fix(setup-gbrain): invoke gstack-memory-ingest/gstack-gbrain-sync via bun run + .ts

/setup-gbrain's transcript-ingest steps told the agent to run
bin/gstack-memory-ingest and bin/gstack-gbrain-sync by BARE name. Neither
exists — only the .ts files ship (mode 644, no bin alias) — so the agent
dutifully reported 'script missing at install root' and the ingest/full-
sync steps dead-ended on every host (hit live under Codex; the Claude
render carries the same text).

All four template sites (probe, silent-bulk, post-answer full sync, the
preamble-hook incremental mention) and the four memory.md reference-doc
sites now use the repo's established form: 'bun run <path>/gstack-memory-
ingest.ts …' / 'bun run <path>/gstack-gbrain-sync.ts …' — matching what
sync-gbrain already does. Generated SKILL.md regenerated from the template
in the same commit.

Re-derived from PR #2409 by @SomSamantray per the wave's screening rule
(the PR edited the generated SKILL.md directly; the generated file must
come from gen:skill-docs). The contributor's structural test rides along
as-is: bare-invocation regexes with negative .ts lookahead and backslash-
continuation coverage pin every site, so the drift can't return. The
referenced-paths ratchet in test/setup-claude-skill-assets.test.ts drops
its two #2250 known-broken entries — the class-2 assertion now guards
these paths again.

Verified against #2250's site list (template lines 690/735/784-area, all
covered) plus a fresh grep: zero bare invocations remain in the template
or memory.md; the one prose mention ('gstack-memory-ingest now persists…')
is not an invocation and stays.

Fixes #2250
Fixes #2393

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

setup-gbrain/SKILL.md

@@ -1503,7 +1503,7 @@ curated `~/.gstack/` artifacts into gbrain so the retrieval surface
 
 Run the probe to size the operation:
 ```bash
-~/.claude/skills/gstack/bin/gstack-memory-ingest --probe
+bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --probe
 ```
 
 Read the output. If `Total files in window: 0`, skip — there's nothing
@@ -1511,7 +1511,7 @@ to ingest. Set `gstack-config set transcript_ingest_mode incremental`
 silently and continue to Step 8.
 
 If `New (never ingested)` is < 200 AND total bytes are < 100MB: silent
-bulk via `gstack-memory-ingest --bulk --quiet`. Set
+bulk via `bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --bulk --quiet`. Set
 `transcript_ingest_mode=incremental` and continue.
 
 Otherwise (the "many transcripts on disk" path): AskUserQuestion with
@@ -1548,14 +1548,14 @@ Options:
 After answer:
 ```bash
 ~/.claude/skills/gstack/bin/gstack-config set transcript_ingest_mode <choice>
-~/.claude/skills/gstack/bin/gstack-gbrain-sync --full --no-brain-sync
+bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --full --no-brain-sync
 ```
 (`--no-brain-sync` because Step 7 already wired that path; this just
 runs the code import + memory ingest stages. Brain-sync will run on the
 next preamble hook.)
 
 If A/D/E, ingest is incremental from this point on; preamble-boundary
-hook runs `gstack-gbrain-sync --incremental --quiet` on every skill
+hook runs `bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --incremental --quiet` on every skill
 start (cheap mtime fast-path).
 
 Reference doc for users: `setup-gbrain/memory.md` (linked from CLAUDE.md

setup-gbrain/SKILL.md.tmpl

@@ -736,7 +736,7 @@ curated `~/.gstack/` artifacts into gbrain so the retrieval surface
 
 Run the probe to size the operation:
 ```bash
-~/.claude/skills/gstack/bin/gstack-memory-ingest --probe
+bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --probe
 ```
 
 Read the output. If `Total files in window: 0`, skip — there's nothing
@@ -744,7 +744,7 @@ to ingest. Set `gstack-config set transcript_ingest_mode incremental`
 silently and continue to Step 8.
 
 If `New (never ingested)` is < 200 AND total bytes are < 100MB: silent
-bulk via `gstack-memory-ingest --bulk --quiet`. Set
+bulk via `bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --bulk --quiet`. Set
 `transcript_ingest_mode=incremental` and continue.
 
 Otherwise (the "many transcripts on disk" path): AskUserQuestion with
@@ -781,14 +781,14 @@ Options:
 After answer:
 ```bash
 ~/.claude/skills/gstack/bin/gstack-config set transcript_ingest_mode <choice>
-~/.claude/skills/gstack/bin/gstack-gbrain-sync --full --no-brain-sync
+bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --full --no-brain-sync
 ```
 (`--no-brain-sync` because Step 7 already wired that path; this just
 runs the code import + memory ingest stages. Brain-sync will run on the
 next preamble hook.)
 
 If A/D/E, ingest is incremental from this point on; preamble-boundary
-hook runs `gstack-gbrain-sync --incremental --quiet` on every skill
+hook runs `bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --incremental --quiet` on every skill
 start (cheap mtime fast-path).
 
 Reference doc for users: `setup-gbrain/memory.md` (linked from CLAUDE.md

setup-gbrain/memory.md

@@ -47,9 +47,9 @@ v1.33.0.0 — off by default. To re-enable it (adds ~4-8 min to cold runs
 on a large transcript corpus), use either:
 
 ```bash
-gstack-memory-ingest --bulk --scan-secrets
+bun run bin/gstack-memory-ingest.ts --bulk --scan-secrets
 # or
-GSTACK_MEMORY_INGEST_SCAN_SECRETS=1 gstack-memory-ingest --bulk
+GSTACK_MEMORY_INGEST_SCAN_SECRETS=1 bun run bin/gstack-memory-ingest.ts --bulk
 ```
 
 When enabled, gitleaks covers:
@@ -155,7 +155,7 @@ verdict block. If a row is RED, the row tells you what to do.
 Common cases:
 
 - **Salience block is empty** — your transcripts may not be ingested
-  yet. Run `gstack-gbrain-sync --full` to do a full pass.
+  yet. Run `bun run bin/gstack-gbrain-sync.ts --full` to do a full pass.
 
 - **"gbrain CLI missing" in the preamble output** — gbrain isn't on
   your PATH. Run `/setup-gbrain` to install/wire it.
@@ -166,7 +166,7 @@ Common cases:
   --pglite && gbrain import <brain-remote-clone-dir>`.
 
 - **A page has stale or wrong content** — `gbrain delete_page <slug>`,
-  then re-run `gstack-gbrain-sync --incremental` to re-ingest from
+  then re-run `bun run bin/gstack-gbrain-sync.ts --incremental` to re-ingest from
   source if the source file is still on disk and unchanged.
 
 ## Privacy + audit

test/setup-claude-skill-assets.test.ts

@@ -43,11 +43,7 @@ const BUILT_ARTIFACT_ALLOWLIST = [
  * do not add entries without an issue + a scheduled fix.
  */
 const KNOWN_BROKEN_CLASS2: Record<string, string> = {
-  // #2250: setup-gbrain's docs call both scripts by bare name; only the .ts
+  // (empty — #2250's bare bin names were the last entries; keep it that way)
-  // files exist. Fixed by the wave's c24 (PR #2409 re-derive) — remove these
-  // entries in that commit.
-  'bin/gstack-memory-ingest': '#2250 — fixed by setup-gbrain .ts invocation-path commit',
-  'bin/gstack-gbrain-sync': '#2250 — fixed by setup-gbrain .ts invocation-path commit',
 };
 
 /** Extract a named shell function body (through its closing brace) from setup. */

test/setup-gbrain-bin-invocation-paths.test.ts

@@ -0,0 +1,133 @@
+// setup-gbrain bin invocation path lint.
+//
+// Pins the correct bun-run + .ts invocation form for gstack-memory-ingest
+// and gstack-gbrain-sync wherever setup-gbrain's docs instruct the agent
+// to run them. Regression coverage for #2393 / #2250: both scripts are
+// .ts files with no package.json bin alias stripping the extension, so a
+// bare name (no `bun run` prefix, no `.ts` suffix) fails with "No such
+// file or directory" the moment an agent follows the doc literally.
+//
+// Why a structural test instead of a full Agent SDK E2E:
+//   - The failure is entirely in the prose an agent reads, not in
+//     runtime behavior a service test could exercise. A grep-based
+//     regression on the template/reference-doc text is fast (<200ms),
+//     free, and catches the same drift a full E2E would, without the
+//     token cost. Same rationale as test/setup-gbrain-path4-structure.test.ts.
+//   - The correct invocation form and the stale one differ only by
+//     `bun run ` + `.ts`, right next to each other in the same files —
+//     exactly the kind of drift a cheap structural check exists to catch,
+//     matching this repo's convention (e.g. test/memory-ingest-no-put_page.test.ts
+//     pinning fix #1346).
+
+import { describe, test, expect } from 'bun:test';
+import * as fs from 'fs';
+import * as path from 'path';
+
+const ROOT = path.resolve(import.meta.dir, '..');
+const TMPL = path.join(ROOT, 'setup-gbrain', 'SKILL.md.tmpl');
+const MEMORY_DOC = path.join(ROOT, 'setup-gbrain', 'memory.md');
+
+const tmpl = fs.readFileSync(TMPL, 'utf-8');
+const memoryDoc = fs.readFileSync(MEMORY_DOC, 'utf-8');
+
+// A "bare invocation" is the tool name immediately followed by a flag/arg
+// with no `.ts` in between — the exact stale shape #2393/#2250 reported.
+// The negative lookahead means `gstack-memory-ingest.ts --probe` (correct)
+// does NOT match, while `gstack-memory-ingest --probe` (stale) does.
+// `(?:\s|\\\r?\n)+` also spans a backslash line-continuation between the
+// name and its flag (e.g. `gstack-memory-ingest \` newline `  --probe`),
+// a style this same template already uses for other commands (see the
+// read_secret_to_env invocation a few hundred lines up) — a plain `\s+`
+// would miss a stale invocation reintroduced in that form.
+const bareMemoryIngest = /\bgstack-memory-ingest\b(?!\.ts)(?:\s|\\\r?\n)+--/;
+const bareGbrainSync = /\bgstack-gbrain-sync\b(?!\.ts)(?:\s|\\\r?\n)+--/;
+
+describe('setup-gbrain/SKILL.md.tmpl — bin invocation paths', () => {
+  test('no bare gstack-memory-ingest invocation remains', () => {
+    expect(tmpl).not.toMatch(bareMemoryIngest);
+  });
+
+  test('no bare gstack-gbrain-sync invocation remains', () => {
+    expect(tmpl).not.toMatch(bareGbrainSync);
+  });
+
+  test('the probe step uses bun run + .ts (R1)', () => {
+    expect(tmpl).toContain(
+      'bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --probe'
+    );
+  });
+
+  test('the silent-bulk mention uses bun run + .ts (R2)', () => {
+    expect(tmpl).toContain(
+      'bun run ~/.claude/skills/gstack/bin/gstack-memory-ingest.ts --bulk --quiet'
+    );
+  });
+
+  test('the post-answer full-sync step uses bun run + .ts (R3)', () => {
+    expect(tmpl).toContain(
+      'bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --full --no-brain-sync'
+    );
+  });
+
+  test('the preamble-hook incremental-sync mention uses bun run + .ts (R4)', () => {
+    expect(tmpl).toContain(
+      'bun run ~/.claude/skills/gstack/bin/gstack-gbrain-sync.ts --incremental --quiet'
+    );
+  });
+
+  test('the neighboring gstack-config line in the post-answer block is untouched (bash script, no extension)', () => {
+    expect(tmpl).toContain(
+      '~/.claude/skills/gstack/bin/gstack-config set transcript_ingest_mode <choice>'
+    );
+  });
+
+  test('the prose-only mention naming the tool as a sentence subject is left unchanged (KTD4 — not a literal invocation)', () => {
+    expect(tmpl).toContain('gstack-memory-ingest now persists staged transcripts to');
+  });
+});
+
+describe('setup-gbrain/memory.md — bin invocation paths', () => {
+  test('no bare gstack-memory-ingest invocation remains', () => {
+    expect(memoryDoc).not.toMatch(bareMemoryIngest);
+  });
+
+  test('no bare gstack-gbrain-sync invocation remains', () => {
+    expect(memoryDoc).not.toMatch(bareGbrainSync);
+  });
+
+  test('the secret-scanning example uses bun run + .ts (R5)', () => {
+    expect(memoryDoc).toContain('bun run bin/gstack-memory-ingest.ts --bulk --scan-secrets');
+    expect(memoryDoc).toContain(
+      'GSTACK_MEMORY_INGEST_SCAN_SECRETS=1 bun run bin/gstack-memory-ingest.ts --bulk'
+    );
+  });
+
+  test('the troubleshooting full-pass mention uses bun run + .ts (R5)', () => {
+    expect(memoryDoc).toContain('Run `bun run bin/gstack-gbrain-sync.ts --full` to do a full pass.');
+  });
+
+  test('the troubleshooting incremental-reingest mention uses bun run + .ts (R5)', () => {
+    expect(memoryDoc).toContain(
+      're-run `bun run bin/gstack-gbrain-sync.ts --incremental` to re-ingest from'
+    );
+  });
+
+  test('the already-correct reference line at the top of the file is unchanged', () => {
+    expect(memoryDoc).toContain('bun run bin/gstack-memory-ingest.ts --probe` (which');
+  });
+});
+
+describe('bare-invocation regex — backslash line-continuation coverage', () => {
+  // This template writes multi-line commands with a trailing backslash
+  // continuation elsewhere (e.g. the read_secret_to_env invocation), so a
+  // stale invocation reintroduced in that same style must still be caught.
+  test('catches a bare invocation split across a backslash continuation', () => {
+    const staleContinuation = 'gstack-memory-ingest \\\n  --probe';
+    expect(staleContinuation).toMatch(bareMemoryIngest);
+  });
+
+  test('does not flag a correct invocation split across a backslash continuation', () => {
+    const fixedContinuation = 'bun run bin/gstack-gbrain-sync.ts \\\n  --incremental';
+    expect(fixedContinuation).not.toMatch(bareGbrainSync);
+  });
+});