mirror of
https://github.com/garrytan/gstack.git
synced 2026-10-31 22:14:25 +01:00
* fix(land-and-deploy): never merge over red, pending or missing CI (#2995) /land-and-deploy gated only on required checks, so a repo with none (every GitHub Free private repo, any repo without branch protection) skipped the CI wait and could merge over red CI. The verdict now comes from bin/gstack-ci-gate, a tested helper bound to the head SHA: all checks by default, cancel/unknown buckets fail, NO_CHECKS needs explicit per-head approval, and both merge commands (including --auto) run only after a PASS for the same SHA they pin with --match-head-commit. Co-authored-by: harshitgavita-07 <harshit.gavita@gmail.com> * fix(slug): give nested-group projects their own state bucket and verify checkpoint identity (#3003, #2901) gstack-slug kept only the last two remote path segments, so customer-a/product/repo and customer-b/product/repo shared one bucket and /context-restore could present another project's checkpoint. - One canonical-remote rule (bash + TS, golden vectors) used by all six resolvers: lowercased host, no userinfo/port/.git, Azure DevOps and Bitbucket Server normalized. 3+ segment remotes get <last-two>-<16 hex>; two-segment remotes (all of GitHub) are unchanged. - Versioned sticky cache (v2:) keeps cache hits git-free. - context-save stamps the credential-free remote and root; context-restore only treats this project's checkpoints as latest, reports foreign/unverified ones first, and treats a different root with the same remote (Conductor handoff) as information. - gstack-slug --adopt-legacy copies only files the user confirms. - question-preference-hook reads the same versioned cache, so never-ask preferences fire; unknown identity asks instead of auto-deciding. * fix(gbrain): brain context loads, managed sync works, memory sync stops duplicating (#2883, #2985, #2922) - brain-context-load called 'gbrain list_pages' (an MCP tool name), so every skill's brain context was 0 bytes. It now uses 'gbrain list' with gbrain's real filter flags, and prints one line when queries fail. - The code-stage sync always passes --no-pull (managed gbrain >= 0.51 refuses without it); a gbrain that rejects the flag fails with an upgrade message and is never retried without it. stderr is streamed with a bounded tail and quoted in failure summaries. - /sync-gbrain gains --sources (and GSTACK_MEMORY_INGEST_SOURCES); by default it skips types already synced through a federated curated source, so curated pages stop being re-duplicated. Co-authored-by: Vitor Cepeda Lopes <vitor@vitorcepedalopes.com> Co-authored-by: harshitgavita-07 <harshit.gavita@gmail.com> * fix(codex,cso): clean Codex reviews stop failing; semgrep and trivy SARIF results import (#2769, #3011) - Codex review gate is PASS / FAIL / UNVERIFIED. Untagged output (what a clean 'codex review --base' prints) is UNVERIFIED, not FAIL, and never a false PASS; P2/P3-only output passes. - import-sarif resolves semgrep's undeclared %SRCROOT% and trivy's ROOTPATH (including git-URL targets) through explicit, captured mappings; declared bases must sit inside the source root or the run's checkout, URIs are percent-decoded, and rejected results are counted with rule, base and path instead of silently dropped. * fix: version strings are not IPs, failed update checks are not 'up to date', private security reporting (#2856, #2786, #2984) - The pre-push redact hook exempts a root VERSION file that is only a version string and the version token in CHANGELOG release headings; the engine allowlist is untouched. - gstack-update-check caches an unreadable remote as CHECK_FAILED with a short TTL instead of UP_TO_DATE; /gstack-upgrade never reports 'latest' after a failed check. - SECURITY.md points reporters at GitHub private vulnerability reporting. - pr-title-sync (pull_request_target, write token) pins actions/checkout to a reviewed SHA. * fix(autoplan): publication guard accepts real Claude Code journals (#3007, #3009, #2968, #2977, #2986, #3006) The guard required a session journal to start with a user message, but Claude Code starts it with SessionStart hook attachments (and a forked compacted session starts with a compact_boundary), so every phase was denied for anyone with a SessionStart hook. Windows path spellings (C:/, /c/, drive case) were rejected too. - Reader accepts the captured startup, compact, clear and fork shapes (claude 2.1.284 fixture), computes root uniqueness over every candidate, and returns typed reason codes instead of one string. - Hard denials (competing root, foreign cwd, sidechain, agent, cycle, malformed, identity) say what happened and how to run the reviews by hand; unflushed journals still retry. - A well-formed, stable journal whose root shape is unrecognized now warns ('phase publication was NOT verified') and continues, without overriding Claude Code's own permission check; no bypass switch. - Windows paths fold before validation, so '..' is still rejected. - Periodic drift canary runs the reader on fresh journals from the pinned claude. Co-authored-by: Charles M Berg <77517285+CharlesBerg85@users.noreply.github.com> Co-authored-by: laurenbdaniels-coder <264343335+laurenbdaniels-coder@users.noreply.github.com> Co-authored-by: Jordan Mayer <176445995+maydaycyber@users.noreply.github.com> Co-authored-by: elmesery-dev <elmesery.dev@gmail.com> * v1.91.13.0: severe bug fix wave release notes, stale gbrain verbs in docs * test: seed the versioned slug cache in the memory-injection deny case (#2901) The preference hook now resolves the project bucket from gstack-slug's v2 slug-cache entry instead of basename(cwd); this fixture still wrote preferences under the basename bucket without the entry the skill preamble always writes first. * fix: full-suite follow-ups for the fix wave - gstack-slug's --help heredoc is over 512B, so it gets the BASH_COMPAT=50 pipe-deadlock guard like the other bins. - Refresh the codex/factory ship goldens for the plan-gates slug change. - The periodic lane gains the autoplan journal drift canary (107 paid files); raise the eval-slices cap to 25 so every slice still starts at once until its recorded duration lets the planner pack it. * fix(plan-eng-review): clearer Step 0 sequencing, write routes and section outputs The workflow judge scored plan-eng-review's instructions at actionability 3.67 (bar 4) on text unchanged from main. Resolve the ambiguities it named together, without changing any gate or output: Step 0 names the Read's order and keeps one STOP; the pending-call, read-only, forbidden-write and failed-save rules become one Write routes block; header/label limits and grid checks are concrete; Performance gets a checklist; each section has a stated output shape; Lake Score is defined. Same total length (parity cap). --------- Co-authored-by: harshitgavita-07 <harshit.gavita@gmail.com> Co-authored-by: Vitor Cepeda Lopes <vitor@vitorcepedalopes.com> Co-authored-by: Charles M Berg <77517285+CharlesBerg85@users.noreply.github.com> Co-authored-by: laurenbdaniels-coder <264343335+laurenbdaniels-coder@users.noreply.github.com> Co-authored-by: Jordan Mayer <176445995+maydaycyber@users.noreply.github.com> Co-authored-by: elmesery-dev <elmesery.dev@gmail.com>
282 lines
14 KiB
TypeScript
282 lines
14 KiB
TypeScript
/**
|
|
* /context-save + /context-restore project identity (#3003), end to end.
|
|
*
|
|
* Runs the REAL bash blocks extracted from the generated context-save and
|
|
* context-restore SKILL.md files (with ~/.claude/skills/gstack pointed at this
|
|
* checkout) inside fixture repos, so the checkpoint stamp, the restore
|
|
* classification and the helper modes are exercised together. The bug these
|
|
* pin: GitLab repos in different groups (a/product/repo, b/product/repo) shared
|
|
* one projects/product-repo/ bucket, and /context-restore in project B
|
|
* presented project A's checkpoint as the latest one, with nothing failing.
|
|
*/
|
|
import { describe, test, expect, beforeEach, afterEach } from 'bun:test';
|
|
import { spawnSync } from 'child_process';
|
|
import * as fs from 'fs';
|
|
import * as os from 'os';
|
|
import * as path from 'path';
|
|
|
|
const ROOT = path.resolve(import.meta.dir, '..');
|
|
const SLUG_BIN = path.join(ROOT, 'bin', 'gstack-slug');
|
|
|
|
function bashBlockAfter(skillFile: string, heading: string, contains?: string): string {
|
|
const md = fs.readFileSync(path.join(ROOT, skillFile), 'utf-8');
|
|
const start = md.indexOf(heading);
|
|
if (start < 0) throw new Error(`${skillFile}: heading not found: ${heading}`);
|
|
const re = /```bash\n([\s\S]*?)```/g;
|
|
re.lastIndex = start;
|
|
for (let m = re.exec(md); m; m = re.exec(md)) {
|
|
if (!contains || m[1].includes(contains)) return m[1].replaceAll('~/.claude/skills/gstack', ROOT);
|
|
}
|
|
throw new Error(`${skillFile}: no bash block after ${heading}`);
|
|
}
|
|
|
|
const SAVE_PATH_BLOCK = bashBlockAfter('context-save/SKILL.md', '### Step 4: Write saved-context file').replace(
|
|
'TIMESTAMP=$(date +%Y%m%d-%H%M%S)',
|
|
'TIMESTAMP="$TEST_TIMESTAMP"',
|
|
);
|
|
const STAMP_BLOCK = bashBlockAfter('context-save/SKILL.md', '### Step 4: Write saved-context file', '--stamp-checkpoint');
|
|
const RESTORE_BLOCK = bashBlockAfter('context-restore/SKILL.md', '### Step 1: Find saved contexts');
|
|
|
|
let tmp: string;
|
|
let home: string;
|
|
|
|
function env(extra: Record<string, string> = {}): Record<string, string> {
|
|
const out: Record<string, string> = {};
|
|
for (const [k, v] of Object.entries(process.env)) if (v !== undefined) out[k] = v;
|
|
delete out.GSTACK_PROJECT_SLUG;
|
|
delete out.GSTACK_STATE_ROOT;
|
|
delete out.GSTACK_STATE_DIR;
|
|
delete out.CURRENT_BRANCH;
|
|
return { ...out, HOME: tmp, GSTACK_HOME: home, ...extra };
|
|
}
|
|
|
|
function sh(script: string, cwd: string, extra: Record<string, string> = {}) {
|
|
const r = spawnSync('bash', ['-c', script], { cwd, env: env(extra), encoding: 'utf-8', timeout: 30_000 });
|
|
return { stdout: r.stdout ?? '', stderr: r.stderr ?? '', status: r.status ?? -1 };
|
|
}
|
|
|
|
function repo(name: string, origin?: string): string {
|
|
const dir = path.join(tmp, name);
|
|
fs.mkdirSync(dir, { recursive: true });
|
|
for (const args of [['init', '-q', '-b', 'main'], ...(origin ? [['remote', 'add', 'origin', origin]] : [])]) {
|
|
const r = spawnSync('git', args, { cwd: dir, encoding: 'utf-8', timeout: 30_000 });
|
|
if (r.status !== 0) throw new Error(`git ${args.join(' ')}: ${r.stderr}`);
|
|
}
|
|
return dir;
|
|
}
|
|
|
|
/** The context-save flow: path block, the agent's Write, then the stamp block. */
|
|
function save(cwd: string, title: string, timestamp: string, branch = 'main'): string {
|
|
const paths = sh(SAVE_PATH_BLOCK, cwd, { TITLE_RAW: title, TEST_TIMESTAMP: timestamp });
|
|
expect(paths.status).toBe(0);
|
|
const file = paths.stdout.match(/^FILE=(.+)$/m)![1];
|
|
fs.writeFileSync(file, `---\nstatus: in-progress\nbranch: ${branch}\ntimestamp: ${timestamp}\n---\n\n## Working on: ${title}\n`);
|
|
const stamp = sh(STAMP_BLOCK.replace('FILE="<the FILE path printed above>"', `FILE=${JSON.stringify(file)}`), cwd);
|
|
expect(stamp.stdout).toContain('STAMPED');
|
|
return file;
|
|
}
|
|
|
|
function restore(cwd: string, branch = 'main') {
|
|
const r = sh(RESTORE_BLOCK, cwd, { CURRENT_BRANCH: branch });
|
|
expect(r.status).toBe(0);
|
|
const lines = r.stdout.split('\n').filter(Boolean);
|
|
const candidates = lines.filter((l) => l.startsWith('/'));
|
|
return { out: r.stdout, lines, candidates, latest: candidates[0] };
|
|
}
|
|
|
|
function legacyCheckpoint(bucket: string, name: string, branch = 'main'): string {
|
|
const dir = path.join(home, 'projects', bucket, 'checkpoints');
|
|
fs.mkdirSync(dir, { recursive: true });
|
|
const file = path.join(dir, name);
|
|
fs.writeFileSync(file, `---\nstatus: in-progress\nbranch: ${branch}\n---\n\n## Working on: ${name}\n`);
|
|
return file;
|
|
}
|
|
|
|
beforeEach(() => {
|
|
tmp = fs.realpathSync(fs.mkdtempSync(path.join(os.tmpdir(), 'ctx-identity-')));
|
|
home = path.join(tmp, '.gstack');
|
|
});
|
|
|
|
afterEach(() => {
|
|
fs.rmSync(tmp, { recursive: true, force: true });
|
|
});
|
|
|
|
describe('context-save stamps the project identity', () => {
|
|
test('checkpoint frontmatter gets the canonical remote and physical root; credentials never reach disk', () => {
|
|
const secret = ['sentinel', 'pw', '55120'].join('-');
|
|
const a = repo('a', `https://deploy-bot:${secret}@GitLab.com:8443/customer-a/product/repo.git`);
|
|
const file = save(a, 'auth refactor', '20261001-100000');
|
|
expect(file.startsWith(path.join(home, 'projects', 'product-repo-5eca1fe041984543', 'checkpoints'))).toBe(true);
|
|
const text = fs.readFileSync(file, 'utf-8');
|
|
const frontmatter = text.split('---')[1];
|
|
expect(frontmatter).toContain('\nremote: gitlab.com/customer-a/product/repo\n');
|
|
expect(frontmatter).toContain(`\nproject_root: ${a}\n`);
|
|
expect(text).toContain('## Working on: auth refactor');
|
|
const leaks = spawnSync('grep', ['-rl', secret, home], { encoding: 'utf-8', timeout: 10_000 });
|
|
expect(leaks.stdout).toBe('');
|
|
});
|
|
|
|
test('re-stamping replaces stale identity lines instead of duplicating them', () => {
|
|
const a = repo('a', 'git@gitlab.com:customer-a/product/repo.git');
|
|
const file = save(a, 'x', '20261001-100000');
|
|
fs.writeFileSync(file, fs.readFileSync(file, 'utf-8').replace('remote: gitlab.com/customer-a/product/repo', 'remote: bogus'));
|
|
const r = spawnSync('bash', [SLUG_BIN, '--stamp-checkpoint', file], { cwd: a, env: env(), encoding: 'utf-8', timeout: 30_000 });
|
|
expect(r.status).toBe(0);
|
|
const fm = fs.readFileSync(file, 'utf-8').split('---')[1];
|
|
expect(fm.match(/^remote:/gm)?.length).toBe(1);
|
|
expect(fm.match(/^project_root:/gm)?.length).toBe(1);
|
|
expect(fm).toContain('remote: gitlab.com/customer-a/product/repo');
|
|
});
|
|
});
|
|
|
|
describe('context-restore never presents another project checkpoint as latest', () => {
|
|
test('nested groups: B restores after A saved — A is not presented, the old shared bucket is reported', () => {
|
|
const a = repo('a', 'git@gitlab.com:customer-a/product/repo.git');
|
|
const b = repo('b', 'https://gitlab.com/customer-b/product/repo.git');
|
|
// Before the fix both projects filed into projects/product-repo/.
|
|
legacyCheckpoint('product-repo', '20260901-090000-old-a-work.md');
|
|
const aFile = save(a, 'customer a migration', '20261001-120000');
|
|
|
|
const r = restore(b);
|
|
expect(r.out).not.toContain(aFile);
|
|
expect(r.out).not.toContain('old-a-work');
|
|
expect(r.out).toContain('NO_CHECKPOINTS');
|
|
expect(r.out).toContain('PROJECT_IDENTITY: gitlab.com/customer-b/product/repo');
|
|
expect(r.out).toContain('LEGACY_BUCKET: projects/product-repo holds 1 earlier checkpoint(s)');
|
|
|
|
const own = restore(a);
|
|
expect(own.latest).toBe(aFile);
|
|
});
|
|
|
|
test('two projects still sharing one bucket: the foreign newer checkpoint is never "latest"', () => {
|
|
// github.com/product/repo and gitlab.com/product/repo are 2-segment remotes
|
|
// and so keep the same slug (product-repo); identity tells them apart.
|
|
const gh = repo('gh', 'https://github.com/product/repo.git');
|
|
const gl = repo('gl', 'git@gitlab.com:product/repo.git');
|
|
const own = save(gl, 'gitlab work', '20261001-090000');
|
|
const foreign = save(gh, 'github work', '20261001-110000');
|
|
expect(path.dirname(own)).toBe(path.dirname(foreign));
|
|
|
|
const r = restore(gl);
|
|
expect(r.latest).toBe(own);
|
|
expect(r.candidates).not.toContain(foreign);
|
|
expect(r.lines).toContain(`FOREIGN ${foreign}`);
|
|
expect(r.out).toContain('SHARED_BUCKET: 1 checkpoint(s) here were saved by another project');
|
|
|
|
// With no checkpoint of its own, gh's view of gl's bucket shows nothing as latest.
|
|
fs.rmSync(foreign);
|
|
const other = restore(gh);
|
|
expect(other.latest).toBeUndefined();
|
|
expect(other.out).toContain('NO_VERIFIED_CHECKPOINTS');
|
|
expect(other.lines).toContain(`FOREIGN ${own}`);
|
|
});
|
|
|
|
test('unstamped checkpoints: trusted in an unshared bucket, UNVERIFIED once another project wrote there', () => {
|
|
const gh = repo('gh', 'https://github.com/product/repo.git');
|
|
const gl = repo('gl', 'git@gitlab.com:product/repo.git');
|
|
const old = legacyCheckpoint('product-repo', '20260901-090000-pre-stamp.md');
|
|
|
|
expect(restore(gh).latest).toBe(old);
|
|
|
|
const foreign = save(gl, 'gitlab work', '20261001-110000');
|
|
const r = restore(gh);
|
|
expect(r.latest).toBeUndefined();
|
|
expect(r.lines).toContain(`UNVERIFIED ${old}`);
|
|
expect(r.lines).toContain(`FOREIGN ${foreign}`);
|
|
});
|
|
|
|
test('same remote, different checkout (workspace handoff): restorable, with ROOT_DIFFERS as info', () => {
|
|
const first = repo('workspace-1', 'git@github.com:acme/app.git');
|
|
const second = repo('workspace-2', 'https://github.com/acme/app');
|
|
const file = save(first, 'handoff', '20261001-100000', 'feat-x');
|
|
const r = restore(second, 'main');
|
|
expect(r.latest).toBe(file);
|
|
expect(r.lines).toContain(`ROOT_DIFFERS ${file}`);
|
|
expect(r.out).not.toContain('FOREIGN');
|
|
});
|
|
|
|
test('no remote: identity is the project root', () => {
|
|
const one = repo('solo');
|
|
const file = save(one, 'local only', '20261001-100000');
|
|
expect(fs.readFileSync(file, 'utf-8')).not.toMatch(/^remote:/m);
|
|
expect(restore(one).latest).toBe(file);
|
|
|
|
const elsewhere = repo('elsewhere');
|
|
const r = sh(RESTORE_BLOCK, elsewhere, { CURRENT_BRANCH: 'main', GSTACK_PROJECT_SLUG: 'solo' });
|
|
expect(r.stdout).toContain('NO_VERIFIED_CHECKPOINTS');
|
|
expect(r.stdout).toContain(`FOREIGN ${file}`);
|
|
});
|
|
});
|
|
|
|
describe('gstack-slug --adopt-legacy', () => {
|
|
function adopt(cwd: string, ...args: string[]) {
|
|
const r = spawnSync('bash', [SLUG_BIN, '--adopt-legacy', ...args], {
|
|
cwd,
|
|
env: env(),
|
|
encoding: 'utf-8',
|
|
timeout: 30_000,
|
|
stdio: ['ignore', 'pipe', 'pipe'],
|
|
});
|
|
return { stdout: r.stdout ?? '', status: r.status ?? -1 };
|
|
}
|
|
|
|
test('lists the old bucket with labels and copies only what is named; foreign and traversal refused', () => {
|
|
const a = repo('a', 'git@gitlab.com:customer-a/product/repo.git');
|
|
const b = repo('b', 'git@gitlab.com:customer-b/product/repo.git');
|
|
const oldBucket = path.join(home, 'projects', 'product-repo');
|
|
const unstamped = legacyCheckpoint('product-repo', '20260901-090000-unknown.md');
|
|
// Stamped checkpoints in the old bucket (written by A and B), via the real stamp.
|
|
const mine = legacyCheckpoint('product-repo', '20260902-090000-mine.md');
|
|
const theirs = legacyCheckpoint('product-repo', '20260903-090000-theirs.md');
|
|
spawnSync('bash', [SLUG_BIN, '--stamp-checkpoint', mine], { cwd: a, env: env(), timeout: 30_000 });
|
|
spawnSync('bash', [SLUG_BIN, '--stamp-checkpoint', theirs], { cwd: b, env: env(), timeout: 30_000 });
|
|
fs.writeFileSync(path.join(oldBucket, 'question-preferences.json'), '{"review-finding-fix":"never-ask"}');
|
|
fs.writeFileSync(path.join(oldBucket, 'learnings.jsonl'), '{"x":1}\n');
|
|
const before = spawnSync('bash', ['-c', 'find . -type f | sort | xargs cat | cksum'], { cwd: oldBucket, encoding: 'utf-8', timeout: 10_000 }).stdout;
|
|
|
|
const list = adopt(a);
|
|
expect(list.status).toBe(0);
|
|
expect(list.stdout).toContain('checkpoints/20260902-090000-mine.md [checkpoint: matching this project]');
|
|
expect(list.stdout).toContain('checkpoints/20260903-090000-theirs.md [checkpoint: another project (not copyable)]');
|
|
expect(list.stdout).toContain('checkpoints/20260901-090000-unknown.md [checkpoint: unknown project');
|
|
expect(list.stdout).toContain('question-preferences.json [no project identity');
|
|
expect(list.stdout).toContain('Nothing copied.');
|
|
const newBucket = path.join(home, 'projects', 'product-repo-5eca1fe041984543');
|
|
expect(fs.existsSync(path.join(newBucket, 'question-preferences.json'))).toBe(false);
|
|
|
|
const copy = adopt(
|
|
a,
|
|
'--copy',
|
|
'checkpoints/20260902-090000-mine.md',
|
|
'question-preferences.json',
|
|
'checkpoints/20260903-090000-theirs.md',
|
|
'../product-repo/learnings.jsonl',
|
|
);
|
|
expect(copy.stdout).toContain('COPIED checkpoints/20260902-090000-mine.md');
|
|
expect(copy.stdout).toContain('COPIED question-preferences.json');
|
|
expect(copy.stdout).toContain('SKIPPED checkpoints/20260903-090000-theirs.md');
|
|
expect(copy.stdout).toContain('REFUSED ../product-repo/learnings.jsonl');
|
|
expect(copy.status).toBe(1);
|
|
expect(fs.readFileSync(path.join(newBucket, 'question-preferences.json'), 'utf-8')).toContain('never-ask');
|
|
expect(fs.existsSync(path.join(newBucket, 'checkpoints', '20260903-090000-theirs.md'))).toBe(false);
|
|
expect(fs.existsSync(path.join(newBucket, 'learnings.jsonl'))).toBe(false);
|
|
expect(fs.existsSync(path.join(newBucket, 'checkpoints', path.basename(unstamped)))).toBe(false);
|
|
|
|
// Never overwrites, never touches the old bucket.
|
|
const again = adopt(a, '--copy', 'question-preferences.json');
|
|
expect(again.stdout).toContain('SKIPPED question-preferences.json');
|
|
const after = spawnSync('bash', ['-c', 'find . -type f | sort | xargs cat | cksum'], { cwd: oldBucket, encoding: 'utf-8', timeout: 10_000 }).stdout;
|
|
expect(after).toBe(before);
|
|
|
|
// The adopted checkpoint is now A's latest on restore.
|
|
expect(restore(a).latest).toBe(path.join(newBucket, 'checkpoints', '20260902-090000-mine.md'));
|
|
});
|
|
|
|
test('a project whose slug did not change has nothing to adopt', () => {
|
|
const gh = repo('gh', 'https://github.com/acme/app.git');
|
|
const r = adopt(gh);
|
|
expect(r.status).toBe(0);
|
|
expect(r.stdout).toContain("Nothing to adopt: this project's slug (acme-app) did not change.");
|
|
});
|
|
});
|