fix(dream): stamp incremental extraction watermark (#2636)

The Dream cycle disables sync's inline extraction and routes changed
slugs through extractForSlugs, which flushed link/timeline batches but
never stamped links_extracted_at — so incrementally extracted pages
stayed permanently visible to `extract --stale` / doctor.

Collect processedRefs per successfully processed page and stamp them
via stampExtracted (best-effort) after both batch flushes, non-dry-run
mode 'all' only. Source-id threading from the original PR #2637 already
landed on master via #1503/#1747, so this rebase carries only the
missing watermark stamp plus regression tests.

Takeover of #2637.

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

src/commands/backlinks.ts

@@ -5,8 +5,8 @@
  * checks if back-links exist, and optionally creates them.
  *
  * Usage:
- *   gbrain check-backlinks check [dir] [--dir <brain-dir>] # report missing back-links
- *   gbrain check-backlinks fix [dir] [--dir <brain-dir>]   # create missing back-links
+ *   gbrain check-backlinks check [--dir <brain-dir>]     # report missing back-links
+ *   gbrain check-backlinks fix [--dir <brain-dir>]        # create missing back-links
  *   gbrain check-backlinks fix --dry-run                  # preview fixes
  */
 
@@ -201,40 +201,6 @@ export interface BacklinksResult {
   dryRun: boolean;
 }
 
-export interface ParsedBacklinksArgs {
-  subcommand: string | undefined;
-  brainDir: string;
-  dryRun: boolean;
-}
-
-export function parseBacklinksArgs(args: string[]): ParsedBacklinksArgs {
-  const subcommand = args[0];
-  const dryRun = args.includes('--dry-run');
-  const dirIdx = args.indexOf('--dir');
-  const flagDir = dirIdx >= 0 && args[dirIdx + 1] && !args[dirIdx + 1].startsWith('--')
-    ? args[dirIdx + 1]
-    : undefined;
-
-  let positionalDir: string | undefined;
-  for (let i = 1; i < args.length; i++) {
-    const arg = args[i];
-    if (arg === '--dir') {
-      i++;
-      continue;
-    }
-    if (arg === '--dry-run') continue;
-    if (arg.startsWith('--')) continue;
-    positionalDir = arg;
-    break;
-  }
-
-  return {
-    subcommand,
-    brainDir: flagDir ?? positionalDir ?? '.',
-    dryRun,
-  };
-}
-
 /**
  * Library-level backlinks check/fix. Throws on validation errors; returns a
  * structured result so Minions handlers + autopilot-cycle can surface counts.
@@ -270,14 +236,16 @@ export async function runBacklinksCore(opts: BacklinksOpts): Promise<BacklinksRe
 }
 
 export async function runBacklinks(args: string[]) {
-  const { subcommand, brainDir, dryRun } = parseBacklinksArgs(args);
+  const subcommand = args[0];
+  const dirIdx = args.indexOf('--dir');
+  const brainDir = dirIdx >= 0 ? args[dirIdx + 1] : '.';
+  const dryRun = args.includes('--dry-run');
 
   if (!subcommand || !['check', 'fix'].includes(subcommand)) {
-    console.error('Usage: gbrain check-backlinks <check|fix> [dir] [--dir <brain-dir>] [--dry-run]');
+    console.error('Usage: gbrain check-backlinks <check|fix> [--dir <brain-dir>] [--dry-run]');
     console.error('  check    Report missing back-links');
     console.error('  fix      Create missing back-links (appends to Timeline)');
-    console.error('  dir      Brain directory (default: current directory)');
-    console.error('  --dir    Brain directory override');
+    console.error('  --dir    Brain directory (default: current directory)');
     console.error('  --dry-run  Preview fixes without writing');
     process.exit(1);
   }

src/commands/extract.ts

@@ -1025,6 +1025,10 @@ async function extractForSlugs(
   let linksCreated = 0;
   let timelineCreated = 0;
   let pagesProcessed = 0;
+  // #2636: successfully processed pages get their extraction watermark
+  // stamped after the final flush (mode 'all' only — a partial-mode run
+  // hasn't done the full extraction the watermark asserts).
+  const processedRefs: Array<{ slug: string; source_id: string }> = [];
 
   // Issue #972: read the basename flag once per extract run.
   const globalBasename = await isGlobalBasenameEnabled(engine);
@@ -1113,6 +1117,7 @@ async function extractForSlugs(
         }
 
         pagesProcessed++;
+        if (!dryRun) processedRefs.push({ slug, source_id: sourceId ?? 'default' });
       } catch { /* skip unreadable */ }
       progress.tick(1);
     },
@@ -1120,6 +1125,13 @@ async function extractForSlugs(
 
   await flushLinks();
   await flushTimeline();
+  // #2636: the Dream cycle disables sync's inline extraction and routes
+  // changed slugs through this incremental path — without a stamp here,
+  // those pages never get links_extracted_at and stay permanently visible
+  // to `extract --stale` / doctor. Stamp only after BOTH batches flushed.
+  if (!dryRun && mode === 'all') {
+    await stampExtracted(engine, processedRefs);
+  }
   progress.finish();
 
   if (!jsonMode) {

test/backlinks.test.ts

@@ -4,7 +4,6 @@ import {
   extractPageTitle,
   hasBacklink,
   buildBacklinkEntry,
-  parseBacklinksArgs,
 } from '../src/commands/backlinks.ts';
 
 describe('extractEntityRefs', () => {
@@ -105,26 +104,3 @@ describe('findBacklinkGaps dedupe (v0.36.x #967 regression)', () => {
     }
   });
 });
-
-describe('parseBacklinksArgs', () => {
-  test('uses positional dir for check and fix subcommands', () => {
-    expect(parseBacklinksArgs(['check', '/tmp/brain']).brainDir).toBe('/tmp/brain');
-    expect(parseBacklinksArgs(['fix', '/tmp/brain']).brainDir).toBe('/tmp/brain');
-  });
-
-  test('defaults to cwd when no dir given', () => {
-    expect(parseBacklinksArgs(['check']).brainDir).toBe('.');
-  });
-
-  test('--dir overrides positional dir and preserves dry-run', () => {
-    const parsed = parseBacklinksArgs(['fix', '/tmp/ignored', '--dir', '/tmp/brain', '--dry-run']);
-    expect(parsed.subcommand).toBe('fix');
-    expect(parsed.brainDir).toBe('/tmp/brain');
-    expect(parsed.dryRun).toBe(true);
-  });
-
-  test('--dir missing its value falls back to positional dir', () => {
-    expect(parseBacklinksArgs(['check', '/tmp/brain', '--dir']).brainDir).toBe('/tmp/brain');
-    expect(parseBacklinksArgs(['check', '--dir', '--dry-run']).brainDir).toBe('.');
-  });
-});

test/extract-incremental.test.ts

@@ -60,6 +60,51 @@ async function seedPage(slug: string, body: string): Promise<void> {
 }
 
 describe('runExtractCore — incremental cycle path (#417)', () => {
+  test('Dream incremental all-mode stamps the source-scoped extraction watermark (#2636)', async () => {
+    await engine.executeRaw(
+      `INSERT INTO sources (id, name, local_path) VALUES ($1, $2, $3)`,
+      ['repo-a', 'repo-a', tempDir],
+    );
+    await engine.putPage('people/alice-example', {
+      type: 'person',
+      title: 'alice-example',
+      compiled_truth: '# alice',
+      timeline: '',
+      frontmatter: {},
+      content_hash: 'h',
+    }, { sourceId: 'repo-a' });
+    writeFileSync(join(tempDir, 'people/alice-example.md'), '# alice');
+
+    await runExtractCore(engine as unknown as BrainEngine, {
+      mode: 'all',
+      dir: tempDir,
+      slugs: ['people/alice-example'],
+      sourceId: 'repo-a',
+    });
+
+    const rows = await engine.executeRaw<{ links_extracted_at: string | null }>(
+      `SELECT links_extracted_at FROM pages WHERE slug = $1 AND source_id = $2`,
+      ['people/alice-example', 'repo-a'],
+    );
+    expect(rows[0]?.links_extracted_at).not.toBeNull();
+    expect(await engine.countStalePagesForExtraction({ sourceId: 'repo-a' })).toBe(0);
+  });
+
+  test('Dream incremental dry-run does NOT stamp the watermark', async () => {
+    await seedPage('people/alice-example', '# alice');
+    await runExtractCore(engine as unknown as BrainEngine, {
+      mode: 'all',
+      dir: tempDir,
+      slugs: ['people/alice-example'],
+      dryRun: true,
+    });
+    const rows = await engine.executeRaw<{ links_extracted_at: string | null }>(
+      `SELECT links_extracted_at FROM pages WHERE slug = $1`,
+      ['people/alice-example'],
+    );
+    expect(rows[0]?.links_extracted_at ?? null).toBeNull();
+  });
+
   test('1. slugs: [] returns immediately with zero counts (early-return path)', async () => {
     await seedPage('people/alice-example', '# alice');
     const result = await runExtractCore(engine as unknown as BrainEngine, {