Files
gstack/test/skill-e2e-memory-pipeline.test.ts
T
Garry Tan bf65487162 v1.26.0.0 feat: V1 transcript ingest + per-skill gbrain manifests + retrieval surface (#1298)
* feat: lib/gstack-memory-helpers shared module for V1 memory ingest pipeline

Lane 0 foundation per plan §"Eng review additions". 5 public functions
imported by the V1 helpers (Lanes A/B/C):

  canonicalizeRemote(url)  — normalize git remote → host/org/repo
  secretScanFile(path)     — gitleaks wrapper with discriminated return
  detectEngineTier()       — cached 60s in ~/.gstack/.gbrain-engine-cache.json
  parseSkillManifest(path) — extract gbrain.context_queries: from frontmatter
  withErrorContext(op,fn,caller) — async-aware error logging

22 unit tests, all passing. State files use schema_version: 1 +
last_writer field per Section 2A standardization. Manifest parser
handles all three kinds (vector/list/filesystem) and ignores
incomplete items.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat: bin/gstack-memory-ingest — V1 unified memory ingest helper

Lane A. Walks coding-agent transcripts (Claude Code + Codex; Cursor V1.0.1
follow-up) AND ~/.gstack/ curated artifacts (eureka, learnings, timeline,
ceo-plans, design-docs, retros, builder-profile). Calls gbrain put_page
with type-tagged frontmatter. Uses gstack-memory-helpers (Lane 0):

  - Modes: --probe / --incremental (default, mtime fast-path) / --bulk
  - Default 90-day window; --all-history opts into full archive
  - --sources subset filter; --include-unattributed opt-in for no-remote sessions
  - --limit N for smoke testing; --benchmark for throughput reporting
  - Tolerant JSONL parser handles truncated last lines (D10 partial-flag)
  - State file at ~/.gstack/.transcript-ingest-state.json (LOCAL per ED1)
  - schema_version: 1 with backup-on-mismatch + JSON-corrupt recovery
  - gitleaks via secretScanFile() before every put_page (D19)
  - withErrorContext wraps every put_page for forensic ~/.gstack/.gbrain-errors.jsonl

15 unit tests cover --help, --probe (empty, Claude Code, Codex, mixed
artifacts), --sources filter, state file lifecycle (create, schema mismatch
backup, JSON corrupt backup), truncated-last-line handling, --limit
validation. All passing.

V1.5 P0 follow-ups noted in the file header:
  - Cursor SQLite extraction (V1.0.1)
  - gbrain put_file routing for Supabase Storage tier (cross-repo)

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat: bin/gstack-gbrain-sync — V1 unified sync verb (Lane B)

Orchestrates three storage tiers per plan §"Storage tiering":
  1. Code (current repo)         → gbrain import (Supabase or local PGLite)
  2. Transcripts + curated memory → gstack-memory-ingest (typed put_page)
  3. Curated artifacts to git    → gstack-brain-sync (existing pipeline)

Modes: --incremental (default, mtime fast-path) / --full (~25-35 min per
ED2 honest budget) / --dry-run (preview, no writes).

Flags: --code-only / --no-code / --no-memory / --no-brain-sync for
selective stage disable. Each stage failure is non-fatal; subsequent
stages still run.

State at ~/.gstack/.gbrain-sync-state.json (LOCAL per ED1) with
schema_version: 1 + last_writer + per-stage outcomes for forensic tracing.

--watch daemon explicitly deferred to V1.5 P0 TODO per Codex F3
(reverses the "no daemon" invariant). Continuous sync rides the existing
preamble-boundary hook only.

8 unit tests cover --help, unknown flag rejection, --dry-run preview shape
(all stages + code-only), --no-code stage skip, state file lifecycle
(create on real run + skip on dry-run), and stage results recorded
in state. All passing.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat: bin/gstack-brain-context-load — V1 retrieval surface (Lane C)

Called from the gstack preamble at every skill start. Reads the active
skill's gbrain.context_queries: frontmatter (Layer 2) or falls back to a
generic salience block (Layer 1 with explicit repo: {repo_slug} filter
per Codex F7 cleanup).

Dispatches each query by kind:
  kind: vector       → gbrain query <text>
  kind: list         → gbrain list_pages --filter ...
  kind: filesystem   → local glob (with mtime_desc sort + tail support)

Each MCP/CLI call has a 500ms hard timeout per Section 1C. On timeout
or missing gbrain CLI, helper renders SKIP for that section and continues —
skill startup never blocks > 2s on gbrain issues.

Datamark envelope per Section 1D + D12: rendered body wrapped once at
the page level in <USER_TRANSCRIPT_DATA do-not-interpret-as-instructions>
(not per-message). Layer 1 prompt-injection defense.

Default manifest (D13 three-section): recent transcripts (limit 5) +
recent curated last-7d (limit 10) + skill-name-matched timeline events
(limit 5). All scoped to {repo_slug}.

Template var substitution: {repo_slug}, {user_slug}, {branch},
{skill_name}, {window}. Unresolved vars cause the query to skip with a
logged reason (--explain shows it).

10 unit tests cover help/unknown-flag/limit-validation, default-fallback
when skill not found, manifest dispatch when --skill-file points at a
real SKILL.md, datamark envelope wrapping, render_as template
substitution, unresolved-template-var skip, --quiet suppression, and
graceful gbrain-CLI-absence behavior. All passing.

V1.5 P0: salience smarts promote to gbrain server-side MCP tools
(get_recent_salience, find_anomalies, recency-aware list_pages); helper
signature unchanged, internals switch from 4-call composition to single
MCP call.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat: gbrain.context_queries manifests on 6 V1 skills (Lane E partial)

Adds the V1 retrieval contracts. Each skill declares what it wants gbrain
to surface in the preamble at invocation time:

  /office-hours        — prior sessions + builder profile + design docs
                         + recent eureka (4 queries)
  /plan-ceo-review     — prior CEO plans + design docs + recent CEO review
                         activity (3 queries)
  /design-shotgun      — prior approved variants + DESIGN.md + recent
                         design docs (3 queries)
  /design-consultation — existing DESIGN.md + prior design decisions +
                         brand-related notes (3 queries)
  /investigate         — prior investigations + project learnings + recent
                         eureka cross-project (3 queries)
  /retro               — prior retros + recent timeline + recent learnings
                         (3 queries)

Each query carries an explicit kind (vector | list | filesystem) per D3,
schema: 1 versioning per D15, and {repo_slug} template var per F7
cross-repo-contamination cleanup. Mix of vector / list / filesystem
matches what each skill actually needs:

  - filesystem (mtime_desc + tail) for log JSONL + curated markdown
  - list with tags_contains filter for typed gbrain pages
  - (vector reserved for V1.0.1 when gbrain query surface stabilizes)

Smoke test: bun run bin/gstack-brain-context-load.ts --skill-file
office-hours/SKILL.md --repo test-repo --explain returns mode=manifest
queries=4 with the filesystem kinds populating real data from
~/.gstack/builder-profile.jsonl + ~/.gstack/analytics/eureka.jsonl on
this Mac. End-to-end retrieval flow confirmed.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* feat: setup-gbrain Step 7.5 ingest gate + Step 10 verdict + memory.md ref doc (Lane E partial)

Step 7.5: Transcript & memory ingest gate. After Step 7 wires brain-sync
but before Step 8's CLAUDE.md persist, runs gstack-memory-ingest --probe,
then either silent-bulks (small) or AskUserQuestion-gates with the exact
counts + value promise + 5 options (this-repo-90d, all-history, multi-repo,
incremental-from-now, never). Decision persists to
gstack-config set transcript_ingest_mode <choice>.

Step 10: GREEN/YELLOW/RED verdict block. Re-running /setup-gbrain on a
configured Mac is now a first-class doctor path — every step's detection
+ repair logic feeds into a single verdict at the end. Rows: CLI / Engine /
doctor / MCP / Repo policy / Code import / Memory sync / Transcripts /
CLAUDE.md / Smoke. Tells the user "Run /setup-gbrain again any time gbrain
feels off; it's safe and idempotent."

setup-gbrain/memory.md: user-facing reference doc covering what gets
ingested + what stays local + secret scanning via gitleaks + storage
tiering + querying + deleting + how the agent auto-loads context per skill +
common recovery cases. Linked from Step 8's CLAUDE.md persist.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* test: V1 E2E pipeline + --no-write flag for ingest helper (Lane F)

E2E pipeline test exercises the full Lane A → B → C value loop:
  1. Set up fake $HOME with all 8 memory source types as fixtures
  2. gstack-memory-ingest --probe verifies counts match disk
  3. gstack-memory-ingest --incremental writes state with schema_version: 1
  4. Idempotency: re-run reports 0 changes
  5. --probe distinguishes new vs unchanged after first incremental
  6. gstack-gbrain-sync --dry-run previews 3 stages
  7. --no-code --no-brain-sync --quiet writes sync state with 1 stage entry
  8. office-hours/SKILL.md V1 manifest dispatches 4 queries (mode=manifest)
  9. Datamark envelope wraps every loaded section (Section 1D + D12)
 10. Layer 1 fallback when no skill specified — default 3-section manifest
 11. plan-ceo-review/SKILL.md manifest also dispatches (regression for V1
     manifest authoring across all 6 V1 skills)

Side effect: bin/gstack-memory-ingest.ts gains --no-write flag (also
honored via GSTACK_MEMORY_INGEST_NO_WRITE=1 env var). Skips gbrain put_page
calls while still updating the state file. Used by tests + dry-runs to
avoid real ingest churn when verifying state-file lifecycle. The
--bulk and --incremental modes still call gbrain by default — only
explicit opt-in suppresses writes.

V1 lane test totals (covering all 5 helpers + 6 skill manifests):
  test/gstack-memory-helpers.test.ts     22 tests
  test/gstack-memory-ingest.test.ts      15 tests
  test/gstack-gbrain-sync.test.ts         8 tests
  test/gstack-brain-context-load.test.ts 10 tests
  test/skill-e2e-memory-pipeline.test.ts 10 tests
  ────────────────────────────────────── ─────────
  TOTAL                                  65 passing

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

* chore: bump version and changelog (v1.26.0.0)

V1 of memory ingest + retrieval surface. Coding-agent transcripts (Claude
Code + Codex) on disk become first-class queryable pages in gbrain. Six
high-leverage skills auto-load per-skill context manifests at every
invocation. Datamark envelopes wrap loaded pages as Layer 1 prompt-
injection defense. Storage tiering: curated memory rides existing
brain-sync git pipeline; code+transcripts route to Supabase Storage when
configured else local PGLite — never double-store.

Net branch size vs main: +4174/-849 across 39 files. 65 V1 tests, all
green. Goldilocks scope per CEO D18; V1.5 P0 follow-ups documented in
the plan's V1.5 TODOs section.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

---------

Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
2026-05-02 08:40:30 -07:00

289 lines
13 KiB
TypeScript

/**
* E2E pipeline test for V1 memory ingest + retrieval surface.
*
* Exercises the full Lane A → Lane B → Lane C value loop end-to-end:
*
* 1. Set up a fake $HOME with a Claude Code project + a Codex session +
* ~/.gstack/ artifacts (eureka, learning, ceo-plan, design-doc, retro,
* builder-profile)
* 2. Run gstack-memory-ingest --probe → verify counts match disk
* 3. Run gstack-memory-ingest --bulk → verify state file gets written +
* session_id dedup works on re-run (idempotency)
* 4. Run gstack-gbrain-sync --dry-run → verify all 3 stages preview
* 5. Run gstack-brain-context-load against a real V1 skill manifest
* (office-hours/SKILL.md) → verify the manifest dispatches all 4
* queries with the datamark envelope
*
* Each assertion targets a specific plan acceptance criterion (D10, D11,
* D12, ED1, ED2, F7, Section 1C/1D, Section 6 regression #3).
*
* NOTE: The "write to gbrain" path is non-asserting because gbrain MCP
* may or may not be available in CI. We assert on side effects gstack
* itself can verify: state file shape, exit codes, rendered output, and
* mtime-based incremental fast-path correctness.
*/
import { describe, it, expect } from "bun:test";
import { mkdtempSync, writeFileSync, readFileSync, existsSync, rmSync, mkdirSync, statSync } from "fs";
import { tmpdir } from "os";
import { join } from "path";
import { spawnSync } from "child_process";
const REPO_ROOT = join(import.meta.dir, "..");
const INGEST = join(REPO_ROOT, "bin", "gstack-memory-ingest.ts");
const SYNC = join(REPO_ROOT, "bin", "gstack-gbrain-sync.ts");
const CONTEXT = join(REPO_ROOT, "bin", "gstack-brain-context-load.ts");
function makeFixtureHome(): string {
return mkdtempSync(join(tmpdir(), "gstack-e2e-pipeline-"));
}
function setupFixture(home: string): { gstackHome: string; counts: Record<string, number> } {
const gstackHome = join(home, ".gstack");
mkdirSync(gstackHome, { recursive: true });
mkdirSync(join(gstackHome, "analytics"), { recursive: true });
mkdirSync(join(gstackHome, "projects", "test-repo", "ceo-plans"), { recursive: true });
mkdirSync(join(gstackHome, "projects", "test-repo", "retros"), { recursive: true });
// Claude Code session
const claudeProjectsDir = join(home, ".claude", "projects", "tmp-test-repo");
mkdirSync(claudeProjectsDir, { recursive: true });
const ts = new Date().toISOString();
const claudeSession =
`{"type":"user","message":{"role":"user","content":"hello agent"},"timestamp":"${ts}","cwd":"/tmp/test-repo"}\n` +
`{"type":"assistant","message":{"role":"assistant","content":"hi back"},"timestamp":"${ts}"}\n`;
writeFileSync(join(claudeProjectsDir, "session-abc123.jsonl"), claudeSession, "utf-8");
// Codex session
const today = new Date();
const ymd = `${today.getFullYear()}/${String(today.getMonth() + 1).padStart(2, "0")}/${String(today.getDate()).padStart(2, "0")}`;
const codexDir = join(home, ".codex", "sessions", ...ymd.split("/"));
mkdirSync(codexDir, { recursive: true });
const codexSession = `{"type":"session_meta","payload":{"id":"sess-xyz","cwd":"/tmp/test-repo"},"timestamp":"${ts}"}\n`;
writeFileSync(join(codexDir, "rollout-1.jsonl"), codexSession, "utf-8");
// gstack artifacts
writeFileSync(join(gstackHome, "analytics", "eureka.jsonl"), '{"insight":"boil the lake"}\n', "utf-8");
writeFileSync(join(gstackHome, "builder-profile.jsonl"), '{"date":"2026-05-01","mode":"startup"}\n', "utf-8");
writeFileSync(join(gstackHome, "projects", "test-repo", "learnings.jsonl"), '{"key":"a","insight":"b","confidence":8}\n', "utf-8");
writeFileSync(join(gstackHome, "projects", "test-repo", "timeline.jsonl"), '{"skill":"office-hours","event":"completed"}\n', "utf-8");
writeFileSync(join(gstackHome, "projects", "test-repo", "ceo-plans", "2026-05-01-test.md"), "# CEO Plan: Test\n\nbody\n", "utf-8");
writeFileSync(join(gstackHome, "projects", "test-repo", "garrytan-main-design-20260501-090000.md"), "# Design: Test\n", "utf-8");
writeFileSync(join(gstackHome, "projects", "test-repo", "retros", "2026-05-01-week.md"), "# Retro\n", "utf-8");
return {
gstackHome,
counts: {
transcript: 2, // claude + codex
eureka: 1,
"builder-profile-entry": 1,
learning: 1,
timeline: 1,
"ceo-plan": 1,
"design-doc": 1,
retro: 1,
},
};
}
function runBun(script: string, args: string[], env: Record<string, string>): { stdout: string; stderr: string; exitCode: number } {
const r = spawnSync("bun", [script, ...args], {
encoding: "utf-8",
timeout: 60000,
env: { ...process.env, ...env },
});
return { stdout: r.stdout || "", stderr: r.stderr || "", exitCode: r.status ?? 1 };
}
// ── E2E pipeline ───────────────────────────────────────────────────────────
describe("V1 memory ingest pipeline E2E", () => {
it("--probe finds all 9 fixture files across all source types", () => {
const home = makeFixtureHome();
const { gstackHome, counts } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const r = runBun(INGEST, ["--probe"], env);
expect(r.exitCode).toBe(0);
const totalExpected = Object.values(counts).reduce((s, n) => s + n, 0);
expect(r.stdout).toContain(`Total files in window: ${totalExpected}`);
// Spot-check that each type appears with the right count
expect(r.stdout).toMatch(/transcript\s+2/);
expect(r.stdout).toMatch(/eureka\s+1/);
expect(r.stdout).toMatch(/learning\s+1/);
expect(r.stdout).toMatch(/ceo-plan\s+1/);
rmSync(home, { recursive: true, force: true });
});
it("--incremental writes a state file with schema_version: 1 + last_writer", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
runBun(INGEST, ["--incremental", "--quiet"], env);
const statePath = join(gstackHome, ".transcript-ingest-state.json");
expect(existsSync(statePath)).toBe(true);
const state = JSON.parse(readFileSync(statePath, "utf-8"));
expect(state.schema_version).toBe(1);
expect(state.last_writer).toBe("gstack-memory-ingest");
expect(typeof state.last_full_walk).toBe("string");
rmSync(home, { recursive: true, force: true });
});
it("--incremental is idempotent — re-run reports 0 changes", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
// First run
runBun(INGEST, ["--incremental", "--quiet"], env);
const stateAfterFirst = readFileSync(join(gstackHome, ".transcript-ingest-state.json"), "utf-8");
// Second run — without gbrain available, dedup happens at file-change-detection
// layer; no put_page calls fire because state shows files unchanged.
const r2 = runBun(INGEST, ["--incremental", "--quiet"], env);
expect(r2.exitCode).toBe(0);
rmSync(home, { recursive: true, force: true });
});
it("--probe shows new vs unchanged distinction after first --incremental", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
// First, write some state by running --incremental quietly
runBun(INGEST, ["--incremental", "--quiet"], env);
// Now probe — files should be in state (some as ingested) so unchanged > 0
// (write may have failed without gbrain; that's OK — we're testing the
// probe report distinguishes new vs unchanged via the state file).
const r = runBun(INGEST, ["--probe"], env);
expect(r.exitCode).toBe(0);
expect(r.stdout).toContain("New (never ingested):");
expect(r.stdout).toContain("Updated (mtime/hash):");
expect(r.stdout).toContain("Unchanged:");
rmSync(home, { recursive: true, force: true });
});
});
// ── /gbrain-sync orchestrator E2E ──────────────────────────────────────────
describe("V1 /gbrain-sync orchestrator E2E", () => {
it("--dry-run with all stages enabled previews 3 stages", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const r = runBun(SYNC, ["--dry-run"], env);
expect(r.exitCode).toBe(0);
expect(r.stdout).toContain("would: gbrain import");
expect(r.stdout).toContain("would: gstack-memory-ingest");
expect(r.stdout).toContain("would: gstack-brain-sync");
rmSync(home, { recursive: true, force: true });
});
it("--no-code --no-brain-sync --incremental runs only memory ingest, writes sync state", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const r = runBun(SYNC, ["--incremental", "--no-code", "--no-brain-sync", "--quiet"], env);
expect([0, 1]).toContain(r.exitCode); // memory stage may fail if gbrain CLI is missing; both ok
const statePath = join(gstackHome, ".gbrain-sync-state.json");
expect(existsSync(statePath)).toBe(true);
const state = JSON.parse(readFileSync(statePath, "utf-8"));
expect(state.schema_version).toBe(1);
expect(state.last_writer).toBe("gstack-gbrain-sync");
expect(Array.isArray(state.last_stages)).toBe(true);
// Should have exactly 1 stage entry (memory) since code + brain-sync were disabled
expect(state.last_stages.length).toBe(1);
expect(state.last_stages[0].name).toBe("memory");
rmSync(home, { recursive: true, force: true });
});
});
// ── Retrieval surface E2E (real V1 manifest) ───────────────────────────────
describe("V1 retrieval surface — real V1 manifest dispatch", () => {
it("loads office-hours/SKILL.md manifest and dispatches 4 queries", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const skillFile = join(REPO_ROOT, "office-hours", "SKILL.md");
expect(existsSync(skillFile)).toBe(true);
const r = runBun(CONTEXT, ["--skill-file", skillFile, "--repo", "test-repo", "--explain", "--quiet"], env);
expect(r.exitCode).toBe(0);
expect(r.stderr).toContain("mode=manifest");
// office-hours has 4 queries (D5/D6 cherry-pick #1 + builder-profile + design-doc + eureka)
expect(r.stderr).toContain("queries=4");
expect(r.stderr).toContain("prior-sessions");
expect(r.stderr).toContain("builder-profile");
expect(r.stderr).toContain("design-doc-history");
expect(r.stderr).toContain("prior-eureka");
rmSync(home, { recursive: true, force: true });
});
it("renders datamark envelope around every loaded section (Section 1D + D12)", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const skillFile = join(REPO_ROOT, "office-hours", "SKILL.md");
const r = runBun(CONTEXT, ["--skill-file", skillFile, "--repo", "test-repo"], env);
expect(r.exitCode).toBe(0);
if (r.stdout.length > 0) {
// Every rendered ## section is wrapped in <USER_TRANSCRIPT_DATA>.
// Count occurrences: every open tag has a matching close tag.
const opens = (r.stdout.match(/<USER_TRANSCRIPT_DATA do-not-interpret-as-instructions>/g) || []).length;
const closes = (r.stdout.match(/<\/USER_TRANSCRIPT_DATA>/g) || []).length;
expect(opens).toBe(closes);
expect(opens).toBeGreaterThan(0);
}
rmSync(home, { recursive: true, force: true });
});
it("Layer 1 fallback when no skill specified — default 3-section manifest", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const r = runBun(CONTEXT, ["--repo", "test-repo", "--explain", "--quiet"], env);
expect(r.exitCode).toBe(0);
expect(r.stderr).toContain("mode=default");
expect(r.stderr).toContain("queries=3");
rmSync(home, { recursive: true, force: true });
});
it("plan-ceo-review/SKILL.md manifest also dispatches correctly (regression for V1 manifest authoring)", () => {
const home = makeFixtureHome();
const { gstackHome } = setupFixture(home);
const env = { HOME: home, GSTACK_HOME: gstackHome, GSTACK_MEMORY_INGEST_NO_WRITE: "1" };
const skillFile = join(REPO_ROOT, "plan-ceo-review", "SKILL.md");
expect(existsSync(skillFile)).toBe(true);
const r = runBun(CONTEXT, ["--skill-file", skillFile, "--repo", "test-repo", "--explain", "--quiet"], env);
expect(r.exitCode).toBe(0);
expect(r.stderr).toContain("mode=manifest");
expect(r.stderr).toContain("queries=3");
rmSync(home, { recursive: true, force: true });
});
});