fix(onboard): stop dropping onboard-check remediations on the --apply --auto path

Takeover of #2161. runRemediation ignored onboard-check extras in three
places: the pre-flight plan, the initial recommendation build, and the D7
mid-run recheck that rebuilds recs after every completed step. The --check
path threaded extras correctly, so `gbrain onboard --apply --auto` reported
"Nothing to do" when the only remediable work came from onboard checks —
and even with the first two sites fixed (the original PR diff), any plan
with 2+ steps dropped all remaining extras after step 1 via the recheck.

- Add RemediationOpts.extraRemediations; thread it through the pre-flight
  plan, initial recs, and the mid-run recheck.
- Recheck filters extras to ids not already processed this run: extras
  carry static status:'remediable', so unfiltered threading would resubmit
  completed extras forever.
- Wire the CLI --auto path (onboard.ts) AND the MCP run_onboard auto path
  (operations.ts), which already computed the scope-filtered allowedExtras
  and then dropped it.
- Regression test: extras-only plan on an empty brain runs BOTH extras
  exactly once and terminates (serial file: mock.module queue stub +
  GBRAIN_HOME tmpdir).

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

src/commands/autopilot.ts

@@ -151,100 +151,6 @@ export function shouldSpawnAutopilotWorker(args: string[]): boolean {
   return !args.includes('--no-worker');
 }
 
-/**
- * #1525 — positional subcommand translation.
- *
- * Pre-fix, `gbrain autopilot status` silently fell through to "start daemon"
- * because `runAutopilot()` only branched on flag forms (`--status`, etc.).
- * `status` was treated as a stray positional and ignored.
- *
- * This translator maps known positional subcommands to their flag form so
- * `autopilot status` is equivalent to `autopilot --status`, then rejects
- * any unrecognized positional with a fail-loud error before any side
- * effect (lockfile, daemon spawn, sync dispatch) runs.
- *
- * Scope decisions:
- *   - Known aliases: `status` → `--status`, `install` → `--install`,
- *     `uninstall` → `--uninstall`, `start` → (drop; default daemon launch).
- *   - `stop` is intentionally NOT aliased here. Stopping a running daemon
- *     is a new behavior (read PID from lock, SIGTERM, drain) that deserves
- *     its own design and PR. Users typing `gbrain autopilot stop` today get
- *     the unknown-positional error with the canonical alternatives.
- *   - At most one positional allowed; multiple positionals fail loud.
- */
-// Every flag that consumes the NEXT argv token. Missing one here makes the
-// translator misread the flag's value as a positional subcommand and exit 2
-// (e.g. `--install --target linux-cron`). Keep in sync with parseArg call sites.
-const AUTOPILOT_VALUE_FLAGS = new Set(['--repo', '--interval', '--target']);
-const AUTOPILOT_POSITIONAL_ALIASES: Record<string, string | null> = {
-  status: '--status',
-  install: '--install',
-  uninstall: '--uninstall',
-  start: null, // drop the positional; default behavior is daemon launch
-};
-
-export type PositionalTranslation =
-  | { ok: true; args: string[] }
-  | {
-      ok: false;
-      reason: 'unknown_subcommand' | 'multiple_subcommands';
-      message: string;
-    };
-
-export function translatePositionalSubcommands(args: string[]): PositionalTranslation {
-  const out: string[] = [];
-  let positionalSeen = false;
-  let i = 0;
-  while (i < args.length) {
-    const a = args[i];
-    if (AUTOPILOT_VALUE_FLAGS.has(a)) {
-      // Pass through the flag and its value untouched. If the value is
-      // missing at end-of-argv, fall through so the existing parseArg
-      // path can report the broken usage.
-      out.push(a);
-      if (i + 1 < args.length) {
-        out.push(args[i + 1]);
-        i += 2;
-      } else {
-        i += 1;
-      }
-      continue;
-    }
-    if (a.startsWith('-')) {
-      out.push(a);
-      i += 1;
-      continue;
-    }
-    // Positional subcommand.
-    if (positionalSeen) {
-      const known = Object.keys(AUTOPILOT_POSITIONAL_ALIASES).join(', ');
-      return {
-        ok: false,
-        reason: 'multiple_subcommands',
-        message: `Multiple subcommands given. Use only one of: ${known}.`,
-      };
-    }
-    positionalSeen = true;
-    if (a in AUTOPILOT_POSITIONAL_ALIASES) {
-      const alias = AUTOPILOT_POSITIONAL_ALIASES[a];
-      if (alias) out.push(alias);
-      i += 1;
-      continue;
-    }
-    const known = Object.keys(AUTOPILOT_POSITIONAL_ALIASES).join(', ');
-    return {
-      ok: false,
-      reason: 'unknown_subcommand',
-      message:
-        `Unknown subcommand: \`${a}\`.\n` +
-        `Allowed subcommands: ${known}.\n` +
-        `Or use the flag form: --status, --install, --uninstall.\n` +
-        `Run \`gbrain autopilot --help\` for full usage.`,
-    };
-  }
-  return { ok: true, args: out };
-}
-
 export function isPidAlive(pid: number): boolean {
   if (!Number.isFinite(pid) || pid <= 0) return false;
   try {
@@ -457,11 +363,6 @@ export async function runAutopilot(engine: BrainEngine, args: string[]) {
       '       gbrain autopilot --install [--repo <path>]\n' +
       '       gbrain autopilot --uninstall\n' +
       '       gbrain autopilot --status [--json]\n\n' +
-      'Subcommand aliases:\n' +
-      '       gbrain autopilot status     → --status\n' +
-      '       gbrain autopilot install    → --install\n' +
-      '       gbrain autopilot uninstall  → --uninstall\n' +
-      '       gbrain autopilot start      → (default daemon launch)\n\n' +
       'Self-maintaining brain daemon. Runs the full maintenance cycle\n' +
       '(lint + backlinks + sync + extract + embed + orphans) on an interval.\n\n' +
       'For a one-shot cron-triggered cycle, see `gbrain dream`.',
@@ -469,16 +370,6 @@ export async function runAutopilot(engine: BrainEngine, args: string[]) {
     return;
   }
 
-  // #1525: translate positional subcommands to their flag form BEFORE any
-  // side effect (lockfile, daemon spawn, sync dispatch). Unknown positionals
-  // fail loud here rather than silently starting the daemon.
-  const translated = translatePositionalSubcommands(args);
-  if (!translated.ok) {
-    console.error(translated.message);
-    process.exit(2);
-  }
-  args = translated.args;
-
   if (args.includes('--install')) {
     await installDaemon(engine, args);
     return;

src/commands/onboard.ts

@@ -142,12 +142,16 @@ 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).
+  // 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.
   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/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 },
+      { targetScore, maxUsd, extraRemediations: allowedExtras },
       {},
     );
 

src/core/remediation/run.ts

@@ -66,9 +66,10 @@ 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 });
+  const initialPlan = await computeRemediationPlan(engine, { targetScore, extraRemediations });
   if (initialPlan.target_unreachable) {
     hooks.onTargetUnreachable?.(targetScore, initialPlan.max_reachable_score);
     return {
@@ -87,7 +88,7 @@ export async function runRemediation(
   }
 
   const initialHealth = await engine.getHealth();
-  let recs: RemediationStep[] = computeRecommendations(initialHealth, ctx)
+  let recs: RemediationStep[] = computeRecommendations(initialHealth, ctx, extraRemediations)
     .filter((r) => r.status === 'remediable');
   if (recs.length === 0) {
     hooks.onNothingToDo?.(initialHealth.brain_score, targetScore);
@@ -305,7 +306,13 @@ export async function runRemediation(
       // steps with bumped retry suffix (D1).
       if (recs.length === 0 || stepCount >= maxJobs) break;
       const freshHealth = await engine.getHealth();
-      recs = computeRecommendations(freshHealth, ctx).filter((r) => r.status === 'remediable');
+      // 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');
     }
   };
 

src/core/remediation/types.ts

@@ -63,6 +63,16 @@ 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/autopilot-positional-subcommands.test.ts

@@ -1,197 +0,0 @@
-/**
- * Tests for translatePositionalSubcommands() — the v0.41.x #1525 fix that
- * prevents `gbrain autopilot status` from silently starting the daemon.
- *
- * IRON RULE regression guard: the exact ticket repro (`gbrain autopilot
- * status`) MUST translate to `--status`, not fall through to the default
- * daemon launch. Verified by the "ticket-exact repro" case below.
- */
-
-import { describe, test, expect } from 'bun:test';
-import { translatePositionalSubcommands } from '../src/commands/autopilot.ts';
-
-describe('translatePositionalSubcommands — known aliases', () => {
-  test('IRON RULE — `autopilot status` translates to `--status` (ticket #1525 repro)', () => {
-    const r = translatePositionalSubcommands(['status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status']);
-  });
-
-  test('`install` translates to `--install`', () => {
-    const r = translatePositionalSubcommands(['install']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install']);
-  });
-
-  test('`uninstall` translates to `--uninstall`', () => {
-    const r = translatePositionalSubcommands(['uninstall']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--uninstall']);
-  });
-
-  test('`start` drops the positional (default daemon launch)', () => {
-    const r = translatePositionalSubcommands(['start']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual([]);
-  });
-
-  test('`start --json` drops only the positional, keeps the flag', () => {
-    const r = translatePositionalSubcommands(['start', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--json']);
-  });
-});
-
-describe('translatePositionalSubcommands — flag/positional interleaving', () => {
-  test('`status --json` preserves the trailing flag', () => {
-    const r = translatePositionalSubcommands(['status', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status', '--json']);
-  });
-
-  test('`--json status` preserves the leading flag', () => {
-    const r = translatePositionalSubcommands(['--json', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--json', '--status']);
-  });
-
-  test('`--repo /foo status` does not mis-classify the path as positional', () => {
-    const r = translatePositionalSubcommands(['--repo', '/foo', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo', '/foo', '--status']);
-  });
-
-  test('`--interval 300 install` does not mis-classify the number as positional', () => {
-    const r = translatePositionalSubcommands(['--interval', '300', 'install']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--interval', '300', '--install']);
-  });
-
-  test('`--install --target linux-cron` does not mis-classify the target as positional', () => {
-    // --target is installDaemon's value flag; its value must never be read
-    // as a positional subcommand (regression guard for the review fix).
-    const r = translatePositionalSubcommands(['--install', '--target', 'linux-cron']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install', '--target', 'linux-cron']);
-  });
-
-  test('`install --target macos` keeps the alias translation and the target value', () => {
-    const r = translatePositionalSubcommands(['install', '--target', 'macos']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--install', '--target', 'macos']);
-  });
-
-  test('value-flag at end of argv with missing value passes through (so parseArg can report it)', () => {
-    const r = translatePositionalSubcommands(['--repo']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo']);
-  });
-
-  test('value-flag whose value looks like an alias is NOT translated', () => {
-    // `--repo status` means "use repo path 'status'", not "show status".
-    // Translator must not destructure the value of --repo.
-    const r = translatePositionalSubcommands(['--repo', 'status']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--repo', 'status']);
-  });
-});
-
-describe('translatePositionalSubcommands — pass-through cases', () => {
-  test('empty args returns empty args', () => {
-    const r = translatePositionalSubcommands([]);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual([]);
-  });
-
-  test('flag-only invocation passes through unchanged', () => {
-    const r = translatePositionalSubcommands(['--status', '--json']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['--status', '--json']);
-  });
-
-  test('short flag `-h` passes through unchanged', () => {
-    const r = translatePositionalSubcommands(['-h']);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(['-h']);
-  });
-
-  test('all known bare flags pass through unchanged', () => {
-    const flags = ['--help', '--install', '--uninstall', '--status', '--json', '--inline', '--no-worker'];
-    const r = translatePositionalSubcommands(flags);
-    expect(r.ok).toBe(true);
-    if (r.ok) expect(r.args).toEqual(flags);
-  });
-});
-
-describe('translatePositionalSubcommands — rejection of unknown positionals', () => {
-  test('unknown positional `foo` fails with reason=unknown_subcommand + structured message', () => {
-    const r = translatePositionalSubcommands(['foo']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `foo`');
-      expect(r.message).toContain('status');
-      expect(r.message).toContain('install');
-      expect(r.message).toContain('uninstall');
-      expect(r.message).toContain('--help');
-    }
-  });
-
-  test('unknown positional `stop` fails with reason=unknown_subcommand (NOT silently aliased)', () => {
-    // Stop is mentioned in the ticket but deliberately NOT aliased in this
-    // PR — stopping a running daemon is a new behavior, not just an alias.
-    // Until that feature lands separately, `stop` must fail loud rather
-    // than starting the daemon (the bug we're fixing).
-    const r = translatePositionalSubcommands(['stop']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `stop`');
-    }
-  });
-
-  test('unknown positional `status-detail` (close-but-not-matching) fails', () => {
-    const r = translatePositionalSubcommands(['status-detail']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('Unknown subcommand: `status-detail`');
-    }
-  });
-
-  test('multiple positionals fail with reason=multiple_subcommands (`start install`)', () => {
-    const r = translatePositionalSubcommands(['start', 'install']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('multiple_subcommands');
-      expect(r.message).toContain('Multiple subcommands');
-    }
-  });
-
-  test('multiple positionals fail even when both are known aliases (`status install`)', () => {
-    const r = translatePositionalSubcommands(['status', 'install']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('multiple_subcommands');
-      expect(r.message).toContain('Multiple subcommands');
-    }
-  });
-
-  test('known-then-unknown rejects with multiple_subcommands (first-positional-wins)', () => {
-    // First positional is known, second is not. Rejection comes from the
-    // multiple-positional rule, which fires before the unknown check; the
-    // intent is "only one subcommand allowed."
-    const r = translatePositionalSubcommands(['status', 'garbage']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) expect(r.reason).toBe('multiple_subcommands');
-  });
-
-  test('unknown-then-known rejects on the unknown (unknown fires before second-positional check)', () => {
-    const r = translatePositionalSubcommands(['garbage', 'status']);
-    expect(r.ok).toBe(false);
-    if (!r.ok) {
-      expect(r.reason).toBe('unknown_subcommand');
-      expect(r.message).toContain('garbage');
-    }
-  });
-});

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

@@ -0,0 +1,108 @@
+// 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',
+    ]);
+  });
+});