fix(webhook): extract links for incremental push syncs (#2849)

GitHub webhook + sync-trigger 'sync' jobs never opted into extraction:
the standalone sync handler defaults noExtract to TRUE, so webhook-
imported pages advanced the source bookmark with links_extracted_at
unstamped — and nothing behind a standalone job ever swept them (the
autopilot cycle's extract phase only walks that cycle's changedSlugs).

Fix, both halves of the issue:
- webhook + sync-trigger payloads pass noExtract:false so normal
  incremental pushes get sync's inline link/timeline extraction.
- >100-file deferral branch: performSync surfaces extractDeferred, and
  the sync job handler queues a source-scoped extract{stale:true}
  follow-up job (idempotent sweep; extract handler gains a stale mode
  backed by the same extractStaleFromDB core as `gbrain extract --stale`).

Tests: new sync-large-deferral-extract.serial.test.ts drives a real
101-file incremental sync end-to-end (deferral flag → queued follow-up →
stale sweep recovers links + watermark); webhook payload contract test +
sync-trigger noExtract assertion salvaged from #2850.

Fixes #2849
Takeover of #2850

Co-authored-by: patentsong <patentsong@users.noreply.github.com>
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

src/commands/extract.ts

@@ -1651,7 +1651,7 @@ async function extractTimelineFromDB(
  * make re-extraction idempotent). EVERY processed page is stamped, including
  * zero-link pages — they WERE processed.
  */
-async function extractStaleFromDB(
+export async function extractStaleFromDB(
   engine: BrainEngine,
   opts: {
     dryRun: boolean;

src/commands/jobs.ts

@@ -1479,7 +1479,31 @@ export async function registerBuiltinHandlers(
       embedSkipReason = 'auto_embed_disabled';
     }
 
-    return { ...result, embed_job_id: embedJobId, embed_skip_reason: embedSkipReason };
+    // #2849: large-sync extract deferral follow-up. performSync skips inline
+    // link/timeline extraction when totalChanges > 100, leaving
+    // links_extracted_at unstamped. A standalone sync job (webhook push,
+    // sync trigger) has no autopilot extract phase behind it, so the pages
+    // would stay extraction-stale until a manual `gbrain extract --stale`.
+    // Queue a source-scoped stale sweep instead. Best-effort + idempotent:
+    // a duplicate sweep finds 0 stale pages and no-ops.
+    let extractJobId: number | null = null;
+    if (result.extractDeferred) {
+      try {
+        const { MinionQueue } = await import('../core/minions/queue.ts');
+        const queue = new MinionQueue(engine);
+        const followUp = await queue.add(
+          'extract',
+          { stale: true, ...(sourceId ? { sourceId } : {}) },
+          {
+            idempotency_key: `sync-extract-stale:${sourceId ?? 'default'}:${Math.floor(Date.now() / 30_000)}`,
+            maxWaiting: 1,
+          },
+        );
+        extractJobId = followUp.id;
+      } catch { /* best-effort: extract --stale sweeps it later */ }
+    }
+
+    return { ...result, embed_job_id: embedJobId, embed_skip_reason: embedSkipReason, extract_stale_job_id: extractJobId };
   });
 
   registerBuiltinJob(worker, engine, 'embed', async (job) => {
@@ -1652,6 +1676,20 @@ export async function registerBuiltinHandlers(
   });
 
   worker.register('extract', async (job) => {
+    // #2849: stale-sweep mode — the sync handler's large-sync deferral
+    // follow-up. DB-source (reads page content from the DB, so it runs on
+    // checkout-less brains), source-scopable, idempotent. Same core as
+    // `gbrain extract --stale`.
+    if (job.data.stale === true) {
+      const { extractStaleFromDB } = await import('./extract.ts');
+      return await extractStaleFromDB(engine, {
+        dryRun: !!job.data.dryRun,
+        jsonMode: false,
+        includeFrontmatter: false,
+        sourceIdFilter: typeof job.data.sourceId === 'string' ? job.data.sourceId : undefined,
+        catchUp: false,
+      });
+    }
     const { runExtractCore } = await import('./extract.ts');
     const mode = (typeof job.data.mode === 'string' && ['links', 'timeline', 'all'].includes(job.data.mode))
       ? (job.data.mode as 'links' | 'timeline' | 'all')

src/commands/onboard.ts

@@ -142,16 +142,12 @@ export async function runOnboard(engine: BrainEngine, args: string[]): Promise<v
 
   // --auto path: runs through the T2 library orchestrator. Hooks emit CLI
   // progress to stderr; the final result lands as JSON on stdout (or human
-  // summary). extraRemediations (gathered above from runAllOnboardChecks)
-  // is threaded into the runner so the onboard-check remediations
-  // (extract-ner, extract-timeline-from-meetings, etc.) reach the planner
-  // — the same wiring the --check path uses above.
+  // summary).
   const result = await runRemediation(
     engine,
     {
       targetScore,
       maxUsd,
-      extraRemediations,
       // --auto --yes opts into the prompt_required tier too; library
       // doesn't distinguish auto_apply vs prompt_required, it just runs
       // every remediation in the plan. The plan-building side (T12 render)

src/commands/serve-http.ts

@@ -2146,8 +2146,13 @@ export async function runServeHttp(engine: BrainEngine, options: ServeHttpOption
   //     Other event types (ping, pull_request, etc.) return 202 'ignored'
   //     so GitHub doesn't retry.
   // D15.5: HMAC compare uses the shared safeHexEqual helper.
-  // D18: submits 'sync' job with auto_embed_backfill=true and priority -10
-  //     (above autopilot's 0).
+  // D18: submits 'sync' job with extraction + auto_embed_backfill enabled and
+  //     priority -10 (above autopilot's 0). noExtract:false opts normal
+  //     incremental pushes into sync's inline link/timeline extraction (#2849
+  //     — the standalone sync handler defaults noExtract to TRUE, which left
+  //     webhook-imported pages permanently stale). Large (>100 file) pushes
+  //     defer inline extract; the sync handler queues an extract --stale
+  //     follow-up job for that branch.
   // ---------------------------------------------------------------------------
   const githubWebhookLimiter = rateLimit({
     windowMs: 60_000,
@@ -2267,6 +2272,7 @@ export async function runServeHttp(engine: BrainEngine, options: ServeHttpOption
           'sync',
           {
             sourceId: source.id,
+            noExtract: false,
             auto_embed_backfill: true,
             embed_reason: 'webhook',
           },

src/commands/sync.ts

@@ -222,6 +222,14 @@ export interface SyncResult {
    * everything," the exact misdiagnosis in the #1794 recurrence report.
    */
   bankedFiles?: number;
+  /**
+   * #2849: true when extraction was REQUESTED (noExtract false) but this sync
+   * skipped inline link/timeline extraction because totalChanges > 100 (the
+   * #1794 large-sync deferral). links_extracted_at stays unstamped for the
+   * imported pages. The standalone `sync` job handler queues a source-scoped
+   * `extract --stale` follow-up when set; CLI runs print the manual hint.
+   */
+  extractDeferred?: boolean;
 }
 
 /**
@@ -1379,6 +1387,10 @@ See also:
     {
       sourceId: sourceIdArg,
       repoPath: source.local_path,
+      // #2849: opt in to inline extraction — the standalone sync handler
+      // defaults noExtract to TRUE (dedupe for doctor's [sync, extract]
+      // remediation plan), which would leave triggered syncs extraction-stale.
+      noExtract: false,
       auto_embed_backfill: true,
       embed_reason: 'sync_trigger',
     },
@@ -3287,11 +3299,16 @@ async function performSyncInner(engine: BrainEngine, opts: SyncOpts): Promise<Sy
   // the stale sweep scans the whole source, so banked-across-runs pages are
   // covered regardless.
   const extractOpts = opts.sourceId ? { sourceId: opts.sourceId } : undefined;
+  let extractDeferred = false;
   if (!opts.noExtract && totalChanges > 100 && pagesAffected.length > 0) {
+    // #2849: surface the deferral to callers. A standalone sync job (webhook
+    // push, sync trigger) has no autopilot extract phase behind it, so the
+    // job handler queues an `extract --stale` follow-up off this flag.
+    extractDeferred = true;
     slog(
       `  Large sync: deferring link/timeline extraction. ` +
       `Run 'gbrain extract --stale${opts.sourceId ? ` --source-id ${opts.sourceId}` : ''}' ` +
-      `(or let the autopilot cycle's extract phase sweep it).`,
+      `(sync jobs queue this follow-up automatically).`,
     );
   }
   if (!opts.noExtract && totalChanges <= 100 && pagesAffected.length > 0) {
@@ -3400,6 +3417,7 @@ async function performSyncInner(engine: BrainEngine, opts: SyncOpts): Promise<Sy
     chunksCreated,
     embedded,
     pagesAffected,
+    extractDeferred,
   };
 }
 

src/core/operations.ts

@@ -4957,7 +4957,7 @@ const run_onboard: Operation = {
     // typo, the underlying queue.add would reject. Defense-in-depth.
     const result = await runRemediation(
       ctx.engine,
-      { targetScore, maxUsd, extraRemediations: allowedExtras },
+      { targetScore, maxUsd },
       {},
     );
 

src/core/remediation/run.ts

@@ -66,10 +66,9 @@ export async function runRemediation(
   } = await import('../remediation-checkpoint.ts');
 
   const ctx = await loadRecommendationContext(engine);
-  const extraRemediations = opts.extraRemediations ?? [];
 
   // Pre-flight ceiling check via the shared plan computation.
-  const initialPlan = await computeRemediationPlan(engine, { targetScore, extraRemediations });
+  const initialPlan = await computeRemediationPlan(engine, { targetScore });
   if (initialPlan.target_unreachable) {
     hooks.onTargetUnreachable?.(targetScore, initialPlan.max_reachable_score);
     return {
@@ -88,7 +87,7 @@ export async function runRemediation(
   }
 
   const initialHealth = await engine.getHealth();
-  let recs: RemediationStep[] = computeRecommendations(initialHealth, ctx, extraRemediations)
+  let recs: RemediationStep[] = computeRecommendations(initialHealth, ctx)
     .filter((r) => r.status === 'remediable');
   if (recs.length === 0) {
     hooks.onNothingToDo?.(initialHealth.brain_score, targetScore);
@@ -306,13 +305,7 @@ export async function runRemediation(
       // steps with bumped retry suffix (D1).
       if (recs.length === 0 || stepCount >= maxJobs) break;
       const freshHealth = await engine.getHealth();
-      // Extras carry a static status:'remediable' — a fresh health snapshot
-      // never ages them out the way health-derived steps drop. Filter out
-      // ids this run already processed (any terminal status), or the recheck
-      // would resubmit completed extras every iteration, forever.
-      const processedIds = new Set(submitted.map((s) => s.id));
-      const pendingExtras = extraRemediations.filter((r) => !processedIds.has(r.id));
-      recs = computeRecommendations(freshHealth, ctx, pendingExtras).filter((r) => r.status === 'remediable');
+      recs = computeRecommendations(freshHealth, ctx).filter((r) => r.status === 'remediable');
     }
   };
 

src/core/remediation/types.ts

@@ -63,16 +63,6 @@ export interface RemediationOpts {
   resumePlanHash?: string;
   /** Whether to attempt resume at all (default false). */
   resume?: boolean;
-  /**
-   * Caller-supplied RemediationStep entries threaded into the planner.
-   * Mirrors RemediationPlanOpts.extraRemediations so onboard's --apply
-   * --auto path (and MCP run_onboard auto modes) forward the same
-   * onboard-check remediations the --check path already passes through
-   * computeRemediationPlan. Without this the runner saw only generic
-   * brain_score remediations and reported "Nothing to do" whenever the
-   * only applicable work was an extra (e.g. extract-ner).
-   */
-  extraRemediations?: RemediationStep[];
 }
 
 /**

test/remediation-run-extras.serial.test.ts

@@ -1,108 +0,0 @@
-// test/remediation-run-extras.serial.test.ts
-// Regression for PR #2161 takeover: `gbrain onboard --apply --auto` dropped
-// onboard-check extraRemediations. Two distinct halves of the bug:
-//   1. runRemediation built the pre-flight plan + initial recs WITHOUT the
-//      extras, so an extras-only plan reported "Nothing to do".
-//   2. The D7 mid-run recheck rebuilt recs WITHOUT the extras after every
-//      completed step, so with 2+ plannable steps all remaining extras were
-//      dropped after step 1. The recheck must also filter out extras this
-//      run already processed — extras carry static status:'remediable', so
-//      unfiltered threading would resubmit completed extras forever.
-//
-// SERIAL: mock.module (queue + wait-for-completion stubs, R2) + GBRAIN_HOME
-// env mutation so checkpoint files land in a tmpdir, not ~/.gbrain.
-
-import { describe, expect, test, beforeAll, afterAll, mock } from 'bun:test';
-import { mkdtempSync, rmSync } from 'node:fs';
-import { tmpdir } from 'node:os';
-import { join } from 'node:path';
-import { PGLiteEngine } from '../src/core/pglite-engine.ts';
-import { makeRemediationStep } from '../src/core/remediation-step.ts';
-
-// Stub the Minion queue: every submitted job is immediately 'completed'.
-// runRemediation only calls queue.add + waitForCompletion(queue, id).
-let nextJobId = 1;
-const submittedJobs: Array<{ name: string }> = [];
-mock.module('../src/core/minions/queue.ts', () => ({
-  MinionQueue: class {
-    async add(name: string) {
-      submittedJobs.push({ name });
-      return { id: nextJobId++, status: 'completed' };
-    }
-  },
-}));
-mock.module('../src/core/minions/wait-for-completion.ts', () => ({
-  waitForCompletion: async () => ({ status: 'completed' }),
-}));
-
-let engine: PGLiteEngine;
-let home: string;
-const prevHome = process.env.GBRAIN_HOME;
-
-beforeAll(async () => {
-  home = mkdtempSync(join(tmpdir(), 'gbrain-remextras-'));
-  process.env.GBRAIN_HOME = home;
-  engine = new PGLiteEngine();
-  await engine.connect({});
-  await engine.initSchema();
-}, 120_000);
-
-afterAll(async () => {
-  await engine.disconnect();
-  if (prevHome === undefined) delete process.env.GBRAIN_HOME;
-  else process.env.GBRAIN_HOME = prevHome;
-  rmSync(home, { recursive: true, force: true });
-});
-
-function extra(id: string, job: string) {
-  return makeRemediationStep({
-    id,
-    job,
-    params: {},
-    severity: 'medium',
-    est_seconds: 5,
-    est_usd_cost: 0,
-    rationale: 'synthetic onboard-check extra',
-    status: 'remediable',
-  });
-}
-
-describe('runRemediation extraRemediations threading', () => {
-  test('extras-only plan runs BOTH extras and terminates (no Nothing-to-do, no resubmit loop)', async () => {
-    // Empty PGLite brain → zero health-derived recommendations. Without the
-    // fix, half 1 makes this run return submitted: [] via onNothingToDo.
-    // With only half 1 (the original PR #2161 diff), the mid-run recheck
-    // drops the second extra after step 1 — submitted has 1 entry, not 2.
-    const { runRemediation } = await import('../src/core/remediation/run.ts');
-    let nothingToDo = false;
-    const result = await runRemediation(
-      engine,
-      {
-        targetScore: 1,
-        extraRemediations: [
-          extra('onboard.extract_ner', 'extract-ner'),
-          extra('onboard.extract_timeline', 'extract-timeline-from-meetings'),
-        ],
-        // Safety bound: an unfiltered recheck would resubmit completed
-        // extras forever; maxJobs turns that regression into a fast fail
-        // (extra count > 1 below) instead of a hung test.
-        maxJobs: 5,
-      },
-      { onNothingToDo: () => { nothingToDo = true; } },
-    );
-
-    expect(nothingToDo).toBe(false);
-    const ids = result.submitted.map((s) => s.id);
-    expect(ids).toContain('onboard.extract_ner');
-    expect(ids).toContain('onboard.extract_timeline');
-    // Each extra ran exactly once — the recheck must not re-plan extras the
-    // run already processed.
-    expect(ids.filter((i) => i === 'onboard.extract_ner').length).toBe(1);
-    expect(ids.filter((i) => i === 'onboard.extract_timeline').length).toBe(1);
-    expect(result.submitted.every((s) => s.status === 'completed')).toBe(true);
-    expect(submittedJobs.map((j) => j.name).sort()).toEqual([
-      'extract-ner',
-      'extract-timeline-from-meetings',
-    ]);
-  });
-});

test/sources-webhook.test.ts

@@ -16,6 +16,7 @@
  */
 import { describe, test, expect } from 'bun:test';
 import { createHmac } from 'node:crypto';
+import { readFileSync } from 'node:fs';
 import { safeHexEqual } from '../src/core/timing-safe.ts';
 
 const GITHUB_SECRET = 'super-secret-webhook-key';
@@ -123,3 +124,25 @@ describe('Branch ref construction (D5)', () => {
     expect(pushedRef === `refs/heads/${trackedBranch}`).toBe(false);
   });
 });
+
+describe('Webhook sync job extraction contract (#2849)', () => {
+  test('opts into extraction before the pushed commit is consumed', () => {
+    const serveSource = readFileSync(
+      new URL('../src/commands/serve-http.ts', import.meta.url),
+      'utf8',
+    );
+    const routeStart = serveSource.indexOf("'/webhooks/github'");
+    const queueStart = serveSource.indexOf('const job = await queue.add(', routeStart);
+    const responseStart = serveSource.indexOf('res.status(202)', queueStart);
+    expect(routeStart).toBeGreaterThanOrEqual(0);
+    expect(queueStart).toBeGreaterThan(routeStart);
+    expect(responseStart).toBeGreaterThan(queueStart);
+
+    const routeSource = serveSource.slice(queueStart, responseStart);
+    const payload = routeSource.match(
+      /queue\.add\(\s*'sync',\s*\{([\s\S]*?)\}\s*,\s*\{/,
+    );
+    expect(payload).not.toBeNull();
+    expect(payload?.[1]).toMatch(/\bnoExtract:\s*false\b/);
+  });
+});

test/sync-large-deferral-extract.serial.test.ts

@@ -0,0 +1,130 @@
+/**
+ * #2849 — large-sync extract deferral queues an `extract --stale` follow-up.
+ *
+ * performSync's incremental path skips inline link/timeline extraction when
+ * totalChanges > 100 (the #1794 large-sync deferral), leaving
+ * links_extracted_at unstamped. Pre-fix, a standalone sync job (webhook push,
+ * `gbrain sync trigger`) had NOTHING behind it to sweep those pages — the
+ * autopilot cycle's extract phase only walks that cycle's changedSlugs — so a
+ * large webhook push left extraction permanently stale until a manual
+ * `gbrain extract --stale`.
+ *
+ * Pins:
+ *   (a) performSync surfaces `extractDeferred: true` on the >100 branch and
+ *       leaves the pages unstamped/unlinked.
+ *   (b) the `sync` job handler queues an `extract` job with
+ *       { stale: true, sourceId? } when extractDeferred is set.
+ *   (c) the `extract` handler's stale mode actually sweeps: links created +
+ *       watermark stamped (end-to-end recovery, no manual step).
+ *
+ * Marked .serial.test.ts — spawns git subprocesses + shares one PGLite engine.
+ */
+
+import { describe, test, expect, beforeAll, afterAll } from 'bun:test';
+import { mkdtempSync, writeFileSync, rmSync, mkdirSync } from 'fs';
+import { execSync } from 'child_process';
+import { tmpdir } from 'os';
+import { join } from 'path';
+import { PGLiteEngine } from '../src/core/pglite-engine.ts';
+import { MinionWorker } from '../src/core/minions/worker.ts';
+import { MinionQueue } from '../src/core/minions/queue.ts';
+import { registerBuiltinHandlers } from '../src/commands/jobs.ts';
+
+let engine: PGLiteEngine;
+let worker: MinionWorker;
+let repoPath: string;
+
+function git(cmd: string): void { execSync(cmd, { cwd: repoPath, stdio: 'pipe' }); }
+
+describe('#2849 — large sync defers extract and queues a stale sweep', () => {
+  beforeAll(async () => {
+    engine = new PGLiteEngine();
+    await engine.connect({});
+    await engine.initSchema();
+    worker = new MinionWorker(engine, { queue: 'test' });
+    await registerBuiltinHandlers(worker, engine, { quiet: true });
+
+    repoPath = mkdtempSync(join(tmpdir(), 'gbrain-large-defer-'));
+    git('git init');
+    git('git config user.email "t@t.com"');
+    git('git config user.name "T"');
+    mkdirSync(join(repoPath, 'people'), { recursive: true });
+    mkdirSync(join(repoPath, 'notes'), { recursive: true });
+    writeFileSync(join(repoPath, 'people/alice.md'), [
+      '---', 'type: person', 'title: Alice', '---', '', 'Alice is a founder.',
+    ].join('\n'));
+    git('git add -A && git commit -m "initial"');
+
+    // Seed: full first sync imports the anchor page + sets last_commit.
+    const { performSync } = await import('../src/commands/sync.ts');
+    await performSync(engine, { repoPath, full: true, noPull: true, noEmbed: true });
+
+    // Second commit: 101 new pages → incremental totalChanges > 100.
+    for (let i = 0; i < 101; i++) {
+      writeFileSync(join(repoPath, `notes/n${i}.md`), [
+        '---', 'type: note', `title: Note ${i}`, '---', '',
+        `[Alice](people/alice) appears in note ${i}.`,
+      ].join('\n'));
+    }
+    git('git add -A && git commit -m "add 101 pages"');
+  }, 120_000);
+
+  afterAll(async () => {
+    if (repoPath) rmSync(repoPath, { recursive: true, force: true });
+    if (engine) await engine.disconnect();
+  }, 60_000);
+
+  test('sync handler defers inline extract and queues extract{stale} follow-up; stale sweep recovers', async () => {
+    const syncHandler = (worker as unknown as { handlers: Map<string, (job: unknown) => Promise<unknown>> })
+      .handlers.get('sync');
+    expect(syncHandler).toBeDefined();
+
+    // Same payload shape the webhook submits (minus embed backfill noise).
+    const result = await syncHandler!({
+      data: { repoPath, noExtract: false, noPull: true, auto_embed_backfill: false },
+      signal: { aborted: false },
+      updateProgress: async () => {},
+    }) as { status: string; extractDeferred?: boolean; extract_stale_job_id?: number | null };
+
+    expect(result.status).toBe('synced');
+    // (a) inline extract was deferred, pages left stale.
+    expect(result.extractDeferred).toBe(true);
+    const staleBefore = await engine.countStalePagesForExtraction();
+    expect(staleBefore).toBeGreaterThan(100);
+    expect(await engine.getLinks('notes/n0')).toHaveLength(0);
+
+    // (b) a follow-up extract job with stale:true was queued.
+    expect(result.extract_stale_job_id).toBeGreaterThan(0);
+    const queue = new MinionQueue(engine);
+    const extractJobs = await queue.getJobs({ name: 'extract', limit: 5 });
+    expect(extractJobs.length).toBe(1);
+    expect((extractJobs[0].data as { stale: boolean }).stale).toBe(true);
+
+    // (c) running the extract handler's stale mode recovers: links + stamps.
+    const extractHandler = (worker as unknown as { handlers: Map<string, (job: unknown) => Promise<unknown>> })
+      .handlers.get('extract');
+    await extractHandler!({
+      data: extractJobs[0].data,
+      signal: { aborted: false },
+      updateProgress: async () => {},
+    });
+    const links = await engine.getLinks('notes/n0');
+    expect(links.some(l => l.to_slug === 'people/alice')).toBe(true);
+    const rows = await engine.executeRaw<{ links_extracted_at: string | null }>(
+      `SELECT links_extracted_at FROM pages WHERE slug = 'notes/n0'`,
+    );
+    expect(rows[0]?.links_extracted_at).not.toBeNull();
+  }, 180_000);
+
+  test('sub-threshold sync does NOT set extractDeferred (no spurious follow-up)', async () => {
+    // One more small commit → inline extract path, no deferral.
+    writeFileSync(join(repoPath, 'notes/small.md'), [
+      '---', 'type: note', 'title: Small', '---', '', 'No big deal.',
+    ].join('\n'));
+    git('git add -A && git commit -m "one small page"');
+    const { performSync } = await import('../src/commands/sync.ts');
+    const result = await performSync(engine, { repoPath, noPull: true, noEmbed: true });
+    expect(result.status).toBe('synced');
+    expect(result.extractDeferred).toBeFalsy();
+  }, 60_000);
+});

test/sync-trigger-cli.test.ts

@@ -100,6 +100,10 @@ describe('runSyncTrigger', () => {
     const job = jobs[0];
     expect(job.priority).toBe(-10);
     expect((job.data as { sourceId: string }).sourceId).toBe('default');
+    // #2849: opt in to inline extraction — the standalone sync handler
+    // defaults noExtract to TRUE, which would leave triggered syncs
+    // extraction-stale.
+    expect((job.data as { noExtract: boolean }).noExtract).toBe(false);
     expect((job.data as { auto_embed_backfill: boolean }).auto_embed_backfill).toBe(true);
   });