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/embed.ts

@@ -107,14 +107,6 @@ export interface EmbedOpts {
    * runs lock every source in sorted order. dryRun skips it.
    */
   singleFlight?: boolean;
-  /**
-   * #394: suppress human stdout summaries (the `[dry-run] Would embed ...` /
-   * `Embedded N chunks ...` slog lines). Set by structured-output callers —
-   * the cycle's embed phase (dream --json must keep stdout JSON-clean per
-   * docs/progress-events.md) reports counts via its own PhaseResult instead.
-   * Errors/warnings still go to stderr regardless.
-   */
-  quiet?: boolean;
 }
 
 /**
@@ -261,7 +253,7 @@ export async function runEmbedCore(engine: BrainEngine, opts: EmbedOpts): Promis
     for (const s of opts.slugs) {
       if (isAborted(opts.signal)) break; // #1737: stop the per-slug loop on abort
       try {
-        await embedPage(engine, s, !!opts.dryRun, result, opts.sourceId, opts.signal, opts.quiet);
+        await embedPage(engine, s, !!opts.dryRun, result, opts.sourceId, opts.signal);
       } catch (e: unknown) {
         serr(`  Error embedding ${s}: ${e instanceof Error ? e.message : e}`);
       }
@@ -355,7 +347,6 @@ export async function runEmbedCore(engine: BrainEngine, opts: EmbedOpts): Promis
         catchUp: opts.catchUp,
         pacer,
         paceMaxConcurrency,
-        quiet: opts.quiet,
       }, opts.signal);
     } finally {
       // E1: surface pacing telemetry (human + structured) when pacing was on.
@@ -385,7 +376,7 @@ export async function runEmbedCore(engine: BrainEngine, opts: EmbedOpts): Promis
     return result;
   }
   if (opts.slug) {
-    await embedPage(engine, opts.slug, !!opts.dryRun, result, opts.sourceId, opts.signal, opts.quiet);
+    await embedPage(engine, opts.slug, !!opts.dryRun, result, opts.sourceId, opts.signal);
     return result;
   }
   throw new Error('No embed target specified. Pass { slug }, { slugs }, { all }, or { stale }.');
@@ -530,7 +521,6 @@ async function embedPage(
   result: EmbedResult,
   sourceId?: string,
   signal?: AbortSignal,
-  quiet?: boolean,
 ) {
   const opts = sourceId ? { sourceId } : undefined;
   const page = await engine.getPage(slug, opts);
@@ -575,7 +565,7 @@ async function embedPage(
   result.skipped += chunks.length - toEmbed.length;
 
   if (toEmbed.length === 0) {
-    if (!quiet) slog(`${slug}: all ${chunks.length} chunks already embedded`);
+    slog(`${slug}: all ${chunks.length} chunks already embedded`);
     result.pages_processed++;
     return;
   }
@@ -612,7 +602,7 @@ async function embedPage(
   }
   result.embedded += toEmbed.length;
   result.pages_processed++;
-  if (!quiet) slog(`${slug}: embedded ${toEmbed.length} chunks`);
+  slog(`${slug}: embedded ${toEmbed.length} chunks`);
 }
 
 async function embedAll(
@@ -630,8 +620,6 @@ async function embedAll(
     pacer?: DbPacer;
     /** Resolved concurrency cap (E-1: the worker count, no separate permit). */
     paceMaxConcurrency?: number;
-    /** #394: suppress human stdout summaries (structured-output callers). */
-    quiet?: boolean;
   },
   signal?: AbortSignal,
 ) {
@@ -775,12 +763,10 @@ async function embedAll(
   });
 
   // Stdout summary preserved for scripts/tests that grep for counts.
-  if (!staleOpts?.quiet) {
-    if (dryRun) {
-      slog(`[dry-run] Would embed ${result.would_embed} chunks across ${pages.length} pages`);
-    } else {
-      slog(`Embedded ${result.embedded} chunks across ${pages.length} pages`);
-    }
+  if (dryRun) {
+    slog(`[dry-run] Would embed ${result.would_embed} chunks across ${pages.length} pages`);
+  } else {
+    slog(`Embedded ${result.embedded} chunks across ${pages.length} pages`);
   }
 }
 
@@ -816,8 +802,6 @@ async function embedAllStale(
     pacer?: DbPacer;
     /** Resolved concurrency cap (E-1: the worker count, no separate permit). */
     paceMaxConcurrency?: number;
-    /** #394: suppress human stdout summaries (structured-output callers). */
-    quiet?: boolean;
   },
   signature?: string,
   externalSignal?: AbortSignal,
@@ -835,7 +819,7 @@ async function embedAllStale(
       signature,
       ...(sourceId && { sourceId }),
     });
-    if (invalidated > 0 && !staleOpts?.quiet) {
+    if (invalidated > 0) {
       slog(`[embed] invalidated ${invalidated} chunk(s) embedded under a prior model signature`);
     }
   }
@@ -846,12 +830,10 @@ async function embedAllStale(
     dryRun && signature ? { ...sourceOpt, signature } : sourceOpt,
   );
   if (staleCount === 0) {
-    if (!staleOpts?.quiet) {
-      if (dryRun) {
-        slog('[dry-run] Would embed 0 chunks (0 stale found)');
-      } else {
-        slog('Embedded 0 chunks (0 stale found)');
-      }
+    if (dryRun) {
+      slog('[dry-run] Would embed 0 chunks (0 stale found)');
+    } else {
+      slog('Embedded 0 chunks (0 stale found)');
     }
     return;
   }
@@ -860,7 +842,7 @@ async function embedAllStale(
     result.would_embed += staleCount;
     result.total_chunks += staleCount;
     if (onProgress) onProgress(1, 1, 0);
-    if (!staleOpts?.quiet) slog(`[dry-run] Would embed ${staleCount} stale chunks`);
+    slog(`[dry-run] Would embed ${staleCount} stale chunks`);
     return;
   }
 
@@ -1100,7 +1082,7 @@ async function embedAllStale(
     if (budgetTimer) clearTimeout(budgetTimer);
   }
 
-  if (!staleOpts?.quiet) slog(`Embedded ${result.embedded} chunks across ${totalProcessedPages} pages`);
+  slog(`Embedded ${result.embedded} chunks across ${totalProcessedPages} pages`);
 
   // #1946 (OV2a): a catch-up pass that completed without being aborted but left
   // chunks unembedded means those chunks are stuck (a non-transient embed

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/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/cycle.ts

@@ -1214,9 +1214,7 @@ async function runPhaseEmbed(engine: BrainEngine, dryRun: boolean, signal?: Abor
     // 10-15 min one) bails within a batch instead of running to completion
     // after the job was killed — which left gbrain_cycle_locks held and
     // wedged every subsequent autopilot cycle.
-    // #394: quiet — the cycle reports embed counts via its own PhaseResult;
-    // raw `[dry-run] Would embed ...` stdout lines would corrupt `dream --json`.
-    const result = await runEmbedCore(engine, { stale: true, dryRun, signal, quiet: true });
+    const result = await runEmbedCore(engine, { stale: true, dryRun, signal });
     const embeddedCount = dryRun ? result.would_embed : result.embedded;
     return {
       phase: 'embed',

test/dream.test.ts

@@ -292,33 +292,6 @@ describe('runDream — output format', () => {
     expect(parsed).toHaveProperty('totals');
   });
 
-  // #394 / takeover of #854: the embed phase's `[dry-run] Would embed ...`
-  // summary must not leak onto stdout ahead of the JSON CycleReport.
-  test('--dry-run --json emits only JSON even when embed has stale chunks', async () => {
-    await engine.putPage('concepts/testing', {
-      type: 'concept',
-      title: 'Testing',
-      compiled_truth: 'Testing keeps JSON contracts honest.',
-      timeline: '',
-    });
-    await engine.upsertChunks('concepts/testing', [
-      { chunk_index: 0, chunk_text: 'Testing keeps JSON contracts honest.', chunk_source: 'compiled_truth' },
-    ]);
-
-    const lines: string[] = [];
-    const logSpy = spyOn(console, 'log').mockImplementation((msg: string) => { lines.push(String(msg)); });
-    await runDream(engine, ['--dir', repo, '--phase', 'embed', '--dry-run', '--json']);
-    logSpy.mockRestore();
-
-    const output = lines.join('\n');
-    expect(output.trimStart().startsWith('{')).toBe(true);
-    const parsed = JSON.parse(output);
-    expect(parsed.schema_version).toBe('1');
-    expect(parsed.phases[0].phase).toBe('embed');
-    // The stale chunk was still counted in the structured report.
-    expect(parsed.phases[0].details.would_embed).toBe(1);
-  });
-
   test('human output for clean status mentions "Brain is healthy"', async () => {
     const lines: string[] = [];
     const logSpy = spyOn(console, 'log').mockImplementation((msg: string) => { lines.push(String(msg)); });

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);
   });