dashboard)/dashboard/HomePageClient.tsx 1377->1381 (+4). #8292 added prefetch={false} to the sidebar but left /home's five quick-start Links prefetching, so first paint still fired 12 speculative RSC requests — caught by navigation.spec.ts only after the e2e helper bug (APP_ROUTE_PATTERN missing /home) was repaired in the same cycle. Growth is the five prefetch attributes; it was offset first by extracting the repeated className literals (INLINE_LINK x4, DOCS_LINK x1), which collapsed five wrapped <Link> blocks back to one line each — a naive fix measured 1391. Guard: tests/unit/sidebar-prefetch-policy-8281.test.ts.",
-  "_rebaseline_2026_08_02_v3850_agentrouter_responses": "Release v3.8.50 AgentRouter/Codex compatibility reconciliation. open-sse/executors/base.ts 1562->1578: #9190 wires AgentRouter's selected Claude/OpenAI/Responses protocol through the existing executor URL, auth, identity-header and fingerprint chokepoints; the reusable alternate resolver remains outside base.ts. open-sse/utils/stream.ts 2887->2889: #9213 evaluates Responses ID and usage normalization independently so response.completed always receives finite usage.total_tokens instead of short-circuiting after an ID rewrite. tests/unit/chatcore-translation-paths.test.ts 2769->2776: #9191 updates the existing Claude-Code bridge assertions for the dynamic AgentRouter wire image. PR #9224 offsets its own chatCore growth by extracting the AgentRouter protocol decisions into chatCore/agentRouterProtocol.ts, leaving chatCore below its frozen ceiling. Covered by agentrouter executor/chatCore protocol tests, chatcore translation-path tests, and responses-commentary-passthrough tests.",
-  "_rebaseline_2026_08_08_v3850_base_drift_batch_9757": "Base drift on release/v3.8.50, not own growth: the 08-06..08-08 merge batches grew 12 already-frozen (or newly-landed) files without carrying their rebaselines — the dedicated rebaseline PR #9616 was closed as 'superseded' but its file-size entries never actually reached the base, and later merges (#8894 combos page, #9539 EditConnectionModal, #8895 models route, #9294/#9293 catalog, #9541 db/core, #8970 tokenHealthCheck, #8925 mcp schemas+server, #8890 accountFallback, #9467 chat.ts, #8931 openai-to-kiro, ProxyRegistryManager) kept growing them. All 12 values re-measured on THIS branch's tree (= pure tip + this PR's 1-line chat.ts fix, which adds zero lines). This PR's own source changes (chat.ts identifier restore, stream.ts format carve-out) do not grow any frozen file past these values.",
-  "_rebaseline_2026_08_08_migration_135_collision": "fix(db): resolve migration version 135 numbering collision — #9449's 135_connection_runtime_state.sql and #8908's 135_migrate_model_capability_max_token.sql both claimed version 135 (#9449 branched before #8908 merged and never got renumbered before landing on release/v3.8.50), which threw 'Migration version collision detected' the moment ANY code touched the database — a fresh install/deploy from this tip cannot even boot. Renumbered the later-landing file to 140 (next free slot) and added the matching isSchemaAlreadyApplied('140') retroactive guard, matching the established pattern already used for the prior 135/136 -> 137/138 renumber in the same file. Own growth: src/lib/db/migrationRunner.ts 1084->1094 (+10, the new case block) — irreducible, matches the existing per-case guard pattern exactly. Covered by tests/unit/migration-135-numbering-collision.test.ts (2/2), confirmed failing (reproducing the exact live crash) against the pre-fix colliding filenames, passing after.",
-  "_rebaseline_2026_08_02_9259_rolling_rpm": "PR #9259 (issue #8733) own growth: open-sse/services/rateLimitManager.ts baseline 1060->1167 (+107; final source 1153). The existing withRateLimit chokepoint now composes process-local rolling RPM leases with Bottleneck admission, releases pre-dispatch leases on queue timeout/abort/connection disable, preserves caller abort reasons, and wires 429/header state into the extracted rollingRpmGate.ts. The remaining growth is irreducible lifecycle wiring at the dispatch boundary plus the real watchdog test hooks needed to verify queued-wedge recovery; moving it further would obscure lease ownership and Bottleneck cleanup. Covered by the focused rate-limit manager/sliding-window suite (33/33); distributed multi-instance coordination remains explicitly out of scope.",
-  "_rebaseline_2026_07_25_dario_upstream_proxy_selector": "PR #8523 (Dario embedded service): upstream-proxy mode selector replaces the binary CLIProxyAPI toggle with Native/CLIProxyAPI/Dario/Fallback + a fallback-backend picker. ProviderDetailPageClient.tsx 798->804 (+6, new hook fields threaded through to ConnectionsListPanel), ConnectionRow.tsx 942->958 (+16, the mode <select> + conditional fallback-backend <select> replacing a single pill button), useProviderConnections.ts 954->986 (+32, upstreamProxyMode/upstreamProxyFallbackBackend state + handleSetUpstreamProxyMode, handleToggleCliproxyapiMode kept as a thin backward-compat wrapper for the existing hook-shape test). All additive UI/state for the new modes — no unrelated refactor.",
-  "_rebaseline_2026_08_02_9242_token_health_transient": "PR #9242 (fix/refresh-circuit-transient): src/lib/tokenHealthCheck.ts 1021 (new file, above cap 1000). The file consolidates token-refresh health checking logic that was previously scattered across auth.ts and tokenRefresh.ts. Cohesive single-responsibility module for refresh circuit state management; not extractable without splitting the refresh state machine. Covered by tests/unit/tokenHealthCheck-transient.test.ts.",
   "_rebaseline_2026_07_28_8870_firefly_ref_cap_timeout": "PR #8870 (fix/adobe-firefly-gpt-ref-cap-timeout) own growth: adobeFireflyClient.ts 2322->2385 (+63 = gpt-image subject-ref hard cap at 2 + adaptive poll timeout budget (base 300s + 60s/ref, max 600s) + defensive .slice on referenceBlobs for gpt/nano/generic families). Fixes live 504s on multi-screenshot listing jobs (Featured Promo / Box Art) where 3–4+ subject refs stall colligo until the old 180s poll budget expires. Helpers adobeFireflyMaxImageRefs/adobeFireflyImageTimeoutMs live next to the existing payload/poll chokepoint (not extractable without splitting the wire recipe mid-PR). Covered by tests/unit/adobe-firefly.test.ts (ref-cap + timeout cases). Structural shrink tracked in #3501.",
+  "_rebaseline_2026_07_29_8281_home_quickstart_prefetch": "Release v3.8.49 base-red fix (no PR — captain sweep): src/app/(dashboard)/dashboard/HomePageClient.tsx 1377->1381 (+4). #8292 added prefetch={false} to the sidebar but left /home's five quick-start Links prefetching, so first paint still fired 12 speculative RSC requests — caught by navigation.spec.ts only after the e2e helper bug (APP_ROUTE_PATTERN missing /home) was repaired in the same cycle. Growth is the five prefetch attributes; it was offset first by extracting the repeated className literals (INLINE_LINK x4, DOCS_LINK x1), which collapsed five wrapped <Link> blocks back to one line each — a naive fix measured 1391. Guard: tests/unit/sidebar-prefetch-policy-8281.test.ts.",
   "_rebaseline_2026_08_01_8964_xai_agent_tools": "PR #8964 own growth: chatCore.ts 5020->5034 at the existing native-passthrough chokepoint. Adds xAI Agent Tools passthrough for /v1/responses (xai/xai-oauth/xao): resolve nativeXaiResponsesPassthrough, force openai-responses targetFormat, stamp body marker, and OR into the existing nativeCodexPassthrough sites (web-search bypass + requestEndpointPath). Leaf logic in passthroughHelpers, responsesEndpoint, targetFormat, xai executor, responseSanitizer, usageTracking. Cohesive wiring at the Codex passthrough boundary.",
   "_rebaseline_2026_08_01_8964_response_sanitizer": "PR #8964 own growth: responseSanitizer.ts 1115->1128. Keep cost_in_usd_ticks / server_side_tool_usage(_details) through sanitizeResponsesApiResponse allowlists so native xAI tool responses retain usage.",
+  "_rebaseline_2026_08_02_v3850_agentrouter_responses": "Release v3.8.50 AgentRouter/Codex compatibility reconciliation. open-sse/executors/base.ts 1562->1578: #9190 wires AgentRouter's selected Claude/OpenAI/Responses protocol through the existing executor URL, auth, identity-header and fingerprint chokepoints; the reusable alternate resolver remains outside base.ts. open-sse/utils/stream.ts 2887->2889: #9213 evaluates Responses ID and usage normalization independently so response.completed always receives finite usage.total_tokens instead of short-circuiting after an ID rewrite. tests/unit/chatcore-translation-paths.test.ts 2769->2776: #9191 updates the existing Claude-Code bridge assertions for the dynamic AgentRouter wire image. PR #9224 offsets its own chatCore growth by extracting the AgentRouter protocol decisions into chatCore/agentRouterProtocol.ts, leaving chatCore below its frozen ceiling. Covered by agentrouter executor/chatCore protocol tests, chatcore translation-path tests, and responses-commentary-passthrough tests.",
   "_rebaseline_2026_08_05_9323_agentrouter_waf_retry": "PR #9323 (fix(agentrouter): retry on 400 content-blocked + burst guard) own growth: open-sse/executors/base.ts 1578->1623 (check-file-size.mjs conta via split(\"\\n\").length; wc -l ve 1622). As +45 linhas sao o WAF_RETRY_CONFIG + o burst guard via gateOutboundRequest() para o WAF do agentrouter.org, com comentarios explicando o porque de cada mitigacao e cobertos por tests/unit/base-executor-waf-retry.test.ts e tests/unit/wafRateLimit.test.ts. Crescimento funcional legitimo, nao inchaco.",
   "_rebaseline_2026_08_05_9529_own_growth": "PR #9529 own growth (base release/v3.8.50 medida EXATAMENTE nos frozen antigos, entao o modo base-relative #8522 nao cobre): open-sse/services/rateLimitManager.ts 1060->1105 (+45: helper applyLimiterSettings() que re-arma o heartbeat do reservoir apos updateSettings — fix do bug Bottleneck 2.19.5 que congelava a fila weighted; TDD em tests/unit/ratelimit-reservoir-refresh.test.ts); tests/integration/chat-pipeline.test.ts 1592->1598 (+6: User-Agent do codex derivado de getCodexClientVersion() em vez de literal pinado — teste-irmao alinhado ao contrato); tests/unit/provider-validation-specialty.test.ts 2980->2985 (+5: cobertura NOVA claude-web 429 -> valid:false, alinhamento #9406); open-sse/translator/response/openai-responses.ts 1174->1204 (+30: buildResponsesReasoningSummaryDelta MOVIDA do leaf pureHelpers.ts para o host — a funcao do #9500 muta stream state e violava o contrato do leaf puro; o LOC total do par host+leaf nao cresceu, o pureHelpers encolheu o mesmo tanto). Crescimento por fix de producao + cobertura adicional + realocacao arquitetural, nao inchaco.",
   "_rebaseline_2026_08_06_v3850_inherited_drift_reconcile": "Reconciliacao 2026-08-06 do drift ACUMULADO da release/v3.8.50 apos o lote de merges de 08-05/06: 13 arquivos acima do frozen no tip puro 8180b49ce1 (medidos pelo proprio gate). O modo PR base-relative (#8522) deixa PRs inocentes passarem, e os rebaselines individuais dos PRs se perderam nas resolucoes sucessivas de conflito deste hot-file — o drift so aparece no modo absoluto (nightly/local). Crescimentos funcionais dos PRs mergeados: #9024 topology click-nav src/app/(dashboard)/dashboard/HomePageClient.tsx; #9324 OpenRouter enrich src/app/(dashboard)/dashboard/providers/page.tsx; #9329 quota card ordering src/app/(dashboard)/dashboard/usage/components/ProviderLimits/index.tsx; #9193 context-window suffixes src/sse/handlers/chat.ts; #9332 nested Claude server tool ids open-sse/executors/base.ts; #9228 strip orphaned tool outputs open-sse/executors/codex.ts; #9236 nvidia tool-name normalize open-sse/executors/default.ts; #9314 nested tool_call validation open-sse/executors/kiro.ts; #9260 caller identity REST hops open-sse/mcp-server/server.ts; #8934 cache breakpoints tests tests/unit/chatcore-translation-paths.test.ts; #9193 suffix tests tests/unit/combo-routing-engine.test.ts; #9196 reasoning-on-tool-finish tests tests/unit/sse-auth.test.ts; #9163 GPT-5.6 Max reasoning tests tests/unit/translator-openai-to-kiro.test.ts. default.ts e kiro.ts entram no frozen (estavam sem entrada, acima do cap 1000). Atualizacao pos-medicao (a base avancou durante o ciclo do PR): src/sse/handlers/chat.ts 1857->1877 (#9184 affinity EOF evict) e open-sse/executors/default.ts 1027->1042 (#9005 Kimi K3 tool-name backfill).",
   "_rebaseline_2026_08_06b_v3850_sweepreds_drift": "Segunda reconciliacao de 2026-08-06 (/sweep-reds sobre o tip puro 2ddbbc61a6): 3 arquivos voltaram a passar do frozen apos os merges do mesmo dia, com atribuicao 1:1 por commit. (1) src/app/(dashboard)/dashboard/providers/page.tsx 1928->1944 e (2) open-sse/executors/base.ts 1635->1640, ambos do #9515 (feat(radar): flag-gated signed free-model catalog overlay, commit e7f6b1d130) — o overlay do Radar entra por wiring nos chokepoints ja existentes (a resolucao/verificacao do catalogo assinado mora fora destes dois arquivos); +16 e +5 linhas liquidas nao sao extraiveis sem inventar um leaf por callsite. (3) open-sse/services/accountFallback.ts 1966->1972 do #8704 (commit c4527f97bd), +6 linhas de dados em CREDITS_EXHAUSTED_SIGNALS ('has been exhausted', fixes #8631). src/sse/handlers/chat.ts 1880>1877 tambem estava violando e NAO entra aqui de proposito: e drenado por encolhimento na PR #9598, sem rebaseline. Crescimento proprio DESTA PR: src/lib/db/migrationRunner.ts 1077->1084 (+7) — o guard retroativo em isSchemaAlreadyApplied para os arquivos renumerados 137/138, exigido pela propria mensagem de erro de colisao do runner (ambas as migracoes sao ALTER TABLE ADD COLUMN puro, nao idempotente). Dois `case` + dois `return hasColumn(...)` + 3 linhas de comentario dentro do switch existente; nao extraivel.",
   "_rebaseline_2026_08_06c_v3850_sweepreds_pr2": "Segunda PR do /sweep-reds (fix/release-v3.8.50-basereds-0806b): tests/unit/provider-models-route.test.ts 1784->1787 (medido pelo gate, que conta split(\"\\n\").length) (+2 apos compressao de comentarios) — alinhamento de contrato forcado por dois merges do dia: #9106 tornou gemini-3.1-pro-high user-callable (a entry do alias entra na lista esperada do teste de discovery-retry, +1 linha de dado + 1 de comentario) e ff012ff420 adicionou onboardUser como bootstrap hop (exclusao no mock, ja comprimida a 1 linha). Nao ha o que encolher sem apagar o comentario que explica o porque.",
-  "_rebaseline_2026_08_07_v3850_sweepreds_pr2_toolnamemap": "tests/unit/translator-openai-to-gemini.test.ts 1616->1619 (+3). O frozen estava EXATAMENTE no tamanho da base, entao qualquer linha nova viola. #9568 (c9a3361e5a) fez buildChangedToolNameMap emitir entradas IDENTIDADE (o Gemini minusculiza nomes de tool nas respostas, entao o tradutor de resposta precisa da chave para mapear de volta), o que passou a incluir `_toolNameMap` no envelope Antigravity de qualquer request com tools. As 3 linhas sao: a chave nova na lista esperada de Object.keys, 1 comentario explicando POR QUE ela aparece (sem ele o proximo leitor tenta remove-la de novo) e 1 assert do CONTEUDO do map — presenca de chave sozinha nao provaria a entrada identidade, que e justamente o comportamento novo. Nao ha o que extrair: e alinhamento de contrato dentro de um teste existente.",
-  "_rebaseline_2026_06_22_4644_deepseek_web_tools": "PR #4644 (BugsBag/robust deepseek-web tool-call parsing): open-sse/executors/deepseek-web.ts 1117->1125 (+8). The new agentic tool-call path emits surrounding text + reasoning before tool_calls and swaps to the dedicated deepseekWebTools.ts parser; the +8 lines are cohesive wiring at the existing transformSSE chokepoint (the parser itself lives in the new deepseekWebTools.ts file, already under cap). The PR's own fast-gate (PR->release) does not run check:file-size, so this surfaced only at release reconcile. Covered by tests/unit/deepseek-web-tools-variants.test.ts + deepseek-web-tools-execute.test.ts.",
-  "_rebaseline_2026_06_23_4712_deepseek_web_tool_results": "PR for #4712 (deepseek-web drops role:tool): open-sse/executors/deepseek-web.ts 1125->1148 (+23). messagesToPrompt() now folds role:\\\"tool\\\" results into the single-prompt transcript (recovering the tool name from the preceding assistant tool_calls by tool_call_id) instead of silently dropping them; the lines are cohesive wiring inside the existing function. Covered by tests/unit/deepseek-web-tool-result-prompt-4712.test.ts.",
-  "_rebaseline_2026_06_24_headroom_strategy": "Headroom-aware connection selection (dario technique): combo.ts 3168->3180 (+12 = a new `else if (strategy === \\\"headroom\\\")` dispatch branch in handleComboChat that delegates to orderTargetsByHeadroom + its log line, plus the import). The actual logic lives OUT of the god-file: the pure ranker rankByHeadroom/computeHeadroom is the new leaf open-sse/services/combo/headroomRanking.ts (91 LOC, <cap) and the async orderer orderTargetsByHeadroom is appended to the existing open-sse/services/combo/quotaStrategies.ts (<cap) next to its sibling reset-aware/reset-window orderers (reuses their connection-expansion machinery). headroom = 1 - max(util_5h, util_7d) from getSaturation (src/lib/quota/saturationSignals.ts), prefers the connection with the most free capacity. Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. fill-first stays default; all existing strategies untouched. Covered by tests/unit/combo-headroom-ranking.test.ts (pure helper) + tests/unit/combo-headroom-strategy.test.ts (orderer, saturation injected). Structural shrink of combo.ts tracked in #3501.",
-  "_rebaseline_2026_06_24_quota_share_strategy": "Dedicated quota-share strategy (Phase 3 #9): combo.ts 3180->3190 (+10 = one new `else if (strategy === \\\"quota-share\\\")` dispatch branch in handleComboChat that delegates 100% to selectQuotaShareTarget + its log line, plus the import). All the new logic lives OUT of the god-file in two new leaves under open-sse/services/combo/: quotaShareInflight.ts (in-flight counter with TTL/lease, ~150 LOC <cap) and quotaShareStrategy.ts (per-model bucket gating via isBucketSaturated + DRR proportional to weight + P2C over in-flight, ~240 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors the headroom/reset-aware/reset-window/context-optimized branches); not extractable without hiding the call site. ZERO existing strategy cases were modified — only this branch was added, and the qtSd/ combos switched from fill-first to quota-share in src/lib/quota/quotaCombos.ts. Covered by tests/unit/quota-share-strategy.test.ts (gating, DRR fairness, P2C in-flight, fail-open, activation). Structural shrink of combo.ts tracked in #3501.",
-  "_rebaseline_2026_06_24_task_aware_routing": "Task-aware routing strategy (port PR #2045, OmniRoute #4945): combo.ts 3190->3225 (+35) = one new `else if (strategy === \\\"task-aware\\\")` dispatch branch delegating 100% to selectTaskAwareTarget + its imports/log lines. All scoring/classification logic lives OUT of the god-file in the new leaf open-sse/services/taskAwareRouting.ts (553 LOC <cap). Only the dispatch wiring is irreducible at the existing combo strategy chokepoint (mirrors quota-share/headroom/reset-aware branches). ZERO existing strategy cases modified. Covered by tests/unit/combo-task-aware.test.ts (35 tests). Structural shrink of combo.ts tracked in #3501.",
-  "_rebaseline_2026_06_26_fidelity_gate_extraction": "Milestone-B fidelity-gate wiring residual: bodyToText+gateAdvance extracted to fidelityGateStep.ts (889->854, -35), but the StackOptions.fidelityGate field, the `const fidelityGate` reads at the two stacked-loop dispatch chokepoints, and the import of FidelityGateConfig are irreducible wiring that cannot leave strategySelector without an architectural refactor of the pre-existing stacked pipeline. Net: 889->854 (+6 vs the pre-Milestone-B frozen 848). Covered by tests/unit/compression/*.test.ts (940 pass).",
-  "_rebaseline_2026_06_27_5193_5203_antigravity_oauthmodal": "Antigravity remote-login own growth: OAuthModal.tsx 960->969 (gate units). #5193 (+~4: remote paste instruction shown for all remote incl. Google + its rationale comment) and #5203 (+~5: handleManualSubmit credential-blob branch + button guard; submit logic extracted to oauthBlobSubmit.ts to minimize). Frozen set to the SUM so either merge order passes. Cohesive at the existing manual-submit chokepoint.",
-  "_rebaseline_2026_06_27_5193_antigravity_basered": "Base-red (pre-existing release drift, fast-gate PR->release skips check:file-size): accountFallback.ts 1773->1777 and src/app/api/providers/[id]/test/route.ts 924->940 were already over their frozen caps on release/v3.8.39 independent of any antigravity change. Owner chose to rebaseline (keep the documented issue-reference comments #1846/#1449/#347 etc.) rather than accept the contributor comment-stripping in #5200/#5198. Reverted #5200 to restore the comments; bumped these two frozen caps to the actual base sizes. No logic change.",
-  "_rebaseline_2026_06_28_5237_impersonation_ua_refresh": "PR #5237 (refresh impersonation UAs): grok-web.ts 1871->1873 (+2), muse-spark-web.ts 1284->1302 (+18), perplexity-web.ts 1013->1032 (+19). Net semantic change in each file is a single User-Agent constant (Chrome 147->149 for grok/muse; perplexity kept at Firefox 148 to stay matched with the firefox_148 TLS profile — the contributor's 152 bump was reverted to avoid a UA-vs-JA3 mismatch, #2459). The growth is Prettier reflow that lint-staged unavoidably applies to these grandfathered long-line files the moment they are touched; not extractable. src/sse/services/auth.ts 2336->2401 in the same reconcile is #5222's antigravity-LRU-retry growth that merged via --admin without a baseline bump.",
-  "_rebaseline_2026_06_28_5243_risk_gate_prepass": "PR #5243 (compression risk-gate pre-pass) own growth: open-sse/services/compression/strategySelector.ts 854->899 (+45). The three exported entry points (applyCompression/applyStackedCompression/applyStackedCompressionAsync) become thin wrappers over pure-extracted private bodies (runCompression/runStackedCompression/runStackedCompressionAsync) so the risk-gate mask->run->restore wrapper sits strictly OUTSIDE the per-step loop — a single universal integration point. The wrapper logic itself (resolveRiskGate/withRiskGate) lives in the new riskGate/strategyWrap.ts (<cap); the residual growth is the duplicated thin-wrapper signatures + the extracted bodies' dispatch boundary, guarded by a byte-identical parity test (riskGateIntegration). Default off (DEFAULT_COMPRESSION_CONFIG unchanged). Not extractable without hiding the dispatch boundary, mirroring prior compression rebaselines. Structural shrink tracked in #3501.",
-  "_rebaseline_2026_06_28_5275_correlation_id_extract": "Extraction of the safe CorrelationId subset of #5275 (hartmark) — request correlation id stored in call_logs (migration 109) and returned via the X-Correlation-Id response header, WITHOUT the combo/resilience or build/lazy-loading changes (those stay in #5275). Own growth: callLogs.ts 975->985 (correlation_id column on CallLogSummaryRow + read/map), usageHistory.ts 983->988 (correlationId metadata normalize), chat.ts 1575->1632 (withCorrelationId response wiring + combo-failure log carrying correlationId), chatHelpers.ts new 811 (withCorrelationId helper + reqId threading; was 791<cap pre-feature). Cohesive request/logging chokepoint wiring; structural shrink of chat.ts tracked in #3501.",
-  "_rebaseline_2026_06_29_4038_cas_guard": "PR (#4038) own growth: tokenRefresh.ts 2103->2181 (+78 = the compare-and-swap guard on the refresh persist — runWithCasGuard/getActiveCasGuard AsyncLocalStorage pair mirroring runWithOnPersist, casGuardShouldSkipPersist that rereads the row right before persisting and skips the write when a concurrent writer already rotated the refresh_token past the one presented, plus getCasGuardStats counters). Fixes the sibling-rotation-revert → token-family-revocation storm. Gated behind an active guard (opt-in; no guard => byte-identical). Wiring lives at the two persist chokepoints inside getAccessToken; the comparison reuses wasRefreshTokenRotated from refreshSerializer. Not extractable without splitting the refresh hot path.",
-  "_rebaseline_2026_06_29_5286_memoization": "PR #5286 own growth: strategySelector.ts 899->960 (+61 = the opt-in result-memoization branches in applyCompression/applyCompressionAsync — principal+determinism gate, makeMemoKey lookup/store with model+supportsVision folded into the key, recompute-with-memo-off). Default off (memoizeCompressionResults), so zero behavior change. The memo helpers live in the leaf resultMemo.ts (<cap); the chokepoint wiring here is not extractable. Structural shrink of this hot-path file tracked in #3501.",
-  "_rebaseline_2026_07_01_v3843_release_5609": "Rebaseline v3.8.43 (PR #5609 release reconciliation). DRIFT dos 109 commits do ciclo: 8 god-files existentes cresceram (ApiManagerPageClient 2983->3017, combos/page 4594->4608, AddApiKeyModal 868->869, providerPageHelpers 974->996, chat.ts 1635->1647, auth.ts 2401->2403, batchProcessor 828->915, combo.ts 3368->3387) + 2 novos acima do cap (huggingchat.ts 813, tests web-cookie-providers-new 827) + 4 test files cresceram. Modularizacao deferida (blast-radius mid-release); congelado no estado atual p/ o proximo ciclo ratchetar daqui.",
-  "_rebaseline_2026_07_02_5816_qoder": "PR #5816 (@AgentKiller45, qoder PAT via qodercli): qoderCli.ts 666->989, new-above-cap frozen (owner-approved baseline freeze). The growth is the legitimate PAT job-token exchange + quota parsing CLI transport (the pure-JS Cosy path 500'd on every PAT request); extracting the spawn/parse helpers now would just add indirection to a contributor PR mid-merge. Test frozen also raised for this PR's coverage growth: providers-page-utils.test.ts 1052->1092. Additionally clears an inherited base-red from the already-merged #5933 (codex json_schema->text.format): translator-openai-responses-req.test.ts 1097->1172 (+75 regression tests, no offending branch left). All remain frozen (cannot grow further); release captain's rebaseline-at-release supersedes.",
-  "_rebaseline_2026_07_09_6126_clinepass_dual_auth": "PR #6126 (@hajilok, dual-auth ClinePass) own growth: tokenRefresh.ts 2181->2182 (+1 = a single `case \\\"clinepass\\\":` fallthrough label added to the existing `case \\\"cline\\\":` in _getAccessTokenInternal's provider switch, so clinepass token refresh dispatches to the already-shared refreshClineToken() instead of silently falling through to the generic OAuth refresh). Irreducible 1-line switch-case wiring at the existing chokepoint; the header-building logic for the same feature was extracted to a new leaf src/shared/utils/clineAuth.ts::buildClinepassHeaders() (well under cap) to avoid growing open-sse/executors/default.ts. Covered by tests/unit/clinepass-provider.test.ts.",
-  "_rebaseline_2026_07_09_6363_kiro_external_idp": "PR #6363 (@artickc, Kiro external IdP) own growth: tokenRefresh.ts 2182->2249 (+67 = the external_idp refresh branch inside refreshKiroToken — standard public-client OAuth2 refresh_token grant against the org IdP tokenEndpoint via buildExternalIdpRefreshParams/isExternalIdpAuthMethod from the new leaf open-sse/services/kiroExternalIdp.ts, with invalid_grant/invalid_client -> unrecoverable_refresh_error mapping). Cohesive addition at the existing refreshKiroToken chokepoint. Covered by tests/unit/kiro-external-idp.test.ts.",
-  "_rebaseline_2026_07_09_6587_kiro_api_key_auth": "PR #6587 (@strangersp) own growth for Kiro long-lived API-key auth, merged onto v3.8.47 tip: openai-to-kiro.ts 890->912 (+22, auth-header selection for API-key-vs-OAuth-token connections), providerLimits.ts 998->1000 (+2, API-key auth-type branch), translator-openai-to-kiro.test.ts 1234->1257 (+23), providers-page-utils.test.ts 1109->1107 (net -2 after merging with parallel release drift; connectionMatchesProviderCard api_key coverage added), provider-validation-specialty.test.ts 2856->2980 (+124 net after merge with parallel release drift; this PR also removed the file's `@typescript-eslint/no-explicit-any` eslint-suppression entry by fixing all `any` usages, adding typed replacements). Cohesive additive feature growth, well tested; not extractable without splitting the existing chokepoints mid-merge.",
-  "_rebaseline_2026_07_09_6678_routing_strategy_9router": "#6678 (SeaXen) — 9router-parity Routing Strategy settings card + per-provider/combo sticky-round-robin override. Own growth: ProviderDetailPageClient.tsx 784->786 (single ProviderAccountRoutingCard mount + import), auth.ts 2448->2458 (providerStrategies override resolution: fallbackStrategy/stickyRoundRobinLimit per-provider cascade in getProviderCredentials). Both additive, zero unrelated refactor; new UI/logic lives in new files (ProviderAccountRoutingCard.tsx, RoutingStrategyCard.tsx, rrState.ts::resolveComboStickyRoundRobinLimit). chat.ts value below reflects the current release tip (grown by other concurrent PRs, e.g. #6640), not this PR own change.",
-  "_rebaseline_2026_07_10_6318_omp_letta": "PR #6318 (@hamsa0x7, omp+letta CLI integrations) own growth: cliTools.ts (+53 = 2 registry entries incl. omp docsUrl) and cliRuntime.ts (+18 = runtime-detection wiring for the 2 new tools). Cohesive registry/wiring growth at the existing chokepoints; scope reduced from the original 5 tools (pi/codewhale/jcode shipped separately).",
-  "_rebaseline_2026_07_10_gcf_v3_2_decode": "PR #6838 own growth: new vendored file open-sse/services/compression/engines/headroom/gcf/decode_generic.ts frozen at 880 (> 800 cap). It is the vendored GCF generic-profile decoder (spec v3.2 nested flattening plus the prototype-pollution / hasOwnProperty hardening added in this PR's Gemini review). Kept as one file faithful to upstream gcf-typescript so re-vendoring stays a clean copy rather than a re-split each cycle (sibling generic.ts/scalar.ts stay < cap; extraction would also fragment the file's frozen eslint no-explicit-any suppressions). Round-trip + prototype-pollution regression coverage in tests/unit/compression/headroom-smartcrusher.test.ts. Frozen: only shrinks from here.",
-  "_rebaseline_2026_07_12_v3847_mergeprs_tail": "v3.8.47 /merge-prs tail (owner-approved): src/lib/localDb.ts NEW>800 (799->805, +6 re-exports countFreeProxies + recordFreeProxySyncErrors/clearFreeProxySyncErrors/getFreeProxySyncErrors + FreeProxySyncErrors type for #6909 free-pool relay-repair; re-export-only per Hard Rule #2, not extractable).",
-  "_rebaseline_2026_07_15_7070_combos_memo": "PR #7070 (perf/p1-memo) own growth: src/app/(dashboard)/dashboard/combos/page.tsx 4655->4656 (+1 = React.memo wrapping of ComboCard). Covered by tests/unit/ui/combos-page-smoke.test.tsx.",
-  "_rebaseline_2026_07_18_7399_xai_oauth_modal": "PR #7399 (xAI OAuth PKCE) own growth: OAuthModal.tsx 993->998 (+5 = provider entry + PKCE flow branch wiring at the existing provider-switch chokepoint; the provider logic itself lives in src/lib/oauth/providers/xai-oauth.ts, new leaf). Third irreducible wiring bump on this modal (969->989->993->998); structural shrink tracked in #3501.",
-  "_rebaseline_2026_07_19_6636_codex_session_json": "#6636 own growth: OAuthModal.tsx 998->1030 (gate units, split(\\\"\\\\n\\\").length incl. trailing newline; +32 = session-JSON paste branch for handleManualSubmit plus a shared submitCodexAccessToken() helper extracted from the pre-existing bare-JWT branch, mirroring the #5203 oauthBlobSubmit.ts extraction precedent; the normalizer logic itself lives in the new src/lib/oauth/utils/codexSessionImport.ts leaf module, not here). Fourth irreducible wiring bump on this modal (969->989->993->998->1030); structural shrink tracked in #3501.",
-  "_rebaseline_2026_07_19_7546_ghe_copilot_modal": "PR #7546 (GHE Copilot OAuth provider) own growth: OAuthModal.tsx 1030->1056 (gate units). Adds a gheUrl input state, routes ghe-copilot through the existing device-code branch, and threads gheUrl into the device-code request/poll extraData at the existing provider-switch chokepoints (+~24 lines, cohesive with the same pattern as #7399/#6636). The standalone GHE enterprise-URL config step JSX (originally +31 lines inline) was extracted to the new src/shared/components/oauthModal/GheConfigStep.tsx leaf component to minimize the bump; what remains is the irreducible provider-branch wiring. Fifth bump on this modal (969->989->993->998->1030->1056); structural shrink tracked in #3501.",
-  "_rebaseline_2026_07_19_7787_ic2_localdb_reexports": "PR #7787 (IC2 raw connections cache + lazy-decrypt) own growth: localDb.ts 805->807 (gate units, +2). localDb.ts is the re-export-only layer (hard rule #2 — no logic); the PR adds 4 new db/readCache re-exports (touchConnectionLastUsed, getCachedRawProviderConnections, getCachedProviderConnectionById, getCachedProviderNodes) required by existing barrel importers. Irreducible for a re-export list; frozen so it can only shrink.",
-  "_rebaseline_2026_07_20_7779_routingcombo_thread": "PR #7779 own growth: chatHelpers.ts 876->877 (+1, thread routingComboId into executeChatWithBreaker for compression-combo assignment). Frozen so it can only shrink.",
-  "_rebaseline_2026_07_20_7819_autocandidateoverrides_reexport": "PR for #7819 (Level 1+2: read-only auto/* candidate transparency + per-API-key exclusions) own growth: localDb.ts 807->808 (+1). Adds a single `export * from \\\"./db/autoCandidateOverrides\\\"` barrel re-export (hard rule #2 — no logic) for the new DB module backing per-apiKey candidate exclusions. Irreducible for a re-export list; frozen so it can only shrink.",
-  "_rebaseline_2026_07_21_8027_grok_cli_auth_json_paste": "PR #8027 (RaviTharuma, fix(grok-cli) #7610) own growth: OAuthModal.tsx 1080->1100 (gate units). Requires the full ~/.grok/auth.json (with refresh_token) on the paste-import path instead of a bare JWT, at the existing paste-token chokepoint (renamed tab label, updated instructions/placeholder, textarea for the auth.json blob, inline error surface). The validation logic itself (parseGrokCliPasteToken, previously an inline ~75-line function) was extracted to the new src/lib/oauth/utils/grokCliAuthJson.ts leaf module — mirroring the #6636/#7546 extraction precedent — so only the irreducible UI wiring remains here. Sixth bump on this modal (969->989->993->998->1030->1056->1100); structural shrink tracked in #3501.",
-  "_rebaseline_2026_07_21_8034_compression_exclusions_sidebar": "#8034 (compression exclusions dashboard tab) own growth: sections.ts 796->806 (+10, one new COMPRESSION_CONTEXT_GROUP sidebar item linking /dashboard/compression/exclusions). The file was already 796/800 before this PR (organic growth from prior sidebar entries), so a single new nav item pushed it 6 lines over cap. Freezing at 806 (cannot grow further); the sidebar item array is data, not extractable logic.",
-  "_rebaseline_2026_07_22_7936_namespace_roundtrip": "#7936 (@RCrushMe, Responses-Chat namespace round-trip identity seam) own growth: open-sse/translator/response/openai-responses.ts 1092->1125 (+33) and open-sse/utils/stream.ts 2814->2869 (+55) — threading the namespace-identity seam through the Responses↔Chat translation + stream paths so tool-call namespaces survive the round-trip. Cohesive translation/stream wiring at existing chokepoints, frozen at new size.",
-  "_rebaseline_2026_07_22_8010_codex_responses_engine": "PR #8010 (@JxnLexn) own growth: open-sse/mcp-server/schemas/tools.ts 1497->1505 (+8 = threading the new \\\"codex-responses\\\" literal into the compressionConfigureInput strategy/autoTriggerMode Zod enums and setCompressionEngineInput engine enum, mirroring the existing rtk/omniglyph enum entries; no new tool). open-sse/services/compression/strategySelector.ts 1043->1054 (+11 = one new `if (mode === \\\"codex-responses\\\")` dispatch branch in runCompression that delegates 100% to the new codexResponsesEngine.apply, mirroring the existing rtk single-mode dispatch, plus threading config.codexResponsesConfig.preserveToolNames into the shared adaptBodyForCompression call at the 3 existing call sites). src/lib/db/compression.ts (untracked, new-file cap 800) 794->845 (+51 = normalizeCodexResponsesConfig, mirroring the existing normalizeRtkConfig normalizer, plus registering \\\"codex-responses\\\" in the COMPRESSION_MODES/STACKED_PIPELINE_ENGINE_IDS/SINGLE_MODE_ENGINE sets and the getCompressionSettings load/save switch) — added to the baseline at its current size. All three are cohesive dispatch/normalizer wiring at existing chokepoints (mirroring the prior compression-mode rebaselines #6534/#6556), not extractable without hiding the mode-dispatch boundary. Covered by tests/unit/compression/codex-responses.test.ts (6) + omniglyph-registries.test.ts/types.test.ts (22, updated for the new mode).",
-  "_rebaseline_2026_07_22_8034_compression_exclusions_persistence": "#8034 (compression exclusions) own growth: src/lib/db/compression.ts 845->850 (+5 = threading the new compressionExclusions field through the existing getCompressionSettings/saveCompressionSettings load/save switch over the shared key_value compression namespace — no new table, no raw SQL). Mirrors the prior compression-field rebaselines (#8010 codex-responses normalizer at the same chokepoint); the load/save switch is a single dispatch boundary, not extractable without hiding it. Covered by the PR's 8 node:test + 3 vitest cases.",
-  "_rebaseline_2026_07_22_8050_model_lockout_exact_family": "#8050 (@AndrianBalanescu) own growth: accountFallback.ts 1864->1892 (+28) — exact-vs-family model-lockout scoping (getModelLockKey/isModelLocked/clearModelLock/getModelLockoutInfo) so an Antigravity 404 for one bare model no longer hijacks the whole family cooldown. Cohesive lockout logic; frozen at new size.",
-  "_rebaseline_2026_07_22_8081_reasoning_placeholder_guard": "#8081 (@Dingding-leo) own growth: openai-responses.ts 1125->1137 (+12) restructuring the reasoning-placeholder guard so it skips only the empty content block and still emits finish_reason/tool_calls in the same chunk. Cohesive translator wiring; frozen at new size.",
-  "_rebaseline_2026_07_22_8210_openrouter_midstream_error": "PR #8210 (hartmark, fix/openrouter-midstream-error-surfacing) own growth: open-sse/translator/response/openai-responses.ts 1137->1163 (+26) measured on the merged tip (release 1137 + this PR own growth). Adds a single new branch inside openaiToOpenAIResponsesResponse() that detects an OpenRouter-style mid-stream aggregator error (HTTP 200 SSE chunk with empty choices + a top-level error object) and surfaces it as state.upstreamError instead of silently falling through to the no-op/awaitingTrailingUsage path, which previously masked the failure as a false empty-success completion and skipped combo fallback. Irreducible call-site addition at the existing chunk-dispatch chokepoint (mirrors the Gemini-to-OpenAI translator's #4177 precedent for the same class of upstream error surfacing). Note: this baseline entry does NOT cover the separate pre-existing +11 drift already on the release tip from #8081/#8162 (1125->1136, unrelated reasoning-placeholder-stripping fix merged after this PR branched) — that drift belongs to the maintainer's rebaseline, not this PR.",
-  "_rebaseline_2026_07_22_8211_gemini_malformed_tool_choice": "PR #8211 (hartmark, fix/gemini-malformed-function-call-tool-choice) own growth: open-sse/translator/response/gemini-to-openai.ts 771->821 (+50, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds MALFORMED_FUNCTION_CALL/UNEXPECTED_TOOL_CALL handling inside geminiToOpenAIResponse(): synthesizes a `malformed_tool_call` tool_calls entry so finish_reason normalizes to the standard \\\"tool_calls\\\" instead of an unrecognized raw enum value that OpenAI-compatible clients (e.g. OpenClaw) silently ignore, and always synthesizes (rather than skipping when a real tool call already exists) so a malformed attempt alongside a real one in the same turn is not silently discarded. Irreducible cohesive addition at the existing candidate/finishReason translation chokepoint (mirrors the 9router#2462 raw-finish-reason precedent immediately below it in the same function). Covered by the PR's own tests/unit test additions for both the malformed-only and malformed-plus-real-call cases.",
-  "_rebaseline_2026_07_22_8213_chat_abandoned_target_abort": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/sse/handlers/chat.ts 1794->1860 (+66, measured against the PR's own merge-base — the release tip separately carries an unrelated -5 net shrink from #8013's antigravity callable-catalog alignment, which this PR's branch does not include and this entry does not cover). Adds resolveDispatchClientRawRequest(): merges a per-target modelAbortSignal into clientRawRequest.signal (via mergeAbortSignals) so a combo target abandoned by comboTargetTimeoutMs actually observes its own abort and reaches its cleanup path, instead of hanging forever inside withRateLimit/acquireAccountSemaphore and leaking a permanent 'pending' dashboard entry (live incident, log id 1784418258231-14961a). Also wires combo-exhausted rejection logging to capture request body + attempted models via the new rejectedRequestUsage helper. Irreducible additions at the existing chat dispatch chokepoint. Covered by the PR's own combo-config + integration test additions.",
-  "_rebaseline_2026_07_22_8213_combo_cooldown_wait_recording": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/combo.ts 3548->3604 (+56, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Fixes combo cooldown-wait state recording so a bogus 503 is no longer crystallized when the cooldown-wait vars reset every setTry, adds an OpenAI-format SSE error frame path for combo-exhausted rejections (capturing request body + attempted models), and gives an abandoned per-target dispatch its own timeout instead of leaking a permanent 'pending' dashboard entry. Irreducible additions at the existing handleComboChat dispatch/retry chokepoint (mirrors the prior quota-share/headroom/task-aware strategy-branch precedents already frozen in this file). Covered by the PR's own combo-config + Gemini TPM-ceiling benchmark test additions.",
-  "_rebaseline_2026_07_22_8213_gemini_tpm_quota_cooldown_wait": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: open-sse/services/accountFallback.ts 1857->1932 on the merged tip (release 1892 incl #8050 +35, plus this PR own growth +40); measured against the PR own merge-base was 1857->1898 (+41 — the release tip separately carries an unrelated +34 from #8050's antigravity 404 model-not-found lockout scoping, which this PR's branch does not include and this entry does not cover). Own growth is the Gemini TPM-ceiling classification + cooldown-wait wiring feeding into the combo cooldown-wait state machine (rate-limit wedge recovery) introduced by this PR's commit series. Irreducible additions at the existing account-fallback/model-lockout chokepoint. Covered by the PR's own gemini-rate-limit-tracker and TPM-ceiling benchmark test additions.",
-  "_rebaseline_2026_07_22_8213_health_unblock_model_cooldowns": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/app/(dashboard)/dashboard/health/page.tsx 1094->1165 (+71, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip). Adds handleUnblockAll/handleUnblockOne dashboard actions (DELETE /api/resilience/model-cooldowns) so an operator can manually clear a Gemini TPM-wedge model lockout surfaced by this PR's cooldown-wait fixes, instead of waiting out the ceiling. Irreducible UI wiring at the existing health-page action chokepoint. Covered by the PR's own dashboard/resilience test additions.",
-  "_rebaseline_2026_07_22_8213_requestloggerdetail_unblock_ui": "PR #8213 (hartmark, fix/gemini-tpm-quota-cooldown-wait) own growth: src/shared/components/RequestLoggerDetail.tsx 799->941 (+142, entirely this PR's diff — no other commit touched this file between the PR's merge-base and the release tip; crosses the general 800-line new-file cap so is frozen here for the first time). Adds a collapsible section header (open/expand-less toggle) plus per-log-entry unblock (`unblocking`/`cleared` state, isCombo503 detection) so the request-logger detail panel surfaces the same Gemini TPM cooldown-wait / model-lockout unblock action introduced by this PR at the individual-request level (mirrors the health-page bulk unblock action added in the same PR). Covered by the PR's own dashboard/resilience test additions.",
-  "_rebaseline_2026_07_22_fusion_8013_8098_antigravity": "Fusion of #8013 (backryun, catalog/IDE-CLI-split rewrite) + #8098 (nguyenha935, protocol-fidelity/fail-closed/credits/tool-cloaking): open-sse/services/usage/antigravity.ts NEW 802 (>cap 800, +2 — #8098 credits/tier usage service on #8013's profile-aware headers). Test growth (models-catalog-route 1605->1608, provider-models-route 1752->1757 from #8013 Gemini 3.6 catalog) tracked in testFrozen.",
-  "_rebaseline_2026_07_23_8127_grok_weekly_quota": "#8127 (@apoapostolov) own growth: src/sse/handlers/chat.ts 1861->1865 (+4) — weekly quota tracking for grok-web wires a quota-fetch hook at the existing dispatch chokepoint. Thin wiring mirroring adjacent provider-quota branches; not extractable. Covered by tests/unit/grok-quota-fetcher.test.ts.",
-  "_rebaseline_2026_07_23_8143_empty_catch_logging": "#8143 (@chirag127) own growth: open-sse/utils/stream.ts 2869->2887 (+18) — replacing empty catch blocks in the SSE stream subsystem with console.debug logging (Rule #6 silent-swallow fix, issues #8138-#8142). Cohesive logging additions at the existing catch chokepoints, not extractable; frozen at new size. Covered by tests/unit/stream-handler-catch-logging-8143.test.ts.",
-  "_rebaseline_2026_07_23_8219_cache_ttl_settings_sidebar": "#8219 (@oyi77) own growth: sections.ts 806->813 (+7) — configurable model-catalog cache-TTL settings adds a new sidebar nav entry + its visibility wiring. Sidebar item array is data, not extractable logic; frozen at new size.",
-  "_rebaseline_2026_07_23_8247_8248_model_unhealthy": "#8247+#8248 own growth: accountFallback.ts 1940->1941 (+1, irreducible import statement only — the substantive #8248 DEGRADED-pattern classifier was extracted into open-sse/config/errorConfig.ts, which has ample headroom, instead of growing this frozen file; #8247's fix is a single existing-line condition change, net zero lines). Scoping the credits-exhausted 403/429 branch to isCompatibleProvider() (per-model-quota openai/anthropic-compatible-* nicknames) so it stays model-scoped instead of terminalling the whole connection, and classifying NVIDIA NIM 'Function ... DEGRADED' 400 bodies as model-access-denied instead of a raw passthrough 400. Covered by tests/unit/8247-accountfallback-model-unhealthy.test.ts and tests/unit/8248-accountfallback-nvidia-degraded.test.ts.",
-  "_rebaseline_2026_07_23_8252_combo_400_advance": "#8252 (@RaviTharuma) own growth: accountFallback.ts 1932->1940 (+8) + combo.ts 3604->3630 (+26) — advance combo on model-scoped 400s wrapped as invalid/Bad-Request. Irreducible wiring at existing account-fallback + combo dispatch chokepoints. Covered by combo-model-scoped-400-advance.test.ts.",
-  "_rebaseline_2026_07_23_8266_alibaba_media": "#8266 (@backryun) own growth: imageRegistry.ts 821->979 (+158) — Alibaba-family media models (Qwen image/video, Bailian, Wan) added to the image/video registry. Registry model data, not extractable logic; frozen at new size.",
-  "_rebaseline_2026_07_24_8388_compression_detail_persist": "#8388 (compression engine DETAIL settings — Headroom/session-dedup/CCR — dropped on save) own growth: src/lib/db/compression.ts 866->872 (+6 = irreducible call-site wiring at the existing getCompressionSettings/updateCompressionSettings chokepoint: one import line, one `...buildDetailConfigDefaults()` spread in the seed config, and one `case \\\"sessionDedup\\\": case \\\"ccr\\\": applyDetailConfigUpdate(config, key, parsed); break;` load-switch case, mirroring the existing headroom/#8056 case immediately above it). The actual normalizer logic (normalizeSessionDedupConfig/normalizeCcrConfig, matching SESSION_DEDUP_SCHEMA/CCR_SCHEMA bounds) was EXTRACTED into a new leaf src/lib/db/compressionDetailNormalizers.ts (well under cap) so this frozen file only carries the minimal dispatch wiring. Covered by tests/unit/8388-compression-detail-persist.test.ts (schema-accept + full DB save->reload round-trip for both new sub-objects, plus a no-regression assertion on the existing headroom round-trip).",
-  "_rebaseline_2026_07_24_responses_toolcalls_log_summary": "hartmark, fix/responses-tool-calls-log-summary own growth: open-sse/translator/response/openai-responses.ts 1163->1174 (+11). closeToolCall() now also writes the completed tool call into the shared state.toolCalls Map (already populated by the openai-to-claude / claude-to-openai / gemini-to-openai response translators) so stream.ts's completion-log summary builder (which reads state.toolCalls, not this translator's own funcCallIds/funcNames/funcArgsBuf bookkeeping) reports finish_reason \\\"tool_calls\\\" and message.tool_calls for openai->openai-responses translated streams instead of always logging \\\"stop\\\" with no tool_calls — the actual client-facing SSE events were already correct; only the persisted call-log summary was wrong. Irreducible call-site addition at the existing tool-call-close chokepoint. Covered by the new regression test in tests/unit/translator-resp-openai-responses.test.ts.",
-  "_rebaseline_2026_07_25_8476_combo_input_bound_homogeneous_scope": "PR #8476 (herjarsa, fix/8375-8459-combo-image-fixes, #8375) own growth: open-sse/services/combo.ts 3642->3679 (+37 net: +29 the PR's own isInputBoundFailure short-circuit for deterministic context_length_exceeded/context_window_exceeded failures, +8 a /green-prs pre-merge fix scoping that short-circuit to homogeneous remainders only — the shipped code fired unconditionally on ANY target, regressing the intentional heterogeneous-combo fallback #6637/isContextOverflow400 protects, exactly as flagged by this PR's own review evidence but never actually implemented in the branch). The fix compares orderedTargets[i+1..] modelStr against the failing target's modelStr at the existing executeTarget dispatch chokepoint (mirrors the sameProviderNext precedent a few lines below) — irreducible call-site wiring, not extractable without hiding the dispatch boundary. Covered by tests/unit/combo-input-bound-failure-8375.test.ts (homogeneous pool still short-circuits) and the new tests/unit/combo-input-bound-heterogeneous-8375.test.ts (heterogeneous combo now correctly falls through to the larger-context target).",
-  "_rebaseline_2026_07_25_adobe_firefly_reference_images": "Follow-up to #8006: storage upload + referenceBlobs for image/video and /v1/images/edits dispatch. adobeFireflyClient.ts 1958->2317 (+upload helpers, extract sources, resolve blob ids). Note: 2317 not 2316 — check-file-size.mjs counts LOC via split(\\\"\\\\n\\\").length (counts the trailing-newline empty element), which is 1 higher than `wc -l` on a file ending in \\\\n; the PR's original entry (2316) was measured with wc -l and undercounted by 1 against the actual gate.",
-  "_rebaseline_pr1043_minimax_tts": "Upstream port decolua/9router#1043 (toanalien) own growth: audioSpeech.ts 965->1061 (+96). Adds MiniMax T2A v2 TTS dispatch (handleMinimaxSpeech + hexToBytes helper) — provider entry was already in audioRegistry (format: minimax-tts) but no handler existed, falling through to the OpenAI-compatible default that fails (T2A has custom shape + hex-encoded audio + base_resp envelope). New branch sits next to the other inline provider branches (xiaomi-mimo, coqui, tortoise, aws-polly) — extracting would just create indirection. Covered by tests/unit/minimax-tts-1043.test.ts (3 tests, GREEN: success, base_resp error, invalid-hex).",
-  "_rebaseline_pr4592_exclude_exhausted_auto": "Reconcile #4592 already-merged growth: combo.ts 2991->3036 (+45, terminal-status quota-cutoff exclusion in buildAutoCandidates + opt-in gate). Fast-gate PR->release does not run check:file-size.",
-  "open-sse/executors/antigravity.ts": "1528",
-  "open-sse/executors/base.ts": "1640",
-  "open-sse/executors/chatgpt-web.ts": "3241",
-  "open-sse/executors/codex.ts": "1562",
-  "open-sse/executors/cursor.ts": "1563",
-  "open-sse/executors/deepseek-web.ts": "1148",
-  "open-sse/executors/grok-web.ts": "1044",
-  "open-sse/executors/muse-spark-web.ts": "1405",
-  "open-sse/handlers/chatCore.ts": "5034",
-  "open-sse/handlers/imageGeneration.ts": "3101",
-  "open-sse/handlers/responseSanitizer.ts": "1128",
-  "open-sse/handlers/search.ts": "1536",
-  "open-sse/handlers/videoGeneration.ts": "1063",
-  "open-sse/mcp-server/schemas/tools.ts": "1553",
-  "open-sse/mcp-server/server.ts": "1448",
-  "open-sse/mcp-server/tools/advancedTools.ts": "1120",
-  "open-sse/services/accountFallback.ts": "1978",
-  "open-sse/services/adobeFireflyClient.ts": "2385",
-  "open-sse/services/claudeCodeCompatible.ts": "1202",
-  "open-sse/services/combo.ts": "3648",
-  "open-sse/services/compression/strategySelector.ts": "1060",
-  "open-sse/services/rateLimitManager.ts": "1167",
-  "open-sse/translator/response/openai-responses.ts": "1204",
-  "open-sse/utils/cursorAgentProtobuf.ts": "1505",
-  "open-sse/utils/stream.ts": "2889",
-  "src/app/(dashboard)/dashboard/HomePageClient.tsx": "1388",
-  "src/app/(dashboard)/dashboard/analytics/ComboHealthTab.tsx": "1031",
-  "src/app/(dashboard)/dashboard/api-manager/ApiManagerPageClient.tsx": "3117",
-  "src/app/(dashboard)/dashboard/cache/media/MediaPageClient.tsx": "1067",
-  "src/app/(dashboard)/dashboard/combos/page.tsx": "4703",
-  "src/app/(dashboard)/dashboard/costs/CostOverviewTab.tsx": "1283",
-  "src/app/(dashboard)/dashboard/costs/quota-share/components/PoolWizard.tsx": "1022",
-  "src/app/(dashboard)/dashboard/endpoint/EndpointPageClient.tsx": "2615",
-  "src/app/(dashboard)/dashboard/health/page.tsx": "1165",
-  "src/app/(dashboard)/dashboard/providers/[id]/components/modals/EditConnectionModal.tsx": "1324",
-  "src/app/(dashboard)/dashboard/providers/page.tsx": "1944",
-  "src/app/(dashboard)/dashboard/runtime/RuntimePageClient.tsx": "1201",
-  "src/app/(dashboard)/dashboard/settings/components/PricingTab.tsx": "1019",
-  "src/app/(dashboard)/dashboard/settings/components/ProxyRegistryManager.tsx": "1470",
-  "src/app/(dashboard)/dashboard/settings/components/ResilienceTab.tsx": "1123",
-  "src/app/(dashboard)/dashboard/settings/components/RoutingTab.tsx": "1629",
-  "src/app/(dashboard)/dashboard/settings/components/SystemStorageTab.tsx": "1573",
-  "src/app/(dashboard)/dashboard/usage/components/BudgetTab.tsx": "1028",
-  "src/app/(dashboard)/dashboard/usage/components/EvalsTab.tsx": "2148",
-  "src/app/(dashboard)/dashboard/usage/components/ProviderLimits/index.tsx": "1119",
-  "src/app/api/providers/[id]/models/route.ts": "2361",
-  "src/app/api/v1/models/catalog.ts": "1590",
-  "src/lib/tokenHealthCheck.ts": "1053",
-  "src/lib/db/apiKeys.ts": "1529",
-  "src/lib/db/core.ts": "1639",
-  "src/lib/db/migrationRunner.ts": "1094",
-  "src/lib/db/models.ts": "1097",
-  "src/lib/db/providers.ts": "1034",
-  "src/lib/memory/retrieval.ts": "1073",
-  "src/lib/tailscaleTunnel.ts": "1202",
-  "src/lib/usage/providerLimits.ts": "1013",
-  "src/shared/components/OAuthModal.tsx": "1134",
-  "src/shared/components/RequestLoggerV2.tsx": "1629",
-  "src/shared/components/analytics/charts.tsx": "1035",
-  "src/shared/services/cliRuntime.ts": "1122",
-  "src/sse/handlers/chat.ts": "1904",
-  "src/sse/services/auth.ts": "2508",
-  "tests/unit/account-fallback-service.test.ts": "1572",
-  "tests/unit/provider-validation-specialty.test.ts": "2985",
-  "open-sse/executors/hyperagent.ts": "1026",
-  "open-sse/executors/default.ts": "1042",
-  "open-sse/executors/kiro.ts": "1069",
-  "open-sse/translator/request/openai-to-kiro.ts": "1057",
-  "open-sse/utils/sseHeartbeat.ts": "142",
-  "_rebaseline_2026_08_04_9305_sse_comments": "#9305 fix: broadened sseCommentsEnabled()"
+  "_rebaseline_2026_08_07_v3850_sweepreds_pr2_toolnamemap": "tests/unit/translator-openai-to-gemini.test.ts 1616->1619 (+3). O frozen estava EXATAMENTE no tamanho da base, entao qualquer linha nova viola. #9568 (c9a3361e5a) fez buildChangedToolNameMap emitir entradas IDENTIDADE (o Gemini minusculiza nomes de tool nas respostas, entao o tradutor de resposta precisa da chave para mapear de volta), o que passou a incluir `_toolNameMap` no envelope Antigravity de qualquer request com tools. As 3 linhas sao: a chave nova na lista esperada de Object.keys, 1 comentario explicando POR QUE ela aparece (sem ele o proximo leitor tenta remove-la de novo) e 1 assert do CONTEUDO do map — presenca de chave sozinha nao provaria a entrada identidade, que e justamente o comportamento novo. Nao ha o que extrair: e alinhamento de contrato dentro de um teste existente."
 }

docs/frameworks/RADAR.md

@@ -1,13 +1,13 @@
 ---
 title: "Radar Free-Model Catalog"
 version: 3.8.50
-lastUpdated: 2026-08-07
+lastUpdated: 2026-08-08
 ---
 
 # Radar Free-Model Catalog
 
 > **Source of truth:** `src/lib/radar/`, `src/lib/db/radar.ts`, `src/app/api/radar/`
-> **Last updated:** 2026-08-07 — v3.8.50
+> **Last updated:** 2026-08-08 — v3.8.50
 
 Radar is an **optional add-on** that overlays a signed, freshly-curated free-model
 catalog on top of the release baseline (`FREE_MODEL_BUDGETS` in
@@ -23,6 +23,23 @@ below.
 
 ---
 
+## Delivery status in v3.8.50
+
+The following status distinguishes what this OSS release implements from later Radar
+workstreams. It is a code-level status, not a promise that a particular hosted deployment
+or external integration is currently available.
+
+| Area                             | Status in this release                                                                                                                                                                                                 |
+| -------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| Signed catalog client            | Implemented behind `RADAR_ENABLED`, with separate opt-in, Ed25519 verification, local encrypted settings/cache, non-destructive overlay, scheduler, and dashboard.                                                     |
+| Contributor activation           | The dashboard links to the server-hosted GitHub claim flow and accepts an existing `omr_…` key. Contributor eligibility is resolved by the private service; the OSS client contains no GitHub token or issuance logic. |
+| Supporter-key activation         | Implemented. The raw key is validated, encrypted at rest, masked on reads, and sent only by the server-side sync. Changing or clearing the key invalidates both entitlement-sensitive feed caches.                     |
+| Referral links                   | Implemented as a separately signed, hourly-refreshed feed. Fixed links are available to the community tier immediately; limited campaigns remain live-tier data.                                                       |
+| Payments and transactional email | Not implemented in the OSS client. Purchase, donation, receipt review, and mail delivery belong to the private service and its later operational workstream.                                                           |
+| Research-agent workstream        | Not part of this client release. Curated feed contents remain server-side data; no autonomous research agent runs in an OmniRoute installation.                                                                        |
+
+---
+
 ## Flag: `RADAR_ENABLED` (default off)
 
 Radar is gated end-to-end by the `RADAR_ENABLED` feature flag
@@ -67,7 +84,8 @@ When both are on, the sync path is:
    plain, unauthenticated-by-default GET. OmniRoute never posts usage data, provider
    configuration, or model traffic to the feed service.
 3. The response is verified, validated, and cached locally (see
-   [Security model](#security-model)). Nothing else touches the network for Radar.
+   [Security model](#security-model)). Radar has exactly two server-side network paths:
+   `syncRadar()` for the catalog and `syncRadarReferrals()` for the standalone referrals feed.
 
 The **supporter key** is an optional Bearer token (`radar_settings.supporter_key`)
 that lets the feed service decide which tier to serve (see
@@ -77,6 +95,9 @@ that lets the feed service decide which tier to serve (see
   helpers (`src/lib/db/encryption.ts`) used for provider credentials.
 - Set via `POST /api/radar/settings` (`{ supporterKey: "omr_" + 40 hex chars }`) and
   **never echoed back** — the response returns a masked form (`omr_****abcd`).
+- Changing or clearing it atomically invalidates both the catalog and referrals caches. The
+  next sync/read resolves the new entitlement server-side; saving a key does not itself make
+  a network request or consume a single-use activation key.
 - Sent to the feed service as a Bearer token on the sync GET — nothing else about the
   key ever leaves the client.
 
@@ -102,10 +123,10 @@ pattern as `RADAR_FEED_URL`) and relayed to the dashboard through the existing
 `GET /api/radar/settings` response (`contributorClaimUrl`, `supporterPlansUrl`) — the
 client component never reads `process.env` itself.
 
-| Var                             | Purpose                                                                                      |
-| -------------------------------- | ---------------------------------------------------------------------------------------------- |
-| `RADAR_CONTRIBUTOR_CLAIM_URL`   | Overrides the contributor-claim URL (default `https://radar.omniroute.online/auth/github`).   |
-| `RADAR_SUPPORTER_PLANS_URL`     | Overrides the supporter-plans URL (default `https://radar.omniroute.online/planos`).          |
+| Var                           | Purpose                                                                                     |
+| ----------------------------- | ------------------------------------------------------------------------------------------- |
+| `RADAR_CONTRIBUTOR_CLAIM_URL` | Overrides the contributor-claim URL (default `https://radar.omniroute.online/auth/github`). |
+| `RADAR_SUPPORTER_PLANS_URL`   | Overrides the supporter-plans URL (default `https://radar.omniroute.online/planos`).        |
 
 Once a visitor has a key (`omr_` + 40 hex chars), the activation screen
 (`src/app/(dashboard)/dashboard/radar/page.tsx`) has a paste-key input as the primary
@@ -117,7 +138,7 @@ as a UX nicety; the server's Zod schema is the authoritative check either way. O
 key is set, the activation screen shows the masked form (`supporterKeyMasked` from
 `GET /api/radar/settings`) instead of an empty input, with a "change key" control to
 paste a new one — the raw key is never redisplayed. The two claim/plans buttons above
-remain the way to *obtain* a key in the first place; this input is where an operator
+remain the way to _obtain_ a key in the first place; this input is where an operator
 who already has one activates it.
 
 ---
@@ -207,11 +228,11 @@ handle.
 ### The served tier comes from a response header, not the signed body
 
 The signed feed **body**'s `tier` field is always `"live"` — the feed service ships
-**one signed artifact per version**, so the body cannot carry a per-request tier
-without invalidating the Ed25519 signature (re-signing per request would defeat the
-point of a pinned, cacheable, verifiable artifact). The tier actually served for a
-given request is instead carried in the **`x-omniroute-feed-tier` response header**,
-decided server-side from the request's `Authorization` key.
+**two signed artifacts per version**: live includes current campaigns and community
+omits them. Each artifact is signed over its own exact bytes. The body still does not
+serve as the entitlement decision; the tier actually selected for a request is carried
+in the **`x-omniroute-feed-tier` response header**, decided server-side from the request's
+`Authorization` key.
 
 `syncRadar()` (`src/lib/radar/sync.ts::parseServedTierHeader()`) is the single place
 that resolves the tier a client should trust:
@@ -223,7 +244,7 @@ that resolves the tier a client should trust:
 2. Fall back to the signed body's `tier` field (always `"live"`) only when step 1
    yields nothing.
 3. The resolved tier is what gets cached and returned as `{ status: "updated",
-   version, tier }` — this is the value the dashboard shows, never the raw body
+version, tier }` — this is the value the dashboard shows, never the raw body
    field.
 
 ---
@@ -265,18 +286,18 @@ Every merged entry carries an `origin` field the UI renders as a badge:
 
 Five local routes back the UI, all under `src/app/api/radar/`:
 
-| Route                   | Method | Purpose                                                                                          |
-| ----------------------- | ------ | -------------------------------------------------------------------------------------------------- |
-| `/api/radar/catalog`    | GET    | Returns the merged catalog (`getRadarCatalog()`) from the local cache.                            |
-| `/api/radar/sync`       | POST   | Triggers `syncRadar()` server-side; returns the resulting status.                                 |
-| `/api/radar/settings`   | GET    | Returns `{ optIn, hasSupporterKey, supporterKeyMasked }` — never the raw key.                     |
-| `/api/radar/settings`   | POST   | Sets opt-in and/or the (encrypted) supporter key.                                                 |
-| `/api/radar/referrals`  | GET    | Returns `{ fixed, campaigns, tier }` from the local cache — see [Referral links](#referral-links-free-credits) below. |
+| Route                  | Method | Purpose                                                                                                               |
+| ---------------------- | ------ | --------------------------------------------------------------------------------------------------------------------- |
+| `/api/radar/catalog`   | GET    | Returns the merged catalog (`getRadarCatalog()`) from the local cache.                                                |
+| `/api/radar/sync`      | POST   | Triggers `syncRadar()` server-side; returns the resulting status.                                                     |
+| `/api/radar/settings`  | GET    | Returns `{ optIn, hasSupporterKey, supporterKeyMasked }` — never the raw key.                                         |
+| `/api/radar/settings`  | POST   | Sets opt-in and/or the (encrypted) supporter key.                                                                     |
+| `/api/radar/referrals` | GET    | Returns `{ fixed, campaigns, tier }` from the local cache — see [Referral links](#referral-links-free-credits) below. |
 
 **Hard rule: these routes never proxy the feed service.** The browser only ever talks
-to the local OmniRoute server; `syncRadar()` is the single module in the whole client
-that touches the network for Radar (`src/lib/radar/sync.ts`), and it always runs
-server-side, never client-side. This keeps the feed URL and any supporter key
+to the local OmniRoute server. The two modules that touch the Radar service are
+`src/lib/radar/sync.ts` (catalog) and `src/lib/radar/referralsSync.ts` (referrals); both
+always run server-side, never client-side. This keeps the feed URL and any supporter key
 out of client-facing network traffic entirely.
 
 All five routes return `404` when `RADAR_ENABLED` is off (see
@@ -348,11 +369,13 @@ touches the network for referrals, mirroring `syncRadar()`'s contract exactly: f
 Ed25519 signature over the exact response bytes (`verifyFeedBytes`), validates against
 `RadarReferralsFeedSchema`, and caches into the `radar_referrals_cache` table
 (migration `142_radar_referrals_cache.sql`) — a table entirely separate from the
-catalog's `radar_feed_cache`. A 10 MB response cap and a `generatedAt` floor (an
-incoming feed with a `generatedAt` no newer than the cached one is treated as `stale`
-and never overwrites the cache — guards against a replay of an older signed artifact)
-mirror the catalog sync's own `MAX_FEED_BYTES`/version-floor guards. Never throws —
-always returns a status object; errors never carry a stack trace in `reason`.
+catalog's `radar_feed_cache`. A 10 MB response cap and a `generatedAt` floor reject an
+incoming feed older than the cached one, guarding against replay of an older signed
+artifact. An equal timestamp is accepted: the server intentionally gives the community
+and live referral variants the same deterministic `generatedAt`, so the signed payload
+and served tier can change after a supporter-key change without the underlying link set
+changing. Never throws — always returns a status object; errors never carry a stack trace
+in `reason`.
 
 Two triggers keep the referrals cache warm, both independent of the catalog's own
 24h cadence:

open-sse/config/audioRegistry.ts

@@ -603,7 +603,7 @@ export interface ProviderNodeRow {
 }
 
 /** Hosts reachable only from the operator's machine/Docker network. */
-export function isLoopbackNodeHost(baseUrl: string): boolean {
+function isLoopbackNodeHost(baseUrl: string): boolean {
   try {
     const hostname = new URL(baseUrl).hostname;
     return (

open-sse/executors/lmarena/response.ts

@@ -7,8 +7,6 @@ import { isCloudflareChallenge } from "../../services/lmarenaTlsClient.ts";
 import { markLMArenaCatalogModelDead } from "./models.ts";
 import { parseArenaSSE } from "./stream.ts";
 
-const encoder = new TextEncoder();
-
 export function errorResponse(
   status: number,
   message: string,
@@ -167,7 +165,7 @@ function baseChunk(model: string) {
 }
 
 function enqueueSse(controller: ReadableStreamDefaultController, chunk: Record<string, unknown>) {
-  controller.enqueue(encoder.encode(`data: ${JSON.stringify(chunk)}\n\n`));
+  controller.enqueue(new TextEncoder().encode(`data: ${JSON.stringify(chunk)}\n\n`));
 }
 
 function emitStopAndDone(controller: ReadableStreamDefaultController, model: string) {
@@ -175,8 +173,7 @@ function emitStopAndDone(controller: ReadableStreamDefaultController, model: str
     ...baseChunk(model),
     choices: [{ index: 0, delta: {}, finish_reason: "stop" }],
   });
-
-  controller.enqueue(encoder.encode("data: [DONE]\n\n"));
+  controller.enqueue(new TextEncoder().encode("data: [DONE]\n\n"));
   controller.close();
 }
 

open-sse/translator/request/openai-responses.ts

@@ -724,7 +724,7 @@ export function openaiResponsesToOpenAIRequest(
     const reasoningRec = toRecord(root.reasoning);
     const effort = toString(reasoningRec.effort);
     if (effort && result.reasoning_effort === undefined) {
-      result.reasoning_effort = normalizeResponsesReasoningEffort(effort, model ?? root.model);
+      result.reasoning_effort = normalizeResponsesReasoningEffort(effort, model);
     }
     if (
       credentialRecord._copilotClient === true &&

open-sse/utils/sseHeartbeat.ts

@@ -79,8 +79,7 @@ export function sseCommentsEnabled(): boolean {
   if (typeof process === "undefined") return true;
   const v = process.env.OMNIROUTE_SSE_COMMENTS;
   if (v === undefined || v === "") return true;
-  const normalized = v.trim().toLowerCase();
-  return normalized !== "off" && normalized !== "false" && normalized !== "0" && normalized !== "no";
+  return v.trim().toLowerCase() !== "off";
 }
 
 export function createSseHeartbeatTransform({

open-sse/utils/stream.ts

@@ -25,7 +25,6 @@ import {
 } from "./streamHelpers.ts";
 import { calculateCost } from "@/lib/usage/costCalculator";
 import { buildOmniRouteSseMetadataComment } from "@/domain/omnirouteResponseMeta";
-import { sseCommentsEnabled } from "./sseHeartbeat.ts";
 import {
   createStructuredSSECollector,
   buildStreamSummaryFromEvents,
@@ -1002,11 +1001,6 @@ export function createSSEStream(options: StreamOptions = {}) {
     controller: TransformStreamDefaultController,
     finalUsage: UsageTokenRecord | Record<string, unknown> | null | undefined
   ) => {
-    // Skip SSE metadata comment lines when OMNIROUTE_SSE_COMMENTS is disabled
-    // (e.g., "off", "false", "0", "no"). Strict OpenAI-compatible clients that
-    // JSON.parse every SSE line will crash on `: x-omniroute-*` comment lines.
-    if (!sseCommentsEnabled()) return;
-
     const costUsd = finalUsage ? await calculateCost(provider, model, finalUsage) : 0;
     const comment = buildOmniRouteSseMetadataComment({
       provider,

scripts/build/colocateOptionals.mjs

@@ -47,8 +47,7 @@
  */
 
 import { cpSync, existsSync, mkdirSync, readFileSync } from "node:fs";
-import { createRequire } from "node:module";
-import { dirname, join, sep } from "node:path";
+import { dirname, join } from "node:path";
 
 /**
  * Entry packages of the SLM optional stack (the closure roots). `@huggingface/transformers` is
@@ -97,33 +96,6 @@ export function computeDependencyClosure(nodeModulesDir, seeds = SEED_PACKAGES)
   return closure;
 }
 
-/**
- * A package in the target tree counts as PRESENT only when its entrypoint
- * resolves from inside that tree — the same contract the Dockerfile's
- * post-build guard enforces. Next's file tracing can materialize a package
- * PARTIALLY (the package.json lands, the files its `main` points at do not),
- * and a directory-level `existsSync` check then skips the package forever
- * while the runtime dies with "Cannot find module <pkg>/dist/index.js".
- *
- * @param {string} targetNodeModulesDir
- * @param {string} name
- * @returns {boolean}
- */
-function isPackageIntact(targetNodeModulesDir, name) {
-  if (!existsSync(join(targetNodeModulesDir, name))) return false;
-  try {
-    const probe = createRequire(
-      join(targetNodeModulesDir, "__colocate_probe__.js")
-    );
-    const resolved = probe.resolve(name);
-    // A resolution that walked past the target into an ancestor tree does not
-    // prove the target copy is usable.
-    return resolved.startsWith(targetNodeModulesDir + sep);
-  } catch {
-    return false;
-  }
-}
-
 /**
  * Co-locate the SLM optional dependency closure from `<rootDir>/node_modules`
  * into a standalone bundle's `node_modules`.
@@ -170,12 +142,11 @@ export function colocateLlmlinguaOptionals({
 
   const closure = computeDependencyClosure(rootNm, seeds);
 
-  // Check the complete closure rather than only the entry package, and judge
-  // presence by entrypoint integrity — a partially traced directory (see
-  // isPackageIntact) must still receive its missing files.
+  // Check the complete closure rather than only the entry package. A partially
+  // populated bundle must still receive any missing transitive dependencies.
   if (
     closure.length > 0 &&
-    closure.every((name) => isPackageIntact(targetNm, name))
+    closure.every((name) => existsSync(join(targetNm, name)))
   ) {
     return { skipped: true, reason: "already co-located" };
   }
@@ -184,18 +155,11 @@ export function colocateLlmlinguaOptionals({
 
   for (const name of closure) {
     const dest = join(targetNm, name);
-    if (isPackageIntact(targetNm, name)) continue;
+    if (existsSync(dest)) continue;
 
     try {
       mkdirSync(dirname(dest), { recursive: true });
-      // force:false merges into a partially traced directory: files the trace
-      // already materialized are kept, missing ones (the package payload) are
-      // filled in from the root tree.
-      cpSync(join(rootNm, name), dest, {
-        recursive: true,
-        force: false,
-        errorOnExist: false,
-      });
+      cpSync(join(rootNm, name), dest, { recursive: true });
       copied++;
     } catch (err) {
       log(

src/app/(dashboard)/dashboard/radar/setup/page.tsx

@@ -1,10 +1,16 @@
 "use client";
 
-import { useState, useEffect, useCallback, useMemo } from "react";
-import { useTranslations } from "next-intl";
+import { useState, useEffect, useCallback } from "react";
+import { useLocale, useTranslations } from "next-intl";
 import { useSearchParams } from "next/navigation";
 import Link from "next/link";
 import { Card } from "@/shared/components";
+import {
+  firstProviderConnectionId,
+  providerConnectionsRequestUrl,
+  type RadarSetupConnection,
+} from "@/lib/radar/setupConnections";
+import type { RadarLocalizedText } from "@/lib/radar/feedSchema";
 
 // ---------------------------------------------------------------------------
 // Types
@@ -14,17 +20,16 @@ import { Card } from "@/shared/components";
  * Localized text: either a plain string or an {en, pt?} object.
  * The renderer resolves the best locale with EN fallback (D25 compat).
  */
-type LocalizedText = string | { en: string; pt?: string };
-
 interface SetupInfo {
   keyUrl: string | null;
-  steps: LocalizedText[];
+  steps: RadarLocalizedText[];
 }
 
 interface ProviderSetupData {
   provider: string;
   setup: SetupInfo | null;
   configured: boolean;
+  connectionId: string | null;
 }
 
 // ---------------------------------------------------------------------------
@@ -32,9 +37,9 @@ interface ProviderSetupData {
 // ---------------------------------------------------------------------------
 
 /** Resolve a LocalizedText to a display string. */
-function resolveText(text: LocalizedText, locale: string): string {
+function resolveText(text: RadarLocalizedText, locale: string): string {
   if (typeof text === "string") return text;
-  if (locale === "pt" && text.pt) return text.pt;
+  if (locale.toLowerCase().startsWith("pt") && text.pt) return text.pt;
   return text.en;
 }
 
@@ -44,9 +49,9 @@ function resolveText(text: LocalizedText, locale: string): string {
 
 export default function RadarSetupPage() {
   const t = useTranslations("radarSetupPage");
+  const locale = useLocale();
   const searchParams = useSearchParams();
   const provider = searchParams.get("provider");
-  const locale = "en"; // Could be derived from next-intl locale later
 
   const [setupData, setSetupData] = useState<ProviderSetupData | null>(null);
   const [loading, setLoading] = useState(true);
@@ -63,18 +68,25 @@ export default function RadarSetupPage() {
 
     async function load() {
       try {
-        const res = await fetch("/api/radar/catalog");
+        const [res, connectionsRes] = await Promise.all([
+          fetch("/api/radar/catalog"),
+          fetch(providerConnectionsRequestUrl(provider)),
+        ]);
         if (res.status === 404) {
           setError(t("flagDisabled"));
           setLoading(false);
           return;
         }
         if (!res.ok) throw new Error(`HTTP ${res.status}`);
+        if (!connectionsRes.ok) throw new Error(`HTTP ${connectionsRes.status}`);
         const data = await res.json();
+        const connectionsData = (await connectionsRes.json()) as {
+          connections?: RadarSetupConnection[];
+        };
 
         // Find ALL entries for this provider and extract setup from the first one that has it
         const providerEntries = data.entries.filter(
-          (e: { provider: string }) => e.provider === provider,
+          (e: { provider: string }) => e.provider === provider
         );
 
         if (providerEntries.length === 0) {
@@ -85,17 +97,19 @@ export default function RadarSetupPage() {
 
         // Find setup info from feed entries (they carry the setup field)
         const entryWithSetup = providerEntries.find(
-          (e: { setup?: SetupInfo | null }) => e.setup && (e.setup.steps.length > 0 || e.setup.keyUrl),
+          (e: { setup?: SetupInfo | null }) =>
+            e.setup && (e.setup.steps.length > 0 || e.setup.keyUrl)
         );
 
-        // Check if provider is configured (has connections)
-        // We infer this from whether the provider exists in the catalog at all
-        // The actual connection check would need a separate API — for now we show
-        // the guide regardless
+        const connectionId = firstProviderConnectionId(
+          Array.isArray(connectionsData.connections) ? connectionsData.connections : [],
+          provider
+        );
         setSetupData({
           provider,
           setup: entryWithSetup?.setup ?? null,
-          configured: false, // Will be enriched when connection-status API is available
+          configured: connectionId !== null,
+          connectionId,
         });
       } catch (err) {
         setError(err instanceof Error ? err.message : t("loadFailed"));
@@ -109,15 +123,11 @@ export default function RadarSetupPage() {
 
   // Test connection — uses the EXISTING connection-test endpoint
   const handleTestConnection = useCallback(async () => {
-    if (!provider) return;
+    if (!setupData?.connectionId) return;
     setTesting(true);
     setTestResult(null);
     try {
-      // The existing test endpoint is POST /api/providers/[id]/test
-      // We need the connection ID — for now we use the provider ID as a proxy.
-      // In a full implementation, the setup page would list connections for
-      // this provider and test each one. Here we test the first connection.
-      const res = await fetch(`/api/providers/${encodeURIComponent(provider)}/test`, {
+      const res = await fetch(`/api/providers/${encodeURIComponent(setupData.connectionId)}/test`, {
         method: "POST",
         headers: { "Content-Type": "application/json" },
         body: JSON.stringify({}),
@@ -136,7 +146,7 @@ export default function RadarSetupPage() {
     } finally {
       setTesting(false);
     }
-  }, [provider, t]);
+  }, [setupData?.connectionId, t]);
 
   if (!provider) {
     return (
@@ -165,9 +175,7 @@ export default function RadarSetupPage() {
         <p className="text-sm text-text-muted mt-1">{t("setupSubtitle")}</p>
       </div>
 
-      {error && (
-        <div className="p-3 rounded-lg bg-red-500/10 text-red-400 text-sm">{error}</div>
-      )}
+      {error && <div className="p-3 rounded-lg bg-red-500/10 text-red-400 text-sm">{error}</div>}
 
       {loading ? (
         <div className="flex items-center justify-center min-h-[200px]">
@@ -248,15 +256,13 @@ export default function RadarSetupPage() {
               <div className="flex items-center gap-3">
                 <button
                   onClick={handleTestConnection}
-                  disabled={testing}
+                  disabled={testing || !setupData.connectionId}
                   className="px-4 py-2 text-sm font-medium rounded-lg border border-violet-500 text-violet-400 hover:bg-violet-500/10 transition-colors disabled:opacity-50"
                 >
                   {testing ? t("testing") : t("testButton")}
                 </button>
                 {testResult && (
-                  <span
-                    className={`text-sm ${testResult.ok ? "text-green-400" : "text-red-400"}`}
-                  >
+                  <span className={`text-sm ${testResult.ok ? "text-green-400" : "text-red-400"}`}>
                     {testResult.message}
                   </span>
                 )}

src/app/api/memory/[id]/route.ts

@@ -1,10 +1,6 @@
 import { NextResponse } from "next/server";
 import { requireManagementAuth } from "@/lib/api/requireManagementAuth";
-// Import through the module index, NOT "@/lib/memory/manager" directly: the index's
-// import-time side effect is what calls memoryManager.register(sqliteBackend). Importing
-// the bare manager gives an EMPTY registry, so every handler here threw
-// `Primary backend "sqlite" not registered` and returned 500 (#8752).
-import { memoryManager } from "@/lib/memory";
+import { memoryManager } from "@/lib/memory/manager";
 import { validateBody, isValidationFailure } from "@/shared/validation/helpers";
 import { MemoryUpdatePutSchema } from "@/shared/schemas/memory";
 import { sanitizeErrorMessage } from "@omniroute/open-sse/utils/error.ts";

src/app/api/plugins/marketplace/install/route.ts

@@ -1,14 +1,9 @@
 import { NextRequest, NextResponse } from "next/server";
-import { z } from "zod";
 import { CORS_HEADERS, handleCorsOptions } from "@/shared/utils/cors";
 import { buildErrorBody } from "@omniroute/open-sse/utils/error";
 import { installMarketplacePlugin } from "@/lib/plugins/marketplace";
 import { requireManagementAuth } from "@/lib/api/requireManagementAuth";
 
-const InstallBodySchema = z.object({
-  name: z.string().trim().min(1),
-});
-
 export async function OPTIONS() {
   return handleCorsOptions();
 }
@@ -20,14 +15,15 @@ export async function POST(request: NextRequest) {
   const authError = await requireManagementAuth(request);
   if (authError) return authError;
   try {
-    const parsed = InstallBodySchema.safeParse(await request.json());
-    if (!parsed.success) {
+    const body = await request.json();
+    const { name } = body as { name?: string };
+    if (!name || typeof name !== "string") {
       return NextResponse.json(buildErrorBody(400, "Missing or invalid 'name' field"), {
         status: 400,
         headers: CORS_HEADERS,
       });
     }
-    const result = await installMarketplacePlugin(parsed.data.name);
+    const result = await installMarketplacePlugin(name);
     return NextResponse.json(result, { status: 201, headers: CORS_HEADERS });
   } catch (err: unknown) {
     const msg = err instanceof Error ? err.message : "Failed to install marketplace plugin";

src/app/api/services/dario/admin/accounts/route.ts

@@ -11,15 +11,9 @@
  * and never reaches the browser.
  */
 
-import { z } from "zod";
-
 import { forwardToDarioAdmin, requireAdminAuth } from "../_lib";
 import { createErrorResponse } from "@/lib/api/errorResponse";
 
-const DeleteAccountBodySchema = z.object({
-  alias: z.string().trim().min(1).optional(),
-});
-
 export async function GET(request: Request): Promise<Response> {
   const authResponse = await requireAdminAuth(request);
   if (authResponse) return authResponse;
@@ -35,9 +29,9 @@ export async function DELETE(request: Request): Promise<Response> {
 
   if (!alias && request.body !== null) {
     try {
-      const parsed = DeleteAccountBodySchema.safeParse(await request.json());
-      if (parsed.success && parsed.data.alias) {
-        alias = parsed.data.alias;
+      const parsed = await request.json();
+      if (parsed && typeof parsed === "object" && typeof (parsed as { alias?: unknown }).alias === "string") {
+        alias = (parsed as { alias: string }).alias.trim();
       }
     } catch {
       /* fall through to the missing-alias error below */

src/app/api/services/dario/admin/import-from-omniroute/route.ts

@@ -28,7 +28,6 @@
  * pickup, rather than relying on any undocumented hot-reload behavior.
  */
 
-import { z } from "zod";
 import { NextResponse } from "next/server";
 import fs from "node:fs";
 import path from "node:path";
@@ -40,11 +39,6 @@ import { getDarioHomeDir } from "@/lib/services/installers/dario";
 import { createErrorResponse } from "@/lib/api/errorResponse";
 import { sanitizeErrorMessage } from "@omniroute/open-sse/utils/error";
 
-const ImportBodySchema = z.object({
-  connectionId: z.string().trim().min(1).optional(),
-  alias: z.string().trim().min(1).optional(),
-});
-
 const ALIAS_PATTERN = /^[A-Za-z0-9][A-Za-z0-9_\-.]{0,63}$/;
 
 function safeAliasFromSource(email: string | null | undefined, connectionId: string): string {
@@ -88,16 +82,15 @@ export async function POST(request: Request): Promise<Response> {
   const authResponse = await requireAdminAuth(request);
   if (authResponse) return authResponse;
 
-  let raw: unknown;
+  let body: unknown;
   try {
-    raw = await request.json();
+    body = await request.json();
   } catch {
     return createErrorResponse({ status: 400, message: "Invalid JSON body" });
   }
 
-  const parsed = ImportBodySchema.safeParse(raw ?? {});
-  const b = parsed.success ? parsed.data : {};
-  const connectionId = b.connectionId ?? null;
+  const b = (body || {}) as Record<string, unknown>;
+  const connectionId = typeof b.connectionId === "string" ? b.connectionId : null;
   if (!connectionId) {
     return createErrorResponse({ status: 400, message: "connectionId is required" });
   }
@@ -119,7 +112,10 @@ export async function POST(request: Request): Promise<Response> {
     });
   }
 
-  let alias = b.alias || safeAliasFromSource(conn.email as string | null, connectionId);
+  let alias =
+    typeof b.alias === "string" && b.alias.trim()
+      ? b.alias.trim()
+      : safeAliasFromSource(conn.email as string | null, connectionId);
   if (!ALIAS_PATTERN.test(alias)) {
     alias = safeAliasFromSource(conn.email as string | null, connectionId);
   }

src/app/api/services/dario/admin/login-start/route.ts

@@ -8,29 +8,23 @@
  * posts the displayed code to /login-complete.
  */
 
-import { z } from "zod";
 import { forwardToDarioAdmin, requireAdminAuth } from "../_lib";
 import { createErrorResponse } from "@/lib/api/errorResponse";
 
-const LoginStartBodySchema = z.object({
-  alias: z.string().trim().min(1).optional(),
-});
-type LoginStartBody = z.infer<typeof LoginStartBodySchema>;
-
 export async function POST(request: Request): Promise<Response> {
   const authResponse = await requireAdminAuth(request);
   if (authResponse) return authResponse;
 
-  let body: LoginStartBody = {};
+  let body: { alias?: string } = {};
   try {
     if (request.body !== null) {
-      const parsed = LoginStartBodySchema.safeParse(await request.json());
-      if (parsed.success) body = parsed.data;
+      const parsed = await request.json();
+      if (parsed && typeof parsed === "object") body = parsed as { alias?: string };
     }
   } catch {
     return createErrorResponse({ status: 400, message: "Invalid JSON body" });
   }
 
-  const forwardBody = body.alias ? { alias: body.alias } : {};
+  const forwardBody = typeof body.alias === "string" && body.alias.trim() ? { alias: body.alias.trim() } : {};
   return forwardToDarioAdmin({ method: "POST", path: "/admin/login/start", body: forwardBody });
 }

src/app/api/v1/_shared/audioProviderNodes.ts

@@ -24,7 +24,6 @@ import { getCachedProviderNodes } from "@/lib/db/readCache";
 import { isFeatureFlagEnabled } from "@/shared/utils/featureFlags";
 import {
   buildDynamicAudioProvider,
-  isLoopbackNodeHost,
   type AudioProvider,
   type ProviderNodeRow,
 } from "@omniroute/open-sse/config/audioRegistry.ts";
@@ -36,7 +35,19 @@ export const AUDIO_REMOTE_NODES_FLAG = "AUDIO_REMOTE_PROVIDER_NODES";
  * Loopback / private-range hosts that never leave the operator's machine or
  * Docker network. `::1` stays excluded, matching the previous SSRF hardening.
  */
-export { isLoopbackNodeHost as isLocalAudioNodeHost };
+export function isLocalAudioNodeHost(baseUrl: string): boolean {
+  try {
+    const hostname = new URL(baseUrl).hostname;
+    return (
+      hostname === "localhost" ||
+      hostname === "127.0.0.1" ||
+      // Strictly 172.16.0.0/12 (Docker/local)
+      /^172\.(1[6-9]|2[0-9]|3[0-1])\.\d{1,3}\.\d{1,3}$/.test(hostname)
+    );
+  } catch {
+    return false;
+  }
+}
 
 /**
  * Pure selection step — no DB, no flag lookup, so the policy is directly testable.
@@ -61,7 +72,7 @@ export function selectAudioProviderNodes(
       return false;
     }
     if (!node.baseUrl) return false;
-    return isLoopbackNodeHost(node.baseUrl) || allowRemote;
+    return isLocalAudioNodeHost(node.baseUrl) || allowRemote;
   });
 
   const providers: AudioProvider[] = [];

src/app/api/v1/models/catalog.ts

@@ -1376,7 +1376,7 @@ async function buildUnifiedModelsResponseCore(
           continue;
         }
 
-        // #8958/#9034: honor the compatible-provider node prefix (as the synced/custom
+        // #8958: honor the compatible-provider node prefix (as the synced/custom
         // loops do) so an alias-backed entry publishes `prefix/model` instead of the
         // raw provider-node UUID. Without the providerIdToPrefix lookup, `alias` fell
         // through to `providerKey` (the UUID) and the dedupe below — which only checks

src/app/api/v1/vscode/[token]/usableChatModel.ts

@@ -13,9 +13,6 @@
 // happen once.
 
 export type UsableChatModelCandidate = {
-	id?: string;
-	root?: string;
-	name?: string;
 	owned_by?: string;
 	parent?: string | null;
 	type?: string;
@@ -47,19 +44,9 @@ function excludesTextOutputModality(model: UsableChatModelCandidate) {
 	);
 }
 
-function isBuiltinAutoModel(model: UsableChatModelCandidate): boolean {
-	const id = model.id || model.root || model.name || "";
-	const normalized = id.trim().toLowerCase();
-	return normalized === "auto" || normalized.startsWith("auto/");
-}
-
 export function isUsableChatModel(model: UsableChatModelCandidate) {
 	if (typeof model.owned_by === "string" && model.owned_by.trim().toLowerCase() === "combo") {
-		// Allow built-in auto-routing models (e.g. auto, auto/best-coding)
-		// while still excluding operator-created combos.
-		if (!isBuiltinAutoModel(model)) {
-			return false;
-		}
+		return false;
 	}
 	if (typeof model.parent === "string" && model.parent.length > 0) return false;
 	if (typeof model.type === "string" && model.type !== "chat") return false;

src/lib/db/radar.ts

@@ -103,9 +103,7 @@ export function setRadarCache(entry: {
 export function getRadarSettings(): RadarSettings {
   const db = getDbInstance();
   const row = db
-    .prepare(
-      "SELECT opt_in, supporter_key_encrypted, updated_at FROM radar_settings WHERE id = 1"
-    )
+    .prepare("SELECT opt_in, supporter_key_encrypted, updated_at FROM radar_settings WHERE id = 1")
     .get() as { opt_in: number; supporter_key_encrypted: string | null; updated_at: string };
 
   return {
@@ -120,9 +118,9 @@ export function getRadarSettings(): RadarSettings {
  */
 export function setRadarOptIn(optIn: boolean): void {
   const db = getDbInstance();
-  db.prepare(
-    "UPDATE radar_settings SET opt_in = ?, updated_at = datetime('now') WHERE id = 1"
-  ).run(optIn ? 1 : 0);
+  db.prepare("UPDATE radar_settings SET opt_in = ?, updated_at = datetime('now') WHERE id = 1").run(
+    optIn ? 1 : 0
+  );
 }
 
 /**
@@ -133,9 +131,20 @@ export function setRadarOptIn(optIn: boolean): void {
 export function setRadarKey(key: string | null): void {
   const db = getDbInstance();
   const encrypted = key !== null ? encrypt(key) : null;
-  db.prepare(
+  const updateKey = db.prepare(
     "UPDATE radar_settings SET supporter_key_encrypted = ?, updated_at = datetime('now') WHERE id = 1"
-  ).run(encrypted);
+  );
+  const clearCatalogCache = db.prepare("DELETE FROM radar_feed_cache WHERE id = 1");
+  const clearReferralsCache = db.prepare("DELETE FROM radar_referrals_cache WHERE id = 1");
+
+  db.transaction(() => {
+    updateKey.run(encrypted);
+    // Both signed feeds are entitlement-sensitive. Clearing their cached
+    // variants forces the next sync/read to resolve the new key server-side
+    // instead of serving data fetched under the previous entitlement.
+    clearCatalogCache.run();
+    clearReferralsCache.run();
+  })();
 }
 
 // ---------------------------------------------------------------------------

src/lib/radar/applyFeed.ts

@@ -5,12 +5,15 @@
  * The baseline (`FREE_MODEL_BUDGETS`) is NEVER mutated.
  *
  * Merge rules:
- *  1. Feed never overwrites a local override.
+ *  1. Feed never overwrites a local override, except the safety-critical
+ *     `enabled:false` signal for an upstream model confirmed unavailable.
  *  2. `enabled:false` disables the entry with `disabledBy: "radar"` provenance.
  *  3. User-added entry NOT in the feed survives untouched.
  *  4. User deletion tombstone prevents feed from resurrecting the entry.
  */
 
+import type { RadarLocalizedText } from "./feedSchema";
+
 // ---------------------------------------------------------------------------
 // Types
 // ---------------------------------------------------------------------------
@@ -73,7 +76,7 @@ export interface MergedEntry {
   /** Setup guide (key URL + steps) reported by the feed. Undefined for baseline-only entries. */
   setup?: {
     keyUrl: string | null;
-    steps: string[];
+    steps: RadarLocalizedText[];
   } | null;
 }
 
@@ -107,7 +110,7 @@ export interface FeedModel {
   tosRisk: MergedEntry["tos"];
   setup: {
     keyUrl: string | null;
-    steps: string[];
+    steps: RadarLocalizedText[];
   } | null;
   enabled: boolean;
 }
@@ -127,9 +130,7 @@ function entryKey(provider: string, modelId: string): string {
  * Convert a FeedModel budget into a `monthlyTokens` number compatible
  * with the baseline catalog shape.
  */
-function feedBudgetToMonthlyTokens(
-  budget: FeedModel["budget"],
-): number {
+function feedBudgetToMonthlyTokens(budget: FeedModel["budget"]): number {
   if (budget.kind === "per_model") return budget.tokensPerMonth;
   if (budget.kind === "shared_pool") return budget.tokensPerMonth;
   return 0; // rate_only
@@ -138,9 +139,7 @@ function feedBudgetToMonthlyTokens(
 /**
  * Convert a FeedModel budget into a `poolKey` compatible with the baseline.
  */
-function feedBudgetToPoolKey(
-  budget: FeedModel["budget"],
-): string | null {
+function feedBudgetToPoolKey(budget: FeedModel["budget"]): string | null {
   if (budget.kind === "shared_pool") return budget.poolId;
   return null;
 }
@@ -242,7 +241,7 @@ export function applyFeed(input: ApplyFeedInput): MergedEntry[] {
 function mergeOne(
   base: MergedEntry,
   feed: FeedModel,
-  overrides: Partial<MergedEntry> | undefined,
+  overrides: Partial<MergedEntry> | undefined
 ): MergedEntry {
   // Start from baseline
   const result: MergedEntry = { ...base };
@@ -307,6 +306,13 @@ function mergeOne(
     if (overrides.setup !== undefined) result.setup = overrides.setup;
   }
 
+  // Safety exception to rule 1: a model confirmed unavailable upstream is
+  // never resurrected by a stale local enabled:true override.
+  if (!feed.enabled) {
+    result.enabled = false;
+    result.disabledBy = "radar";
+  }
+
   // Origin: "local" if user has overrides, else "radar" (feed updated it)
   result.origin = overriddenKeys.size > 0 ? "local" : "radar";
 
@@ -318,7 +324,7 @@ function mergeOne(
  */
 function feedModelToMerged(
   feed: FeedModel,
-  overrides: Partial<MergedEntry> | undefined,
+  overrides: Partial<MergedEntry> | undefined
 ): MergedEntry {
   const entry: MergedEntry = {
     provider: feed.provider,
@@ -329,8 +335,8 @@ function feedModelToMerged(
     freeType: overrides?.freeType ?? feed.freeType,
     poolKey: overrides?.poolKey ?? feedBudgetToPoolKey(feed.budget),
     tos: overrides?.tos ?? feed.tosRisk,
-    trainsOnPrompts: overrides?.trainsOnPrompts ?? (feed.trainsOnPrompts ?? undefined),
-    enabled: overrides?.enabled ?? feed.enabled,
+    trainsOnPrompts: overrides?.trainsOnPrompts ?? feed.trainsOnPrompts ?? undefined,
+    enabled: feed.enabled ? (overrides?.enabled ?? true) : false,
     origin: overrides ? "local" : "radar",
     contextWindow: overrides?.contextWindow ?? feed.contextWindow,
     capabilities: overrides?.capabilities ?? feed.capabilities,
@@ -338,10 +344,8 @@ function feedModelToMerged(
     setup: overrides?.setup ?? feed.setup,
   };
 
-  // Rule 2 (feed disable) — but rule 1 (local override wins) takes precedence,
-  // matching mergeOne(): only force-disable when the user has NOT explicitly
-  // overridden `enabled` locally.
-  if (!feed.enabled && overrides?.enabled === undefined) {
+  // Rule 2 is the safety exception to local override precedence.
+  if (!feed.enabled) {
     entry.enabled = false;
     entry.disabledBy = "radar";
   }

src/lib/radar/feedSchema.ts

@@ -43,6 +43,16 @@ export type RadarTier = z.infer<typeof RadarTierSchema>;
 
 const IntNullable = z.number().int().nullable();
 
+/** D25: English is canonical; Portuguese is an optional localized companion. */
+export const RadarLocalizedTextSchema = z.union([
+  z.string(), // compatibility with schema-v1 feeds published before D25
+  z.object({
+    en: z.string().min(1),
+    pt: z.string().min(1).optional(),
+  }),
+]);
+export type RadarLocalizedText = z.infer<typeof RadarLocalizedTextSchema>;
+
 /**
  * Budget is a discriminated union on `kind`:
  *   - per_model: tokensPerMonth (positive int)
@@ -80,7 +90,7 @@ const CapabilitiesSchema = z.object({
 const SetupSchema = z
   .object({
     keyUrl: z.string().url().nullable(),
-    steps: z.array(z.string()),
+    steps: z.array(RadarLocalizedTextSchema),
   })
   .nullable();
 
@@ -167,8 +177,8 @@ const QuirkTargetSchema = z.object({
 
 const QuirkSchema = z.object({
   slug: z.string(),
-  title: z.string(),
-  body: z.string(),
+  title: RadarLocalizedTextSchema,
+  body: RadarLocalizedTextSchema,
   severity: SeverityEnum,
   targets: z.array(QuirkTargetSchema),
 });
@@ -182,10 +192,9 @@ export const RadarFeedSchema = z.object({
   schemaVersion: z.literal(1),
   version: z.string(),
   generatedAt: z.string().datetime(),
-  // NOTE: this body field is ALWAYS "live", by design — the community tier
-  // is the exact same signed bytes served from an older snapshot, and there
-  // is only one signed artifact per version (rewriting this field
-  // server-side per request would break the exact-bytes Ed25519 signature).
+  // NOTE: this body field is not the entitlement decision. The server can
+  // publish separate exact-byte live/community artifacts for one version,
+  // while the selected request tier is still communicated by the header.
   // The tier ACTUALLY served is decided by the server per-request based on
   // the Authorization key, and is surfaced via the `x-omniroute-feed-tier`
   // response header instead. NEVER read this field for UI/display — use the

src/lib/radar/referralsSync.ts

@@ -18,10 +18,7 @@
  * Deps are injectable for testing.
  */
 
-import {
-  RadarReferralsFeedSchema,
-  type RadarReferralsFeed,
-} from "./referralsFeedSchema";
+import { RadarReferralsFeedSchema, type RadarReferralsFeed } from "./referralsFeedSchema";
 import { RadarTierSchema, type RadarTier } from "./feedSchema";
 import { verifyFeedBytes } from "./verify";
 import { sanitizeErrorMessage } from "@omniroute/open-sse/utils/error";
@@ -144,9 +141,10 @@ export function shouldSyncReferralsOnRead(
  *  4. Verify Ed25519 signature over exact bytes (same pinned key as the
  *     catalog feed — one key pins both artifacts).
  *  5. Parse+validate with RadarReferralsFeedSchema.
- *  6. Replay guard: incoming `generatedAt` must be strictly newer than the
- *     cache (a same/older `generatedAt` is a no-op — nothing changed, or a
- *     stale replay — either way the cache is left untouched).
+ *  6. Replay guard: reject an incoming `generatedAt` older than the cache.
+ *     Equal timestamps remain valid because the community and live referral
+ *     variants deliberately share one deterministic `generatedAt`; the
+ *     served tier can still change after the supporter key changes.
  *  7. Cache the result.
  *
  * @param deps - Injectable dependencies for testing.
@@ -262,7 +260,8 @@ export async function syncRadarReferrals(
       return { status: "invalid_schema" };
     }
 
-    // Step 7: generatedAt floor — replay/no-op guard.
+    // Step 7: generatedAt floor — reject older signed replays. Equal
+    // timestamps are accepted because entitlement variants share generatedAt.
     let existingCache: RadarReferralsCacheEntry | null = null;
     if (getCacheFn) {
       existingCache = getCacheFn();
@@ -274,7 +273,7 @@ export async function syncRadarReferrals(
     if (existingCache) {
       const existingMs = Date.parse(existingCache.generatedAt);
       const incomingMs = Date.parse(feed.generatedAt);
-      if (Number.isFinite(existingMs) && Number.isFinite(incomingMs) && incomingMs <= existingMs) {
+      if (Number.isFinite(existingMs) && Number.isFinite(incomingMs) && incomingMs < existingMs) {
         return { status: "stale" };
       }
     }
@@ -282,7 +281,8 @@ export async function syncRadarReferrals(
     // Step 8: Resolve served tier — the body carries no `tier` field at all
     // for this feed, so the header is the only source; absent/garbage header
     // degrades to the least-privileged "community" default.
-    const servedTier = parseServedTierHeader(res.headers.get("x-omniroute-feed-tier")) ?? "community";
+    const servedTier =
+      parseServedTierHeader(res.headers.get("x-omniroute-feed-tier")) ?? "community";
 
     // Step 9: Cache the result
     const cacheEntry: RadarReferralsCacheEntry = {

src/lib/radar/setupConnections.ts

@@ -0,0 +1,26 @@
+export interface RadarSetupConnection {
+  id: string;
+  provider: string;
+  isActive?: boolean;
+}
+
+/** Reuse the authenticated provider-connections API with a bounded provider filter. */
+export function providerConnectionsRequestUrl(provider: string): string {
+  return `/api/providers?provider=${encodeURIComponent(provider)}`;
+}
+
+/**
+ * Pick a concrete connection id for the setup test endpoint. Prefer an active
+ * connection, then fall back to the first valid connection for the provider.
+ */
+export function firstProviderConnectionId(
+  connections: readonly RadarSetupConnection[],
+  provider: string
+): string | null {
+  const matching = connections.filter(
+    (connection) => connection.provider === provider && connection.id.length > 0
+  );
+  return (
+    matching.find((connection) => connection.isActive !== false)?.id ?? matching[0]?.id ?? null
+  );
+}

src/lib/radar/sync.ts

@@ -266,7 +266,14 @@ export async function syncRadar(deps: SyncDeps = {}): Promise<SyncStatus> {
       return { status: "invalid_schema" };
     }
 
-    // Step 7: Version floor
+    // Step 7: Resolve the served tier before the version floor. A single-use
+    // supporter key deliberately transitions from live to community after its
+    // first catalog pull, and the community snapshot can be older.
+    const servedTier = parseServedTierHeader(res.headers.get("x-omniroute-feed-tier")) ?? feed.tier;
+
+    // Step 8: Version floor. Same/older versions are rejected within a tier,
+    // but a verified live -> community transition must replace the privileged
+    // cache even when the community snapshot is older.
     let existingCache: RadarCacheEntry | null = null;
     if (getCacheFn) {
       existingCache = getCacheFn();
@@ -275,17 +282,15 @@ export async function syncRadar(deps: SyncDeps = {}): Promise<SyncStatus> {
       existingCache = mod.getRadarCache();
     }
 
-    if (existingCache && compareVersions(feed.version, existingCache.version) <= 0) {
+    const isEntitlementDowngrade = existingCache?.tier === "live" && servedTier === "community";
+    if (
+      existingCache &&
+      !isEntitlementDowngrade &&
+      compareVersions(feed.version, existingCache.version) <= 0
+    ) {
       return { status: "stale" };
     }
 
-    // Step 8: Resolve the served tier.
-    // The `x-omniroute-feed-tier` header reflects the tier ACTUALLY served
-    // (see `parseServedTierHeader`); fall back to the signed body's `tier`
-    // field only when the header is absent or unrecognized — never trust an
-    // arbitrary header value into the cache/UI.
-    const servedTier = parseServedTierHeader(res.headers.get("x-omniroute-feed-tier")) ?? feed.tier;
-
     // Step 9: Cache the result
     const cacheEntry: RadarCacheEntry = {
       version: feed.version,

src/server/authz/policies/management.ts

@@ -253,39 +253,6 @@ export const managementPolicy: RoutePolicy = {
       return allow({ kind: "management_key", id: "cli", label: "local-cli-token" });
     }
 
-    // MCP path carve-out (#9159): accept mcp:connect, manage, or admin
-    // scope for /api/mcp/* from any origin (loopback, private LAN, or remote).
-    // Loopback/LAN requests skip the Tier 1 bypass gate above, so with
-    // requireLogin=true they would fall through to the generic API-key check
-    // which only accepts manage/admin -- rejecting mcp:connect-only keys.
-    // This carve-out mirrors the existing Tier 1 MCP check but without the
-    // locality guard, so it catches the loopback/LAN requests that the Tier 1
-    // gate does not reach.
-    if (path.startsWith("/api/mcp/")) {
-      const apiKey = extractApiKey(ctx.request as unknown as Request, { allowUrl: false });
-      if (apiKey) {
-        try {
-          if (await isValidApiKey(apiKey)) {
-            const meta = await getApiKeyMetadata(apiKey);
-            if (meta && hasMcpConnectOrManageScope(meta.scopes)) {
-              const grantedBy = meta.scopes.includes("admin")
-                ? "admin"
-                : meta.scopes.includes("manage")
-                  ? "manage"
-                  : "mcp-connect";
-              return allow({
-                kind: "management_key",
-                id: meta.id,
-                label: `api-key-${grantedBy}-scope-mcp-carve-out`,
-              });
-            }
-          }
-        } catch {
-          return reject(503, "AUTH_BACKEND_UNAVAILABLE", "Service temporarily unavailable");
-        }
-      }
-    }
-
     // Tier 2: always-protected routes skip the requireLogin=false bypass.
     if (!isAlwaysProtectedPath(path) && !(await isAuthRequired(ctx.request))) {
       return allow({ kind: "anonymous", id: "anonymous", label: "auth-disabled" });

src/shared/utils/classify429.ts

@@ -67,14 +67,6 @@ const QUOTA_PATTERNS: ReadonlyArray<RegExp> = [
   // ~60s against a budget that only resets at UTC midnight.
   /daily free allocation/i,
 
-  // OmniRoute auth-layer synthetic 429 (Issue #9269).
-  // Body: "All antigravity accounts have exhausted their quota (reset after 5m)"
-  // Produced by auth.ts line 1477 when every account for a provider has
-  // exhausted its quota. Without this pattern, the message is classified as
-  // a transient rate-limit and the combo loop burns retries against the
-  // same provider instead of falling back to a healthy one.
-  /have exhausted their quota/i,
-
   // Modal-hosted OpenAI-compatible endpoints (e.g. self-hosted Kimi K3).
   // Body: {"error":"usage limit reached"}, no nested "message"/"quota"/
   // "daily" wording. Without this pattern the 429 falls through to

src/sse/handlers/chat.ts

@@ -714,20 +714,6 @@ async function handleChatImplementation(
     ) => {
       if (isComboLiveTest) return true;
 
-      // #9057: for keys with model restrictions (allowedModels or disableNonPublicModels),
-      // run isModelAllowedForKey even for auto/* models. The API-key policy gate
-      // (validateModelAccess in apiKeyPolicy.ts) treats auto/* as a virtual combo and
-      // skips isModelAllowedForKey, so the per-candidate check here is the only
-      // enforcement point during combo routing. Without it, a key with
-      // disableNonPublicModels=true can reach free/prohibited models through auto/*.
-      const hasModelRestrictions =
-        apiKeyInfo &&
-        (Boolean(apiKeyInfo.allowedModels?.length) || apiKeyInfo.disableNonPublicModels === true);
-      if (hasModelRestrictions && apiKey) {
-        const modelAllowed = await isModelAllowedForKey(apiKey, modelString);
-        if (!modelAllowed) return false;
-      }
-
       // Use getModelInfo to resolve custom prefixes, but prefer the combo
       // target's providerId when available — the model string's provider
       // prefix may differ from the credential provider ID (e.g. model

src/sse/services/auth.ts

@@ -775,15 +775,9 @@ function providerCanUseSyntheticNoAuthFallback(providerId: string): boolean {
 
 async function maybeSyntheticNoAuthFallback(
   providerId: string,
-  excludedConnectionIds: Set<string>,
-  allowedConnections: string[] | null = null
+  excludedConnectionIds: Set<string>
 ) {
   if (!providerCanUseSyntheticNoAuthFallback(providerId)) return null;
-  // #9057: a key pinned to specific connections via allowedConnections must
-  // NOT receive the synthetic "noauth" connection — the synthetic id is
-  // never in an explicit allowlist, so returning it would let a restricted
-  // key reach free providers (felo-chat, etc.) that it should not access.
-  if (Array.isArray(allowedConnections) && allowedConnections.length > 0) return null;
   if (excludedConnectionIds.has(SYNTHETIC_NOAUTH_CONNECTION_ID)) return null;
   // #4954: hydrate per-account proxy/rotation config off the connection row so
   // no-auth executors (opencode, mimocode) actually honor configured proxies.
@@ -1012,14 +1006,7 @@ export async function getProviderCredentials(
         excludeConnectionId,
         options.excludeConnectionIds
       );
-      // #9057: when allowedConnections is set, the synthetic "noauth" connection
-      // is never in the explicit allowlist, so we must NOT return it — fall through
-      // to the normal connection-selection path so the connection allowlist is
-      // respected (the no-auth provider will be rejected if it has no real connections
-      // matching the allowlist, or a real connection row will be selected if present).
-      if (!allowedConnections || allowedConnections.length === 0) {
-        return await maybeSyntheticNoAuthFallback(resolvedId, excludedForNoAuth);
-      }
+      return await maybeSyntheticNoAuthFallback(resolvedId, excludedForNoAuth);
     }
 
     const allowSuppressedConnections = options.allowSuppressedConnections === true;
@@ -1146,8 +1133,7 @@ export async function getProviderCredentials(
         if (terminalConnections.length === allConnections.length) {
           const syntheticFallback = await maybeSyntheticNoAuthFallback(
             resolvedId,
-            excludedConnectionIds,
-            allowedConnections
+            excludedConnectionIds
           );
           if (syntheticFallback) return syntheticFallback;
 
@@ -1167,8 +1153,7 @@ export async function getProviderCredentials(
       }
       const syntheticFallback = await maybeSyntheticNoAuthFallback(
         resolvedId,
-        excludedConnectionIds,
-        allowedConnections
+        excludedConnectionIds
       );
       if (syntheticFallback) return syntheticFallback;
       log.warn("AUTH", `No credentials for ${provider}`);
@@ -1367,8 +1352,7 @@ export async function getProviderCredentials(
       }
       const syntheticFallback = await maybeSyntheticNoAuthFallback(
         resolvedId,
-        excludedConnectionIds,
-        allowedConnections
+        excludedConnectionIds
       );
       if (syntheticFallback) return syntheticFallback;
       log.warn("AUTH", `${provider} | all ${connections.length} accounts unavailable`);

tests/fixtures/radar-feed-canonical.json

@@ -46,7 +46,10 @@
       "setup": {
         "keyUrl": "https://console.groq.com/keys",
         "steps": [
-          "Crie uma conta gratuita no console da Groq",
+          {
+            "en": "Create a free account in the Groq console",
+            "pt": "Crie uma conta gratuita no console da Groq"
+          },
           "Gere uma API key na página de keys",
           "Adicione a key no OmniRoute com o provider groq"
         ]

tests/integration/combo-matrix/context-relay-codex.test.ts

@@ -62,12 +62,6 @@ const { getHandoff } = await import("../../../src/lib/db/contextHandoffs.ts");
 // ── Constants ─────────────────────────────────────────────────────────────────
 
 const CODEX_COMBO_NAME = "m-relay-codex-quota";
-// The control test needs its OWN combo name. resetStorage() unlinks the DB file
-// between tests, but the previous better-sqlite3 handle survives the unlink and
-// keeps writing to the same inode, so reusing the name here failed with
-// `UNIQUE constraint failed: combos.name` — a test-isolation defect, not a
-// routing one (the assertion below does not depend on the name).
-const CONTROL_COMBO_NAME = "m-relay-openai-control";
 const SESSION_HEADER_VALUE = "relay-codex-quota-001";
 const SESSION_ID = `ext:${SESSION_HEADER_VALUE}`;
 
@@ -77,7 +71,8 @@ const CODEX_RESPONSES_HOST = "chatgpt.com/backend-api/codex/responses";
 
 // Summary JSON that parseHandoffJSON will successfully parse.
 const CODEX_SUMMARY_JSON = JSON.stringify({
-  summary: "User is implementing a TypeScript context-relay codex quota-handoff test using TDD.",
+  summary:
+    "User is implementing a TypeScript context-relay codex quota-handoff test using TDD.",
   keyDecisions: ["codex provider selected", "quota threshold at 90%"],
   taskProgress: "writing deterministic integration test for codex handoff",
   activeEntities: ["combo.ts", "codexQuotaFetcher.ts", "contextHandoff.ts"],
@@ -147,11 +142,11 @@ function buildCodexUsageBody(
 
 // ── Request builder ───────────────────────────────────────────────────────────
 
-function codexRequest(withSessionId = true, comboName = CODEX_COMBO_NAME) {
+function codexRequest(withSessionId = true) {
   return buildRequest({
     headers: withSessionId ? { "x-session-id": SESSION_HEADER_VALUE } : {},
     body: {
-      model: comboName,
+      model: CODEX_COMBO_NAME,
       stream: false,
       messages: [{ role: "user", content: "Write a TypeScript hello world." }],
     },
@@ -264,7 +259,9 @@ test("context-relay codex quota handoff: fires and expiresAt matches session-win
     name: CODEX_COMBO_NAME,
     strategy: "context-relay",
     config: { maxRetries: 0, retryDelayMs: 0, stickyRoundRobinLimit: 1 },
-    models: [{ id: "rc-codex-1", kind: "model", providerId: "codex", model: "gpt-5.3-codex" }],
+    models: [
+      { id: "rc-codex-1", kind: "model", providerId: "codex", model: "gpt-5.3-codex" },
+    ],
   });
 
   // 4. Compute quota reset times (future timestamps).
@@ -331,10 +328,12 @@ test("context-relay codex quota handoff: does NOT fire when provider is openai (
   await seedConnection("openai", { apiKey: "sk-openai-control-no-codex-block" });
 
   await combosDb.createCombo({
-    name: CONTROL_COMBO_NAME,
+    name: CODEX_COMBO_NAME,
     strategy: "context-relay",
     config: { maxRetries: 0, retryDelayMs: 0, stickyRoundRobinLimit: 1 },
-    models: [{ id: "rc-openai-ctrl", kind: "model", providerId: "openai", model: "gpt-4o-mini" }],
+    models: [
+      { id: "rc-openai-ctrl", kind: "model", providerId: "openai", model: "gpt-4o-mini" },
+    ],
   });
 
   const seenUrls: string[] = [];
@@ -343,7 +342,7 @@ test("context-relay codex quota handoff: does NOT fire when provider is openai (
     return buildOpenAIResponse("assistant reply ok");
   };
 
-  const r = await handleChat(codexRequest(true, CONTROL_COMBO_NAME));
+  const r = await handleChat(codexRequest(true));
   assert.equal(r.status, 200, "openai request must return 200");
 
   // Give setImmediate time to fire if the block were incorrectly entered.
@@ -359,7 +358,7 @@ test("context-relay codex quota handoff: does NOT fire when provider is openai (
   // No codex quota handoff record in DB.
   // (The universal handoff also does not fire because no prior model is seeded,
   // so getLastSessionModel returns null → no model switch detected.)
-  const handoff = getHandoff(SESSION_ID, CONTROL_COMBO_NAME);
+  const handoff = getHandoff(SESSION_ID, CODEX_COMBO_NAME);
   assert.equal(
     handoff,
     null,

tests/unit/8951-github-gpt56-responses.test.ts

@@ -1,19 +0,0 @@
-import assert from "node:assert/strict";
-import test from "node:test";
-
-import { GithubExecutor } from "../../open-sse/executors/github.ts";
-
-test("#8951 GitHub GPT-5.6 models must use the Responses endpoint", () => {
-  const executor = new GithubExecutor();
-  const urls = Object.fromEntries(
-    ["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"].map((model) => [
-      model,
-      executor.buildUrl(model, false),
-    ])
-  );
-  assert.deepEqual(urls, {
-    "gpt-5.6-sol": "https://api.githubcopilot.com/responses",
-    "gpt-5.6-terra": "https://api.githubcopilot.com/responses",
-    "gpt-5.6-luna": "https://api.githubcopilot.com/responses",
-  });
-});

tests/unit/9034-alias-backed-prefix-id-repro.test.ts

@@ -1,120 +0,0 @@
-/**
- * Regression test for #9034 — /v1/models alias-backed emission block (catalog.ts ~:1298)
- * leaked the raw provider-node UUID as the published model `id` for alias-backed models
- * (synced via `syncManagedAvailableModelAliases`), instead of the operator-configured
- * prefix. The #8327 fix only covered `owned_by`; the routable model `id` was missed.
- *
- * Root cause: the alias-backed block builds `alias` as
- * `providerIdToAlias[canonicalProviderId] || providerKey` and never consults
- * `providerIdToPrefix`. For a compatible provider node, the storage prefix is the raw
- * node UUID (managedAvailableModels.getProviderStoragePrefix()), so `providerKey` is the
- * node UUID and the catalog re-publishes it as the public model `id` (e.g.
- * `openai-compatible-chat-550e8400-.../kimi-k2`) instead of the configured prefix.
- *
- * Fix: resolve `const prefix = providerIdToPrefix[providerKey] ?? providerIdToPrefix[canonicalProviderId]`
- * and `const alias = prefix || providerIdToAlias[canonicalProviderId] || providerKey`,
- * plus add `!prefix` to the includeCanonical guard (mirroring synced :896 / custom :1245).
- */
-import test from "node:test";
-import assert from "node:assert/strict";
-import fs from "node:fs";
-import os from "node:os";
-import path from "node:path";
-
-const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-9034-"));
-process.env.DATA_DIR = TEST_DATA_DIR;
-
-// Type-only imports for the module shape
-type CoreModule = typeof import("../../src/lib/db/core.ts");
-type ProvidersDbModule = typeof import("../../src/lib/db/providers.ts");
-type ModelsDbModule = typeof import("../../src/lib/db/models.ts");
-type CatalogModule = typeof import("../../src/app/api/v1/models/catalog.ts");
-type ManagedAvailableModelsModule = typeof import("../../src/lib/providerModels/managedAvailableModels.ts");
-
-let core: CoreModule;
-let providersDb: ProvidersDbModule;
-let modelsDb: ModelsDbModule;
-let v1ModelsCatalog: CatalogModule;
-let managedAvailableModels: ManagedAvailableModelsModule;
-
-// A realistic provider-node id shape, matching `openai-compatible-chat-<uuid>`
-const NODE_ID = "openai-compatible-chat-550e8400-e29b-41d4-a716-446655440000";
-const UUID_SHAPE_RE = /[0-9a-f]{8}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{4}-[0-9a-f]{12}/i;
-const CONFIGURED_PREFIX = "myprefix";
-const MODEL_NAME = "kimi-k2";
-
-async function resetStorage() {
-  core.resetDbInstance();
-  fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true });
-  fs.mkdirSync(TEST_DATA_DIR, { recursive: true });
-  v1ModelsCatalog.__resetCatalogBuilderRunsForTest();
-}
-
-test.before(async () => {
-  core = await import("../../src/lib/db/core.ts");
-  providersDb = await import("../../src/lib/db/providers.ts");
-  modelsDb = await import("../../src/lib/db/models.ts");
-  v1ModelsCatalog = await import("../../src/app/api/v1/models/catalog.ts");
-  managedAvailableModels = await import("../../src/lib/providerModels/managedAvailableModels.ts");
-  await resetStorage();
-});
-
-test.after(async () => {
-  if (core) core.resetDbInstance();
-  fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true });
-});
-
-test("#9034: alias-backed model id must use the configured prefix, not the raw provider-node UUID", async () => {
-  // Create an openai-compatible provider node with a configured prefix
-  await providersDb.createProviderNode({
-    id: NODE_ID,
-    type: "openai-compatible",
-    name: "test node (probe)",
-    prefix: CONFIGURED_PREFIX,
-    baseUrl: "https://proxy.example.com",
-    chatPath: "/v1/chat/completions",
-    modelsPath: "/v1/models",
-  });
-  await providersDb.createProviderConnection({
-    provider: NODE_ID,
-    authType: "apikey",
-    name: "test-conn",
-    apiKey: "sk-test",
-    isActive: true,
-    testStatus: "active",
-    providerSpecificData: {
-      baseUrl: "https://proxy.example.com",
-      chatPath: "/v1/chat/completions",
-      modelsPath: "/v1/models",
-    },
-  });
-
-  // The real producer path: syncManagedAvailableModelAliases stores aliases as
-  // `<UUID>/<modelId>` (getProviderStoragePrefix returns raw node id for compatible providers).
-  // This creates a key_value alias entry that the alias-backed block in catalog.ts reads.
-  await managedAvailableModels.syncManagedAvailableModelAliases(NODE_ID, [MODEL_NAME]);
-
-  const response = await v1ModelsCatalog.getUnifiedModelsResponse(
-    new Request("http://localhost/api/v1/models")
-  );
-  const body = (await response.json()) as { data: Array<Record<string, unknown>> };
-  const ids = body.data.map((m) => m.id as string);
-
-  assert.equal(response.status, 200);
-
-  // (a) The configured prefix must be used in the model id
-  const expectedId = `${CONFIGURED_PREFIX}/${MODEL_NAME}`;
-  assert.ok(
-    ids.includes(expectedId),
-    `expected model id "${expectedId}" to exist in /v1/models — got: ${JSON.stringify(ids)}`
-  );
-
-  // (b) No entry id should start with the raw node UUID when a prefix is configured
-  for (const id of ids) {
-    assert.equal(
-      id.startsWith(NODE_ID),
-      false,
-      `entry id "${id}" must not start with the raw provider-node UUID "${NODE_ID}" when a prefix ("${CONFIGURED_PREFIX}") is configured`
-    );
-  }
-});

tests/unit/9134-repro-audio-combo-rejection.test.ts

@@ -1,95 +0,0 @@
-// Repro test for #9134 — /v1/audio/transcriptions rejects combo names.
-//
-// Run: node --import tsx/esm --test tests/unit/9134-repro-audio-combo-rejection.test.ts
-// Expected to PASS once the fix is applied, RED before.
-
-import test from "node:test";
-import assert from "node:assert/strict";
-import fs from "node:fs";
-import os from "node:os";
-import path from "node:path";
-
-const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-9134-repro-"));
-process.env.DATA_DIR = TEST_DATA_DIR;
-
-const core = await import("../../src/lib/db/core.ts");
-const { createCombo } = await import("../../src/lib/db/combos.ts");
-const { createProviderNode } = await import("../../src/lib/db/providers.ts");
-const route = await import("../../src/app/api/v1/audio/transcriptions/route.ts");
-
-const originalFetch = globalThis.fetch;
-
-test.after(() => {
-  globalThis.fetch = originalFetch;
-  core.resetDbInstance();
-  fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true });
-});
-
-/** Minimal but structurally valid WAV so nothing rejects the upload shape. */
-function makeWav(): Blob {
-  const dataLen = 1600;
-  const b = Buffer.alloc(44 + dataLen);
-  b.write("RIFF", 0, "ascii");
-  b.writeUInt32LE(36 + dataLen, 4);
-  b.write("WAVE", 8, "ascii");
-  b.write("fmt ", 12, "ascii");
-  b.writeUInt32LE(16, 16);
-  b.writeUInt16LE(1, 20);
-  b.writeUInt16LE(1, 22);
-  b.writeUInt32LE(16000, 24);
-  b.writeUInt32LE(32000, 28);
-  b.writeUInt16LE(2, 32);
-  b.writeUInt16LE(16, 34);
-  b.write("data", 36, "ascii");
-  b.writeUInt32LE(dataLen, 40);
-  return new Blob([b], { type: "audio/wav" });
-}
-
-function transcriptionRequest(model: string) {
-  const fd = new FormData();
-  fd.set("model", model);
-  fd.set("file", makeWav(), "t.wav");
-  return new Request("http://localhost/v1/audio/transcriptions", { method: "POST", body: fd });
-}
-
-test("#9134 combo name is rejected instead of resolved", async () => {
-  await createProviderNode({
-    id: "openai-compatible-audio-transcriptions-test",
-    type: "openai-compatible",
-    name: "Local STT",
-    prefix: "localstt",
-    apiType: "audio-transcriptions",
-    baseUrl: "http://localhost:9000/v1",
-  } as Parameters<typeof createProviderNode>[0]);
-
-  await createCombo({
-    name: "transcricao",
-    strategy: "priority",
-    models: [{ provider: "localstt", model: "whisper-1" }],
-  } as Parameters<typeof createCombo>[0]);
-
-  globalThis.fetch = (async () =>
-    new Response(JSON.stringify({ text: "ok" }), {
-      status: 200,
-      headers: { "Content-Type": "application/json" },
-    })
-  ) as typeof fetch;
-
-  const res = await route.POST(transcriptionRequest("transcricao"));
-  const body = await res.text();
-
-  // The bug: the combo name "transcricao" is NOT resolved. The route returns 400
-  // with "Invalid transcription model: transcricao. Use format: provider/model"
-  // even though /v1/models advertises this combo and chat/embeddings resolve it.
-  // Regression guard: combo names must be resolved before model parsing. This
-  // was failing as `400 Invalid transcription model: transcricao` before the fix.
-  assert.notEqual(
-    res.status,
-    400,
-    `BUG #9134: combo name "transcricao" was rejected as invalid model — got status ${res.status}: ${body}`
-  );
-  assert.ok(
-    !body.includes("Invalid transcription model"),
-    `BUG #9134: combo name was not resolved — got: ${body}`
-  );
-});

tests/unit/api-key-policy-noauth-allowed-connections.test.ts

@@ -1,71 +0,0 @@
-/**
- * #9057 — API key `allowedConnections` MUST gate no-auth synthetic credentials
- *
- * TDD regression test: an API key pinned via `allowedConnections` to a specific
- * connection must NOT receive synthetic no-auth credentials for free providers
- * (e.g. felo-chat).
- */
-
-import test from "node:test";
-import assert from "node:assert/strict";
-import fs from "node:fs";
-import os from "node:os";
-import path from "node:path";
-
-const TEST_DATA_DIR = fs.mkdtempSync(path.join(os.tmpdir(), "omniroute-9057-"));
-process.env.DATA_DIR = TEST_DATA_DIR;
-process.env.API_KEY_SECRET = process.env.API_KEY_SECRET || "9057-test-secret";
-
-const coreDb = await import("../../src/lib/db/core.ts");
-const apiKeysDb = await import("../../src/lib/db/apiKeys.ts");
-const { getProviderCredentials } = await import("../../src/sse/services/auth.ts");
-const { isModelAllowedForKey } = await import("../../src/lib/db/apiKeys.ts");
-
-const RESTRICTED_CONNECTION_UUID = "00000000-0000-4000-8000-000000000001";
-
-test.after(() => {
-  coreDb.resetDbInstance();
-  try { fs.rmSync(TEST_DATA_DIR, { recursive: true, force: true }); } catch {}
-});
-
-test("#9057 LAYER1: restricted key gets NO synthetic credentials for noauth provider felo", async () => {
-  // LAYER1: getProviderCredentials() with explicit allowedConnections
-  // must NOT return synthetic noauth credentials because the synthetic
-  // "noauth" connection is never in an explicit allowed-connections list.
-  const creds = await getProviderCredentials(
-    "felo",
-    null,
-    [RESTRICTED_CONNECTION_UUID], // allowedConnections restricts to a real UUID
-    "felo-chat"
-  );
-  assert.equal(creds, null,
-    "noauth provider felo must not leak synthetic credentials for a connection-restricted key");
-});
-
-test("#9057 LAYER1: unrestricted key still gets synthetic credentials for felo", async () => {
-  const creds = await getProviderCredentials(
-    "felo",
-    null,
-    null,   // allowedConnections=null means unrestricted
-    "felo-chat"
-  );
-  assert(creds, "unrestricted key must receive synthetic credentials for felo");
-  assert.equal(
-    (creds as Record<string, unknown>)?.connectionId,
-    "noauth",
-    "synthetic noauth connection"
-  );
-});
-
-test("#9057 LAYER2: isModelAllowedForKey rejects felo-chat for disableNonPublicModels key", async () => {
-  // Create a key with disableNonPublicModels=true
-  const created = await apiKeysDb.createApiKey("dnp-9057", "machine-dnp");
-  assert(created, "key must be created");
-  const key = created.key;
-  await apiKeysDb.updateApiKeyPermissions(created.id, {
-    disableNonPublicModels: true,
-  });
-
-  const allowed = await isModelAllowedForKey(key, "felo-chat");
-  assert.equal(allowed, false, "disableNonPublicModels key must reject felo-chat");
-});

tests/unit/audio-speech-dynamic-node-9096.test.ts

@@ -1,67 +0,0 @@
-import test from "node:test";
-import assert from "node:assert/strict";
-
-const { parseSpeechModel, parseTranscriptionModel, parseTranslationModel } = await import("../../open-sse/config/audioRegistry.ts");
-
-test("parseSpeechModel resolves dynamic provider (audio-speech apiType) by prefix", () => {
-  const dynamicProviders = [
-    {
-      id: "mytest9096",
-      baseUrl: "http://localhost:9999/v1/audio/speech",
-      authType: "none" as const,
-      authHeader: "none" as const,
-      models: [],
-    },
-  ];
-
-  const result = parseSpeechModel("mytest9096/tts-1", dynamicProviders);
-  assert.equal(result.provider, "mytest9096");
-  assert.equal(result.model, "tts-1");
-});
-
-test("parseSpeechModel returns null for unknown dynamic provider prefix", () => {
-  const dynamicProviders = [
-    {
-      id: "mytest9096",
-      baseUrl: "http://localhost:9999/v1/audio/speech",
-      authType: "none" as const,
-      authHeader: "none" as const,
-      models: [],
-    },
-  ];
-
-  const result = parseSpeechModel("nonexistent/tts-1", dynamicProviders);
-  assert.equal(result.provider, null);
-});
-
-test("parseTranscriptionModel resolves dynamic provider (audio-transcriptions apiType) by prefix", () => {
-  const dynamicProviders = [
-    {
-      id: "mytest9096",
-      baseUrl: "http://localhost:9999/v1/audio/transcriptions",
-      authType: "none" as const,
-      authHeader: "none" as const,
-      models: [],
-    },
-  ];
-
-  const result = parseTranscriptionModel("mytest9096/whisper-1", dynamicProviders);
-  assert.equal(result.provider, "mytest9096");
-  assert.equal(result.model, "whisper-1");
-});
-
-test("parseTranslationModel resolves dynamic provider (audio-transcriptions apiType) by prefix", () => {
-  const dynamicProviders = [
-    {
-      id: "mytest9096",
-      baseUrl: "http://localhost:9999/v1/audio/translations",
-      authType: "none" as const,
-      authHeader: "none" as const,
-      models: [],
-    },
-  ];
-
-  const result = parseTranslationModel("mytest9096/whisper-1", dynamicProviders);
-  assert.equal(result.provider, "mytest9096");
-  assert.equal(result.model, "whisper-1");
-});

tests/unit/classify429.test.ts

@@ -45,13 +45,6 @@ test("classify429: Antigravity 'Individual quota reached' body returns 'quota_ex
   assert.equal(classify429({ status: 429, body: { error: { message: body } } }), "quota_exhausted");
 });
 
-test("classify429: auth-layer synthetic 'have exhausted their quota' returns 'quota_exhausted' (#9269)", () => {
-  const body = "All antigravity accounts have exhausted their quota (reset after 5m)";
-  assert.equal(looksLikeQuotaExhausted(body), true);
-  assert.equal(classify429({ status: 429, body }), "quota_exhausted");
-  assert.equal(classify429({ status: 429, body: { error: { message: body } } }), "quota_exhausted");
-});
-
 test("classify429: Google RESOURCE_EXHAUSTED with a billing-period reset is quota exhausted", () => {
   const body = "Resource has been exhausted (e.g. check quota). (reset after 24h)";
   assert.equal(looksLikeQuotaExhausted(body), true);

tests/unit/colocate-optionals.test.ts

@@ -33,11 +33,6 @@ function mkPkg(
  *   @tensorflow/tfjs   → dep @tensorflow/tfjs-core → dep long
  *   js-tiktoken        → dep base64-js
  *   @huggingface/transformers present at root as a (stale) 4.2.0
- *
- * Each mock package gets a resolvable entrypoint so that isPackageIntact (which
- * checks entrypoint integrity via require.resolve) can validate the co-located
- * copy. The `main` field and corresponding index.js mirror what real npm
- * packages ship.
  */
 function buildRoot(rootDir: string): void {
   const rootNm = join(rootDir, "node_modules");
@@ -45,7 +40,6 @@ function buildRoot(rootDir: string): void {
     rootNm,
     "@atjsh/llmlingua-2",
     {
-      main: "dist/index.js",
       dependencies: { "es-toolkit": "^1.38.0" },
       peerDependencies: {
         "@huggingface/transformers": "*",
@@ -55,27 +49,12 @@ function buildRoot(rootDir: string): void {
     },
     { "dist/index.js": "export const llmlingua = true;\n" }
   );
-  mkPkg(rootNm, "es-toolkit", { main: "index.js" }, { "index.js": "export const esToolkit = true;\n" });
-  mkPkg(
-    rootNm,
-    "@tensorflow/tfjs",
-    { main: "index.js", dependencies: { "@tensorflow/tfjs-core": "4.22.0" } },
-    { "index.js": "export const tfjs = true;\n" }
-  );
-  mkPkg(
-    rootNm,
-    "@tensorflow/tfjs-core",
-    { main: "index.js", dependencies: { long: "^5.0.0" } },
-    { "index.js": "export const tfjsCore = true;\n" }
-  );
-  mkPkg(rootNm, "long", { main: "index.js" }, { "index.js": "export const long = true;\n" });
-  mkPkg(
-    rootNm,
-    "js-tiktoken",
-    { main: "index.js", dependencies: { "base64-js": "^1.5.1" } },
-    { "index.js": "export const tiktoken = true;\n" }
-  );
-  mkPkg(rootNm, "base64-js", { main: "index.js" }, { "index.js": "export const base64 = true;\n" });
+  mkPkg(rootNm, "es-toolkit", {});
+  mkPkg(rootNm, "@tensorflow/tfjs", { dependencies: { "@tensorflow/tfjs-core": "4.22.0" } });
+  mkPkg(rootNm, "@tensorflow/tfjs-core", { dependencies: { long: "^5.0.0" } });
+  mkPkg(rootNm, "long", {});
+  mkPkg(rootNm, "js-tiktoken", { dependencies: { "base64-js": "^1.5.1" } });
+  mkPkg(rootNm, "base64-js", {});
   // Root transformers is the STALE 4.x line — the bug we must not propagate into dist.
   mkPkg(rootNm, "@huggingface/transformers", { version: "4.2.0" });
 }

tests/unit/docker-llmlingua-optionals-9166.test.ts

@@ -55,7 +55,6 @@ function buildLlmlinguaRoot(
     rootNm,
     "@atjsh/llmlingua-2",
     {
-      main: "dist/index.js",
       dependencies: {
         "es-toolkit": "^1.38.0",
       },
@@ -228,103 +227,6 @@ test("#9166 standalone assembly never overwrites an already pinned transformers
   }
 });
 
-test("#9166 co-location completes a partially traced package (package.json without its main)", () => {
-  const root = mkdtempSync(
-    join(tmpdir(), "omniroute-docker-llmlingua-partial-9166-")
-  );
-
-  try {
-    buildLlmlinguaRoot(root);
-    const { distDir, standaloneDir } = createStandalone(root);
-
-    // Next's file tracing materializes @atjsh/llmlingua-2 PARTIALLY in the
-    // standalone: the package.json lands (its "main" points at dist/index.js)
-    // but the dist/ payload does not — the exact state the Docker guard hits
-    // ("Cannot find module .../dist/index.js"). A directory-level no-clobber
-    // sees the dir and skips the package forever.
-    mkPkg(join(standaloneDir, "node_modules"), "@atjsh/llmlingua-2", {
-      main: "dist/index.js",
-    });
-
-    assembleStandalone({
-      distDir,
-      outDir: standaloneDir,
-      projectRoot: root,
-      copyNatives: true,
-    });
-
-    assert.ok(
-      existsSync(
-        join(
-          standaloneDir,
-          "node_modules",
-          "@atjsh",
-          "llmlingua-2",
-          "dist",
-          "index.js"
-        )
-      ),
-      "a partially traced package must be completed, not skipped as already present"
-    );
-  } finally {
-    rmSync(root, { recursive: true, force: true });
-  }
-});
-
-test("#9166 co-location is not skipped when every closure dir exists but one is partial", () => {
-  const root = mkdtempSync(
-    join(tmpdir(), "omniroute-docker-llmlingua-partial-all-9166-")
-  );
-
-  try {
-    buildLlmlinguaRoot(root);
-    const { distDir, standaloneDir } = createStandalone(root);
-    const standaloneNm = join(standaloneDir, "node_modules");
-
-    // Every closure package already has a directory in the standalone (so a
-    // directory-level "already co-located" early-exit would fire), but the
-    // llmlingua-2 one is the partial NFT-trace shell without its main.
-    for (const packageName of [
-      "es-toolkit",
-      "@tensorflow/tfjs",
-      "@tensorflow/tfjs-core",
-      "long",
-      "js-tiktoken",
-      "base64-js",
-      "@huggingface/transformers",
-      "onnxruntime-node",
-    ]) {
-      mkPkg(standaloneNm, packageName, { main: "index.js" }, {
-        "index.js": "export {};\n",
-      });
-    }
-    mkPkg(standaloneNm, "@atjsh/llmlingua-2", { main: "dist/index.js" });
-
-    assembleStandalone({
-      distDir,
-      outDir: standaloneDir,
-      projectRoot: root,
-      copyNatives: true,
-    });
-
-    assert.ok(
-      existsSync(
-        join(
-          standaloneDir,
-          "node_modules",
-          "@atjsh",
-          "llmlingua-2",
-          "dist",
-          "index.js"
-        )
-      ),
-      "the closure-wide early-exit must not fire while any member is partial"
-    );
-  } finally {
-    rmSync(root, { recursive: true, force: true });
-  }
-});
-
 test("#9166 Docker explicitly installs and validates LLMLingua optionals", () => {
   const dockerfile = readFileSync(
     new URL("../../Dockerfile", import.meta.url),

tests/unit/lmarena-stream-readiness-repro-9306.test.ts

@@ -1,94 +0,0 @@
-/**
- * Regression test for #9306 — Arena AI streaming response must produce
- * Uint8Array chunks (not raw strings) so downstream consumers like
- * ensureStreamReadiness / TextDecoder.decode() do not throw TypeError.
- *
- * Run: node --import tsx/esm --test tests/unit/lmarena-stream-readiness-repro-9306.test.ts
- */
-
-import { describe, it } from "node:test";
-import assert from "node:assert/strict";
-import { ensureStreamReadiness } from "../../open-sse/utils/streamReadiness.ts";
-import { createOpenAIArenaStream } from "../../open-sse/executors/lmarena/response.ts";
-
-describe("Arena AI stream readiness (#9306)", () => {
-  it("produces Uint8Array chunks consumable by ensureStreamReadiness", async () => {
-    // Simulate the upstream Arena TLS reader returning Uint8Array chunks
-    const upstreamReader = new ReadableStream<Uint8Array>({
-      start(controller) {
-        controller.enqueue(new TextEncoder().encode('a0:{"text":"Hello"}\n'));
-        controller.enqueue(
-          new TextEncoder().encode('a0:{"text":", world!"}\nad:{}\n')
-        );
-        controller.close();
-      },
-    }).getReader();
-
-    const stream = createOpenAIArenaStream({
-      reader: upstreamReader,
-      model: "test-model",
-      signal: new AbortController().signal,
-    });
-
-    // Wrap in a Response so ensureStreamReadiness can consume it
-    const response = new Response(stream, {
-      status: 200,
-      headers: { "Content-Type": "text/event-stream" },
-    });
-
-    // This should not throw TypeError
-    const result = await ensureStreamReadiness(response, {
-      timeoutMs: 5000,
-      provider: "lmarena",
-      model: "test-model",
-    });
-
-    assert.ok(result.ok, "Stream readiness should succeed, not throw ERR_INVALID_ARG_TYPE");
-    if (result.ok) {
-      // Verify the stream body can be read
-      const reader = result.response.body?.getReader();
-      assert.ok(reader, "Should have a readable body");
-      const decoder = new TextDecoder();
-      let fullText = "";
-      while (true) {
-        const { done, value } = await reader!.read();
-        if (done) break;
-        fullText += decoder.decode(value, { stream: true });
-      }
-      fullText += decoder.decode();
-      // Should contain the SSE data we sent
-      assert.ok(fullText.includes("Hello"), "Stream should contain the SSE text");
-      assert.ok(fullText.includes("world"), "Stream should contain the SSE text");
-      assert.ok(fullText.includes("[DONE]"), "Stream should end with [DONE] marker");
-    }
-  });
-
-  it("does not throw TypeError when reading chunks directly", async () => {
-    const upstreamReader = new ReadableStream<Uint8Array>({
-      start(controller) {
-        controller.enqueue(new TextEncoder().encode('a0:{"text":"Hello"}\nad:{}\n'));
-        controller.close();
-      },
-    }).getReader();
-
-    const stream = createOpenAIArenaStream({
-      reader: upstreamReader,
-      model: "test-model",
-      signal: new AbortController().signal,
-    });
-
-    const reader = stream.getReader();
-    const chunks: Uint8Array[] = [];
-    while (true) {
-      const { done, value } = await reader.read();
-      if (done) break;
-      // If value is a string, this would throw
-      chunks.push(value);
-    }
-    // All chunks should be Uint8Array, not string
-    assert.ok(chunks.length > 0, "Should have produced at least one chunk");
-    for (const chunk of chunks) {
-      assert.ok(chunk instanceof Uint8Array, "Each chunk must be Uint8Array, not string");
-    }
-  });
-});

tests/unit/mcp-connect-scope.test.ts

@@ -56,14 +56,13 @@ test.after(() => {
   else process.env.INITIAL_PASSWORD = ORIGINAL_INITIAL;
 });
 
-function mgmtCtx(headers: Headers, method = "GET", pathname = "/api/keys", peerAddress?: string) {
+function mgmtCtx(headers: Headers, method = "GET", pathname = "/api/keys") {
   return {
     request: {
       method,
       headers,
       url: `http://localhost${pathname}`,
       nextUrl: { pathname },
-      socket: peerAddress ? { remoteAddress: peerAddress } : undefined,
     },
     classification: {
       routeClass: "MANAGEMENT" as const,
@@ -269,37 +268,3 @@ test("mcp:connect key is still rejected for the non-bypassable /api/cli-tools/ru
     assert.equal(out.code, "LOCAL_ONLY");
   }
 });
-
-// ─── 7. #9159 — loopback/LAN mcp:connect with requireLogin ──────────────────
-
-test("#9159 mcp:connect-only key must pass /api/mcp/ from loopback when login is required", async () => {
-  await seedAuthRequired();
-  const created = await apiKeysDb.createApiKey("mcp-loopback-only", "machine-mcp-loopback", [
-    MCP_CONNECT_SCOPE,
-  ]);
-  const out = await managementPolicy.evaluate(
-    mgmtCtx(
-      new Headers({ authorization: `Bearer ${created.key}` }),
-      "POST",
-      "/api/mcp/stream",
-      "127.0.0.1"
-    )
-  );
-  assert.equal(out.allow, true, JSON.stringify(out));
-});
-
-test("#9159 mcp:connect-only key must pass /api/mcp/ from private LAN when login is required", async () => {
-  await seedAuthRequired();
-  const created = await apiKeysDb.createApiKey("mcp-lan-only", "machine-mcp-lan", [
-    MCP_CONNECT_SCOPE,
-  ]);
-  const out = await managementPolicy.evaluate(
-    mgmtCtx(
-      new Headers({ authorization: `Bearer ${created.key}` }),
-      "POST",
-      "/api/mcp/stream",
-      "192.168.1.20"
-    )
-  );
-  assert.equal(out.allow, true, JSON.stringify(out));
-});

tests/unit/provider-registry-github-copilot-targetformat.test.ts

@@ -67,9 +67,6 @@ for (const id of [
   "gpt-5.4-mini",
   "gpt-5.4",
   "gpt-5.5",
-  "gpt-5.6-sol",
-  "gpt-5.6-terra",
-  "gpt-5.6-luna",
   "mai-code-1-flash",
   "gpt-5-mini",
   "oswe-vscode-prime",

tests/unit/quality-rail-gate-membership.test.ts

@@ -48,15 +48,19 @@ test("lint-guard carries the quality-ratchet engine (collect → ratchet → req
 
 test("fast-gates carries the deterministic ratchets and security scanners from the main rail", () => {
   const block = jobBlock("fast-gates");
-  // #8542: all gates run inside a single aggregation step's bash loop.
-  // Check that the gate names appear in the arrays or the loop body.
   for (const needle of [
-    "cycles lockfile duplication dead-code type-coverage compression-budget",
-    "secrets vuln-ratchet workflows openapi-breaking",
-    "typecheck:core",
-    "check:dashboard-typecheck",
+    "npm run check:cycles",
+    "npm run check:lockfile",
+    "npm run check:duplication",
+    "npm run check:dead-code",
+    "npm run check:type-coverage",
+    "npm run check:compression-budget",
+    "npm run check:secrets -- --ratchet",
+    "npm run check:vuln-ratchet -- --ratchet",
+    "npm run check:workflows -- --ratchet",
+    "npm run check:openapi-breaking -- --ratchet",
   ]) {
-    assert.ok(block.includes(needle), `fast-gates must contain "${needle}"`);
+    assert.ok(block.includes(needle), `fast-gates must run "${needle}"`);
   }
   assert.ok(
     block.includes(
@@ -82,7 +86,7 @@ test("fast-gates carries the deterministic ratchets and security scanners from t
 
 test("the complexity ratchet stays on the release rail (G0's written validation criterion)", () => {
   assert.ok(
-    jobBlock("fast-gates").includes("complexity-ratchets"),
+    jobBlock("fast-gates").includes("npm run check:complexity-ratchets"),
     "a complexity regression in a PR→release/** must be blocked by fast-gates"
   );
 });

tests/unit/radar-apply-feed.test.ts

@@ -15,11 +15,7 @@
 
 import test from "node:test";
 import assert from "node:assert/strict";
-import {
-  applyFeed,
-  type MergedEntry,
-  type FeedModel,
-} from "../../src/lib/radar/applyFeed.ts";
+import { applyFeed, type MergedEntry, type FeedModel } from "../../src/lib/radar/applyFeed.ts";
 import {
   getRadarCatalog,
   baselineToMergedEntries,
@@ -70,7 +66,9 @@ function makeBaseline(): MergedEntry[] {
   ];
 }
 
-function makeFeedModel(overrides: Partial<FeedModel> & { provider: string; modelId: string }): FeedModel {
+function makeFeedModel(
+  overrides: Partial<FeedModel> & { provider: string; modelId: string }
+): FeedModel {
   return {
     displayName: overrides.displayName ?? overrides.modelId,
     familyId: null,
@@ -116,7 +114,7 @@ test("rule 1: feed does NOT overwrite a local override field", () => {
   });
 
   const groq = result.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   )!;
 
   // Local override fields must survive
@@ -150,7 +148,7 @@ test("rule 2: feed enabled:false disables entry and carries disabledBy provenanc
   });
 
   const groq = result.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   )!;
 
   assert.equal(groq.enabled, false);
@@ -191,9 +189,7 @@ test("rule 3: user-added entry not in feed survives untouched", () => {
     tombstones: new Set(),
   });
 
-  const custom = result.find(
-    (e) => e.provider === "custom" && e.modelId === "my-local-model",
-  )!;
+  const custom = result.find((e) => e.provider === "custom" && e.modelId === "my-local-model")!;
 
   assert.equal(custom.displayName, "My Local Model");
   assert.equal(custom.monthlyTokens, 50_000);
@@ -242,7 +238,7 @@ test("rule 3b: user-added entry that IS in the feed merges with rule 1", () => {
   });
 
   const groq = result.filter(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   );
 
   // Should be deduplicated to ONE entry
@@ -262,7 +258,7 @@ test("rule 3b: user-added entry that IS in the feed merges with rule 1", () => {
 test("rule 4: tombstone prevents feed from resurrecting a deleted entry", () => {
   // Baseline has an entry for gemini, but user deleted it
   const baseline = makeBaseline().filter(
-    (e) => !(e.provider === "gemini" && e.modelId === "gemini-2.5-flash"),
+    (e) => !(e.provider === "gemini" && e.modelId === "gemini-2.5-flash")
   );
 
   const feed: FeedModel[] = [
@@ -283,9 +279,7 @@ test("rule 4: tombstone prevents feed from resurrecting a deleted entry", () =>
     tombstones,
   });
 
-  const gemini = result.find(
-    (e) => e.provider === "gemini" && e.modelId === "gemini-2.5-flash",
-  );
+  const gemini = result.find((e) => e.provider === "gemini" && e.modelId === "gemini-2.5-flash");
 
   // Must NOT be resurrected
   assert.equal(gemini, undefined);
@@ -353,9 +347,7 @@ test("applyFeed: feed-only entry is added with origin 'radar'", () => {
     tombstones: new Set(),
   });
 
-  const added = result.find(
-    (e) => e.provider === "new-provider" && e.modelId === "new-model",
-  );
+  const added = result.find((e) => e.provider === "new-provider" && e.modelId === "new-model");
 
   assert.ok(added, "feed-only entry should be present");
   assert.equal(added.displayName, "Brand New Model");
@@ -388,7 +380,7 @@ test("applyFeed: feed fields merge over baseline where no local override", () =>
   });
 
   const groq = result.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   )!;
 
   // Feed values win when no local override
@@ -440,7 +432,7 @@ test("applyFeed: duplicate key (baseline + feed) produces single merged entry",
   });
 
   const groqEntries = result.filter(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   );
 
   assert.equal(groqEntries.length, 1, "should be deduplicated to one entry");
@@ -465,9 +457,7 @@ test("rule 4b: tombstoned entry removed even when baseline has it", () => {
     tombstones,
   });
 
-  const groq = result.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
-  );
+  const groq = result.find((e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile");
 
   assert.equal(groq, undefined, "tombstoned entry should be excluded");
 });
@@ -494,9 +484,7 @@ test("origin switches to 'radar' when feed updates a baseline entry", () => {
     tombstones: new Set(),
   });
 
-  const gemini = result.find(
-    (e) => e.provider === "gemini" && e.modelId === "gemini-2.5-flash",
-  )!;
+  const gemini = result.find((e) => e.provider === "gemini" && e.modelId === "gemini-2.5-flash")!;
 
   assert.equal(gemini.origin, "radar");
   assert.equal(gemini.displayName, "Updated Gemini");
@@ -612,7 +600,7 @@ test("getRadarCatalog: valid cache returns merged entries with meta", () => {
 
   // The feed has groq:llama-3.3-70b-versatile, which merges over baseline
   const groq = result.entries.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   )!;
 
   assert.equal(groq.displayName, "Feed Groq Name");
@@ -714,7 +702,7 @@ test("FIX2 mergeOne path: contextWindow/capabilities/limits/setup survive merge
   });
 
   const groq = result.find(
-    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile",
+    (e) => e.provider === "groq" && e.modelId === "llama-3.3-70b-versatile"
   )!;
 
   assert.equal(groq.contextWindow, 131072);
@@ -758,13 +746,11 @@ test("FIX2 feedModelToMerged path: contextWindow/capabilities/limits/setup survi
 });
 
 // ===========================================================================
-// FIX 4 — feedModelToMerged() must honor an `enabled` local override instead
-// of unconditionally forcing `enabled:false` when the feed disables the model.
-// mergeOne() already gets this right (overrides applied AFTER rule 2); this
-// pins the same semantics on the feed-only path.
+// Feed `enabled:false` is the safety exception to local override precedence:
+// a model confirmed dead upstream must not be resurrected locally.
 // ===========================================================================
 
-test("FIX4: feed-only entry with local override enabled:true wins over feed enabled:false", () => {
+test("rule 2: feed-only entry stays disabled even with local enabled:true", () => {
   const baseline = makeBaseline();
   const feed: FeedModel[] = [
     makeFeedModel({
@@ -786,11 +772,11 @@ test("FIX4: feed-only entry with local override enabled:true wins over feed enab
   });
 
   const entry = result.find(
-    (e) => e.provider === "new-provider" && e.modelId === "disabled-model",
+    (e) => e.provider === "new-provider" && e.modelId === "disabled-model"
   )!;
 
-  assert.equal(entry.enabled, true, "local override must win over feed disable");
-  assert.equal(entry.disabledBy, undefined, "must not carry radar disabledBy when overridden on");
+  assert.equal(entry.enabled, false, "a local override must not resurrect a dead upstream model");
+  assert.equal(entry.disabledBy, "radar");
 });
 
 test("FIX4: feed-only entry with NO override still gets disabled with disabledBy provenance", () => {
@@ -811,7 +797,7 @@ test("FIX4: feed-only entry with NO override still gets disabled with disabledBy
   });
 
   const entry = result.find(
-    (e) => e.provider === "new-provider" && e.modelId === "disabled-model-2",
+    (e) => e.provider === "new-provider" && e.modelId === "disabled-model-2"
   )!;
 
   assert.equal(entry.enabled, false);

tests/unit/radar-db.test.ts

@@ -165,7 +165,9 @@ test("setRadarKey encrypts at rest and getRadarSettings decrypts", () => {
   assert.equal(settings.supporterKey, clearKey, "getRadarSettings must return the clear key");
 
   // Direct DB query to prove encryption at rest
-  interface SettingsRow { supporter_key_encrypted: string | null }
+  interface SettingsRow {
+    supporter_key_encrypted: string | null;
+  }
   const row = db
     .prepare("SELECT supporter_key_encrypted FROM radar_settings WHERE id = 1")
     .get() as SettingsRow;
@@ -197,6 +199,36 @@ test("setRadarKey(null) clears the key", () => {
   assert.equal(cleared.supporter_key_encrypted, null, "DB value must be null");
 });
 
+test("changing the supporter key invalidates the entitlement-sensitive referrals cache", () => {
+  radar.setRadarCache({
+    version: "2026.08.07.1",
+    tier: "live",
+    payload: '{"models":[]}',
+    signature: "catalog-live-signature",
+  });
+  radar.setRadarReferralsCache({
+    generatedAt: "2026-08-07T12:00:00.000Z",
+    tier: "live",
+    payload: '{"referrals":{"fixed":[],"campaigns":[{"provider":"groq"}]}}',
+    signature: "live-signature",
+  });
+  assert.ok(radar.getRadarReferralsCache(), "precondition: live referrals cache exists");
+  assert.ok(radar.getRadarCache(), "precondition: live catalog cache exists");
+
+  radar.setRadarKey("omr_" + "d".repeat(40));
+
+  assert.equal(
+    radar.getRadarReferralsCache(),
+    null,
+    "a new key must force the next referrals read to resolve entitlement server-side"
+  );
+  assert.equal(
+    radar.getRadarCache(),
+    null,
+    "a new key must force the next catalog sync to resolve entitlement server-side"
+  );
+});
+
 test("setRadarKey uses existing AES-256-GCM encryption from encryption.ts", () => {
   const db = core.getDbInstance();
   const clearKey = "omr_" + "c".repeat(40);
@@ -271,7 +303,9 @@ test("second setRadarReferralsCache REPLACES the row (still single row)", () =>
   assert.equal(result.tier, "live", "must have the second tier");
   assert.equal(result.payload, '{"new":true}', "must have the second payload");
 
-  const count = db.prepare("SELECT COUNT(*) AS c FROM radar_referrals_cache").get() as { c: number };
+  const count = db.prepare("SELECT COUNT(*) AS c FROM radar_referrals_cache").get() as {
+    c: number;
+  };
   assert.equal(count.c, 1, "must have exactly one row");
 });
 

tests/unit/radar-localized-feed.test.ts

@@ -0,0 +1,45 @@
+import assert from "node:assert/strict";
+import { readFileSync } from "node:fs";
+import { test } from "node:test";
+
+import { RadarFeedSchema } from "../../src/lib/radar/feedSchema.ts";
+
+const fixture = JSON.parse(
+  readFileSync(new URL("../fixtures/radar-feed-canonical.json", import.meta.url), "utf8")
+) as Record<string, unknown>;
+
+test("canonical fixture carries D25 localized setup and accepts localized quirks", () => {
+  const localized = structuredClone(fixture) as {
+    models: Array<{ setup: { steps: unknown[] } | null }>;
+    quirks: Array<{ title: unknown; body: unknown }>;
+  };
+  localized.quirks = [
+    {
+      slug: "shared-pool",
+      title: { en: "Shared quota", pt: "Cota compartilhada" },
+      body: { en: "Models share one pool." },
+      severity: "info",
+      targets: [{ provider: "groq", modelGlob: null }],
+    },
+  ];
+  const parsed = RadarFeedSchema.parse(localized);
+  assert.deepEqual(parsed.models[0]!.setup!.steps[0], {
+    en: "Create a free account in the Groq console",
+    pt: "Crie uma conta gratuita no console da Groq",
+  });
+});
+
+test("RadarFeedSchema preserves schema-v1 legacy setup and quirk strings", () => {
+  const legacy = structuredClone(fixture) as {
+    models: Array<{ setup: { steps: unknown[] } | null }>;
+    quirks: Array<{ title: unknown; body: unknown }>;
+  };
+  legacy.models[0]!.setup!.steps[0] = "Create an account";
+  legacy.quirks[0]!.title = "Shared quota";
+  legacy.quirks[0]!.body = "Models share one pool.";
+
+  const parsed = RadarFeedSchema.parse(legacy);
+  assert.equal(parsed.models[0]!.setup!.steps[0], "Create an account");
+  assert.equal(parsed.quirks[0]!.title, "Shared quota");
+  assert.equal(parsed.quirks[0]!.body, "Models share one pool.");
+});

tests/unit/radar-referrals-sync.test.ts

@@ -118,7 +118,11 @@ test("RadarReferralsFeedSchema: rejects missing referrals section", () => {
   const feed = baseReferralsFeed();
   delete (feed as Record<string, unknown>).referrals;
   const result = referralsFeedSchema.RadarReferralsFeedSchema.safeParse(feed);
-  assert.equal(result.success, false, "referrals section is required (no old-feed compat needed here)");
+  assert.equal(
+    result.success,
+    false,
+    "referrals section is required (no old-feed compat needed here)"
+  );
 });
 
 test("RadarReferralsFeedSchema: rejects a non-https referral url", () => {
@@ -215,7 +219,10 @@ test("syncRadarReferrals: valid signature => cache updated, payload byte-identic
     },
     fetch: (() =>
       Promise.resolve(
-        mockResponse(bytes, { "x-omniroute-feed-signature": sig, "x-omniroute-feed-tier": "community" })
+        mockResponse(bytes, {
+          "x-omniroute-feed-signature": sig,
+          "x-omniroute-feed-tier": "community",
+        })
       )) as unknown as typeof globalThis.fetch,
     now: () => new Date("2026-08-07T12:05:00.000Z"),
   });
@@ -296,32 +303,47 @@ test("syncRadarReferrals: valid sig over garbage JSON => invalid_schema, cache u
 // syncRadarReferrals — generatedAt floor (replay/no-op guard)
 // ===========================================================================
 
-test("syncRadarReferrals: same generatedAt as cache => stale, cache untouched", async () => {
+test("syncRadarReferrals: same generatedAt with a new served tier replaces the cache", async () => {
   const feed = baseReferralsFeed("2026-08-07T12:00:00.000Z");
+  (feed.referrals as { campaigns: Array<Record<string, unknown>> }).campaigns = [
+    {
+      provider: "groq",
+      url: "https://groq.com/?campaign=live",
+      kind: "campanha",
+      validUntil: null,
+      requiredAction: null,
+      isDefault: false,
+    },
+  ];
   const bytes = feedBytes(feed);
   const sig = signBytes(bytes);
-  let cacheWritten = false;
+  const cacheStore: referralsSync.RadarReferralsCacheEntry[] = [];
 
   const result = await referralsSync.syncRadarReferrals({
     getFlag: () => true,
-    getSettings: () => ({ optIn: true, supporterKey: null }),
+    getSettings: () => ({ optIn: true, supporterKey: "omr_" + "a".repeat(40) }),
     getCache: () => ({
       generatedAt: "2026-08-07T12:00:00.000Z",
       tier: "community",
       payload: "{}",
       signature: "old-sig",
     }),
-    setCache: () => {
-      cacheWritten = true;
+    setCache: (entry) => {
+      cacheStore.push(entry);
     },
     fetch: (() =>
       Promise.resolve(
-        mockResponse(bytes, { "x-omniroute-feed-signature": sig })
+        mockResponse(bytes, {
+          "x-omniroute-feed-signature": sig,
+          "x-omniroute-feed-tier": "live",
+        })
       )) as unknown as typeof globalThis.fetch,
   });
 
-  assert.equal(result.status, "stale");
-  assert.equal(cacheWritten, false);
+  assert.equal(result.status, "updated");
+  assert.equal(cacheStore.length, 1);
+  assert.equal(cacheStore[0]!.tier, "live");
+  assert.equal(cacheStore[0]!.payload, bytes.toString("utf-8"));
 });
 
 test("syncRadarReferrals: older generatedAt than cache => stale (replay rejected)", async () => {
@@ -487,7 +509,9 @@ test("syncRadarReferrals: header absent => falls back to 'community' (no body ti
       cacheStore.push(entry);
     },
     fetch: (() =>
-      Promise.resolve(mockResponse(bytes, { "x-omniroute-feed-signature": sig }))) as unknown as typeof globalThis.fetch,
+      Promise.resolve(
+        mockResponse(bytes, { "x-omniroute-feed-signature": sig })
+      )) as unknown as typeof globalThis.fetch,
   });
 
   assert.equal(result.status, "updated");
@@ -596,7 +620,9 @@ test("syncRadarReferrals: HTTP non-200 => error mentioning the status code", asy
     getFlag: () => true,
     getSettings: () => ({ optIn: true, supporterKey: null }),
     fetch: (() =>
-      Promise.resolve(mockResponse(Buffer.from("Internal Server Error"), {}, 500))) as unknown as typeof globalThis.fetch,
+      Promise.resolve(
+        mockResponse(Buffer.from("Internal Server Error"), {}, 500)
+      )) as unknown as typeof globalThis.fetch,
   });
 
   assert.equal(result.status, "error");
@@ -646,7 +672,8 @@ test("FIX: oversized body without a trustworthy Content-Length header => too_lar
     setCache: () => {
       setCacheCalled = true;
     },
-    fetch: (() => Promise.resolve(mockResponse(oversized, {}))) as unknown as typeof globalThis.fetch,
+    fetch: (() =>
+      Promise.resolve(mockResponse(oversized, {}))) as unknown as typeof globalThis.fetch,
   });
 
   assert.deepEqual(result, { status: "too_large" });

tests/unit/radar-setup-connections.test.ts

@@ -0,0 +1,28 @@
+import assert from "node:assert/strict";
+import { test } from "node:test";
+
+import {
+  firstProviderConnectionId,
+  providerConnectionsRequestUrl,
+} from "../../src/lib/radar/setupConnections.ts";
+
+test("providerConnectionsRequestUrl filters the existing providers API", () => {
+  assert.equal(
+    providerConnectionsRequestUrl("openrouter/custom"),
+    "/api/providers?provider=openrouter%2Fcustom"
+  );
+});
+
+test("firstProviderConnectionId selects a real connection id, never the provider slug", () => {
+  assert.equal(
+    firstProviderConnectionId(
+      [
+        { id: "connection-disabled", provider: "groq", isActive: false },
+        { id: "connection-active", provider: "groq", isActive: true },
+      ],
+      "groq"
+    ),
+    "connection-active"
+  );
+  assert.equal(firstProviderConnectionId([], "groq"), null);
+});

tests/unit/radar-sync.test.ts

@@ -409,6 +409,35 @@ test("syncRadar: version floor — incoming older => stale", async () => {
   assert.equal(cacheWritten, false, "cache must NOT be overwritten with older version");
 });
 
+test("syncRadar: an entitlement downgrade replaces a newer live cache with community", async () => {
+  const sig = signBytes(FIXTURE_BYTES);
+  const cacheStore: syncMod.RadarCacheEntry[] = [];
+
+  const result = await syncMod.syncRadar({
+    getFlag: () => true,
+    getSettings: () => ({ optIn: true, supporterKey: "omr_" + "a".repeat(40) }),
+    getCache: () => ({
+      version: "2026.08.02.1",
+      tier: "live",
+      payload: "{}",
+      signature: "old-live-sig",
+    }),
+    setCache: (entry) => cacheStore.push(entry),
+    fetch: (() =>
+      Promise.resolve(
+        mockResponse(FIXTURE_BYTES, {
+          "x-omniroute-feed-signature": sig,
+          "x-omniroute-feed-tier": "community",
+        })
+      )) as unknown as typeof globalThis.fetch,
+  });
+
+  assert.equal(result.status, "updated");
+  assert.equal(cacheStore.length, 1);
+  assert.equal(cacheStore[0]!.tier, "community");
+  assert.equal(cacheStore[0]!.version, "2026.08.01.1");
+});
+
 test("syncRadar: version floor — incoming newer => updated", async () => {
   // Modify fixture to have a newer version
   const fixtureObj = JSON.parse(FIXTURE_STRING);

tests/unit/repro-8542.test.ts

@@ -1,54 +0,0 @@
-import { test } from "node:test";
-import assert from "node:assert/strict";
-import { readFileSync } from "node:fs";
-import { fileURLToPath } from "node:url";
-import { dirname, resolve } from "node:path";
-import { parse } from "yaml";
-
-const __dirname = dirname(fileURLToPath(import.meta.url));
-const repoRoot = resolve(__dirname, "../..");
-const WORKFLOW = resolve(repoRoot, ".github/workflows/quality.yml");
-
-function loadWorkflow(): any {
-  return parse(readFileSync(WORKFLOW, "utf8"));
-}
-
-function invokesGate(run: string): boolean {
-  if (!run) return false;
-  return /npm run (check:|typecheck:)/.test(run) || /npm run "check/.test(run) || /npm run \\"check/.test(run);
-}
-function stepCanFail(step: any): boolean {
-  return step?.["continue-on-error"] !== true;
-}
-
-test("repro #8542: fast-gates must not fail-fast into a later gate", () => {
-  const wf = loadWorkflow();
-  const job = wf.jobs?.["fast-gates"];
-  assert.ok(job, "fast-gates job must exist");
-  const steps: any[] = job.steps ?? [];
-  assert.ok(steps.length >= 5, `fast-gates must have >=5 steps, got ${steps.length}`);
-
-  const gateSteps = steps.map((s, i) => ({ s, i })).filter(({ s }) => invokesGate(s?.run ?? ""));
-  assert.ok(gateSteps.length >= 1, `expected >=1 gate step, got ${gateSteps.length}`);
-
-  const maskedPairs: string[] = [];
-  for (let a = 0; a < gateSteps.length; a++) {
-    const stepA = gateSteps[a];
-    if (!stepCanFail(stepA.s)) continue;
-    for (let b = a + 1; b < gateSteps.length; b++) {
-      const stepB = gateSteps[b];
-      maskedPairs.push(
-        `step ${stepA.i + 1} (${stepA.s.name ?? String(stepA.s.run).split("\n")[0].slice(0, 40)})` +
-          ` can fail and masks step ${stepB.i + 1} (${stepB.s.name ?? String(stepB.s.run).split("\n")[0].slice(0, 40)})`
-      );
-    }
-  }
-
-  assert.deepEqual(
-    maskedPairs,
-    [],
-    `FAIL-FAST MASKING PRESENT (${maskedPairs.length} pair(s)): a failing gate step aborts the job and every later gate reports "skipped". This is the #8542 mechanism.\n` +
-      maskedPairs.slice(0, 12).join("\n") +
-      (maskedPairs.length > 12 ? `\n... (+${maskedPairs.length - 12} more)` : "")
-  );
-});

tests/unit/route-body-validation-t06.test.ts

@@ -1,54 +0,0 @@
-/**
- * Guard for the t06:route-validation gate (Hard Rule #7 — always validate inputs
- * with Zod schemas).
- *
- * The gate (scripts/check/check-route-validation.mjs) is a source scan: any
- * `route.ts` under src/app/api that calls `request.json()` must also call
- * `validateBody()` or `.safeParse()`. It has NO allowlist, so a route that
- * hand-rolls `typeof x === "string"` checks passes review but fails CI — which
- * is exactly how four routes (#9445 marketplace install, #8523's three Dario
- * admin routes) landed on release/v3.8.50 and kept the branch out of
- * release-green (#9737).
- *
- * This test runs the same rule inside the unit suite so the violation surfaces
- * on the PR that introduces it, instead of on the next base-red sweep.
- */
-import test from "node:test";
-import assert from "node:assert/strict";
-import fs from "node:fs";
-import path from "node:path";
-
-const REPO_ROOT = path.resolve(import.meta.dirname, "..", "..");
-const API_ROOT = path.join(REPO_ROOT, "src", "app", "api");
-
-function collectRouteFiles(dir: string): string[] {
-  const files: string[] = [];
-  for (const entry of fs.readdirSync(dir, { withFileTypes: true })) {
-    const full = path.join(dir, entry.name);
-    if (entry.isDirectory()) {
-      files.push(...collectRouteFiles(full));
-    } else if (entry.isFile() && entry.name === "route.ts") {
-      files.push(full);
-    }
-  }
-  return files;
-}
-
-test("every API route reading request.json() validates it with Zod (t06)", () => {
-  const offenders: string[] = [];
-
-  for (const file of collectRouteFiles(API_ROOT)) {
-    const source = fs.readFileSync(file, "utf8");
-    if (!/request\.json\s*\(/.test(source)) continue;
-    if (/\bvalidateBody\s*\(/.test(source) || /\.safeParse\s*\(/.test(source)) continue;
-    offenders.push(path.relative(REPO_ROOT, file));
-  }
-
-  assert.deepEqual(
-    offenders,
-    [],
-    `routes call request.json() without validateBody()/.safeParse() — hand-rolled ` +
-      `typeof checks do not satisfy Hard Rule #7 and fail the t06 CI gate:\n  ` +
-      offenders.join("\n  ")
-  );
-});

tests/unit/sseHeartbeat.test.ts

@@ -23,17 +23,11 @@ test("sseCommentsEnabled defaults to true when the env var is unset", () => {
   withEnv(undefined, () => assert.equal(sseCommentsEnabled(), true));
 });
 
-test("sseCommentsEnabled is false for 'off', 'false', '0', 'no' (case-insensitive)", () => {
+test("sseCommentsEnabled is false only when set to 'off' (case-insensitive)", () => {
   withEnv("off", () => assert.equal(sseCommentsEnabled(), false));
   withEnv("OFF", () => assert.equal(sseCommentsEnabled(), false));
-  withEnv("false", () => assert.equal(sseCommentsEnabled(), false));
-  withEnv("FALSE", () => assert.equal(sseCommentsEnabled(), false));
-  withEnv("0", () => assert.equal(sseCommentsEnabled(), false));
-  withEnv("no", () => assert.equal(sseCommentsEnabled(), false));
-  withEnv("NO", () => assert.equal(sseCommentsEnabled(), false));
   withEnv("on", () => assert.equal(sseCommentsEnabled(), true));
-  withEnv("yes", () => assert.equal(sseCommentsEnabled(), true));
-  withEnv("1", () => assert.equal(sseCommentsEnabled(), true));
+  withEnv("false", () => assert.equal(sseCommentsEnabled(), true));
 });
 
 test("shapeForClientFormat maps known client formats", () => {

tests/unit/triage-bugs-2026-08-02.test.ts

@@ -1,61 +1,120 @@
+/**
+ * #8853 — authenticated HTTP proxy health checks drop credentials
+ *
+ * Root cause: both the auto-test route and the scheduler build proxy URLs
+ * manually as `${proxy.type}://${proxy.host}:${proxy.port}`, dropping
+ * username/password. The `proxyConfigToUrl()` function in proxyDispatcher.ts
+ * already handles URL-encoded credentials correctly.
+ *
+ * We prove the bug by showing that the proxy URL produced by the current
+ * manual construction lacks credentials, and that `proxyConfigToUrl()` with
+ * the same config object includes them — therefore the fix is to reuse it.
+ */
 import test from "node:test";
 import assert from "node:assert/strict";
-import { openaiResponsesToOpenAIRequest } from "../../open-sse/translator/request/openai-responses.ts";
 
-function asRecord(value: unknown): Record<string, unknown> {
-  return value as Record<string, unknown>;
+// The function that fixes the bug — we import it here to verify it works
+import { proxyConfigToUrl } from "@omniroute/open-sse/utils/proxyDispatcher";
+
+// ── proxyConfigToUrl credential tests ──────────────────────────────────────
+
+test("#8853 proxyConfigToUrl encodes username and password into proxy URL", () => {
+  const url = proxyConfigToUrl({
+    type: "http",
+    host: "127.0.0.1",
+    port: 3128,
+    username: "alice",
+    password: "s3cret",
+  });
+  assert.ok(url, "proxyConfigToUrl must return a URL");
+  assert.match(url!, /:\/\/alice:s3cret@/, "URL must contain credentials");
+});
+
+test("#8853 proxyConfigToUrl encodes special characters in credentials", () => {
+  const url = proxyConfigToUrl({
+    type: "http",
+    host: "proxy.example.com",
+    port: 8080,
+    username: "user@domain",
+    password: "p@ss:w0rd",
+  });
+  assert.ok(url, "proxyConfigToUrl must return a URL");
+  assert.match(url!, /:\/\/user%40domain:p%40ss%3Aw0rd@/, "URL must URL-encode special chars");
+});
+
+test("#8853 proxyConfigToUrl omits auth when no username", () => {
+  const url = proxyConfigToUrl({
+    type: "http",
+    host: "127.0.0.1",
+    port: 3128,
+  });
+  assert.ok(url, "proxyConfigToUrl must return a URL");
+  assert.doesNotMatch(url!, /@/, "URL must not contain @ (no auth)");
+});
+
+test("#8853 proxyConfigToUrl handles IPv6 host with family", () => {
+  const url = proxyConfigToUrl({
+    type: "http",
+    host: "[::1]",
+    port: 3128,
+    family: "ipv6",
+  });
+  assert.ok(url, "proxyConfigToUrl must return a URL");
+  assert.match(url!, /\[::1\]/, "IPv6 host must be bracketed");
+});
+
+// ── Simulate the buggy construction ─────────────────────────────────────────
+
+function buggyManualUrl(proxy: { type: string; host: string; port: number }) {
+  return `${proxy.type}://${proxy.host}:${proxy.port}`;
 }
 
-for (const variant of ["gpt-5.6-sol", "gpt-5.6-terra", "gpt-5.6-luna"]) {
-  test(`#8997 ${variant} nested reasoning.effort max survives promotion`, () => {
-    const translated = asRecord(
-      openaiResponsesToOpenAIRequest(
-        variant,
-        { model: variant, input: "hello", reasoning: { effort: "max" } },
-        false,
-        {}
-      )
-    );
-    assert.equal(translated.reasoning_effort, "max");
-  });
+test("#8853 manual URL construction (current bug) drops credentials", () => {
+  const proxy = {
+    type: "http",
+    host: "127.0.0.1",
+    port: 3128,
+    username: "alice",
+    password: "s3cret",
+  };
+  const manualUrl = buggyManualUrl(proxy);
+  assert.doesNotMatch(manualUrl, /alice/, "Buggy URL must NOT contain username");
+  assert.doesNotMatch(manualUrl, /s3cret/, "Buggy URL must NOT contain password");
 
-  test(`#8997 ${variant} flat reasoning_effort max survives promotion`, () => {
-    const translated = asRecord(
-      openaiResponsesToOpenAIRequest(
-        variant,
-        { model: variant, input: "hello", reasoning_effort: "max" },
-        false,
-        {}
-      )
-    );
-    assert.equal(translated.reasoning_effort, "max");
-  });
-}
+  // Compare with proxyConfigToUrl which includes credentials
+  const fixedUrl = proxyConfigToUrl(proxy);
+  assert.ok(fixedUrl);
+  assert.match(fixedUrl!, /alice/, "Fixed URL must contain username");
+  assert.match(fixedUrl!, /s3cret/, "Fixed URL must contain password");
+});
 
-test("non-GPT-5.6 models still get max downgraded to xhigh", () => {
-  const translated = asRecord(
-    openaiResponsesToOpenAIRequest(
-      "gpt-4o",
-      { model: "gpt-4o", input: "hello", reasoning: { effort: "max" } },
-      false,
-      {}
-    )
-  );
-  assert.equal(translated.reasoning_effort, "xhigh");
+// ── Verify the scheduler and auto-test would use proxyConfigToUrl ──────────
 
-// #9140 — VS Code routes filter out built-in auto models
-const { isUsableChatModel } = await import(
-  "../../src/app/api/v1/vscode/[token]/usableChatModel.ts"
-);
+test("#8853 proxyConfigToUrl accepts ProxyRegistryRecord-shaped object", () => {
+  // Simulating the shape of a proxy record returned by listProxies({ includeSecrets: true })
+  const proxyRecord = {
+    id: "p1",
+    name: "test",
+    type: "http",
+    host: "10.0.0.1",
+    port: 8888,
+    username: "bob",
+    password: "p4ss",
+    family: "auto",
+    region: null,
+    notes: null,
+    status: "active",
+    source: "manual",
+    subscriptionId: null,
+    createdAt: "2026-01-01",
+    updatedAt: "2026-01-01",
+  };
+  const url = proxyConfigToUrl(proxyRecord);
+  assert.ok(url, "proxyConfigToUrl must accept ProxyRegistryRecord-shaped objects");
+  assert.match(url!, /bob:p4ss/, "URL must include credentials from the record");
+});
 
-test("#9140 VS Code listing must accept built-in auto routing entries", () => {
-  assert.equal(
-    isUsableChatModel({ id: "auto/best-coding", owned_by: "combo" }),
-    true,
-    "built-in auto/* model should be accepted"
-  );
-  assert.equal(
-    isUsableChatModel({ id: "operator-combo", owned_by: "combo" }),
-    false,
-    "operator-created combo should still be rejected"
-  );
+test("#8853 proxyConfigToUrl returns null for partial config (no host)", () => {
+  const url = proxyConfigToUrl({ type: "http", port: 8080 } as Record<string, unknown>);
+  assert.equal(url, null, "proxyConfigToUrl must return null for partial config without host");
+});