The windows-test-parity ratchet greps test source for fs.rmSync-without-
maxRetries (and six other Windows-portability anti-patterns), failing when an
integer offender COUNT exceeds a frozen baseline (rmSync: 95). A count ratchet
is a Goodhart metric: fixing one offender and adding another keeps the count
constant, so a new defect slips through green. Replace it — and every other
count ratchet in the repo — with a layered, masking-proof design.
Behavioral seam test
- tests/helpers-cleanup.test.cjs proves helpers.cleanup() carries the Windows
EBUSY retry budget. cleanup() delegates retries to Node's fs.rmSync via
maxRetries (it owns no loop), so the test asserts the option contract
(recursive/force/maxRetries>0/retryDelay>0) + real-FS removal + the cwd-guard,
rather than a loop that does not exist. The EBUSY risk is now tested ONCE at
the helper, not approximated textually at every call site.
Write-time ESLint rule (AST-accurate, replaces the grep)
- eslint-rules/no-raw-rmsync-in-tests.cjs (error in tests/**/*.test.cjs) bans
raw fs.rmSync, steering to cleanup(). Catches member, computed (fs['rmSync']),
destructured and aliased forms; escape hatch is inline
`// eslint-disable-next-line local/no-raw-rmsync-in-tests -- <reason>` only.
- Migrated 336 raw fs.rmSync teardown calls across ~116 test files to cleanup().
~18 genuinely load-bearing sites (mid-test SUT/fault-injection removals,
error-swallowing or name-colliding local teardown helpers) keep the raw call
with an inline eslint-disable + reason.
Shared anti-ratchet primitive
- scripts/lib/allowlist-ratchet.cjs:
- assertWithinAllowlist: fails on NOVEL ids (new offender introduced) AND on
STALE ids (a known offender was fixed but not pruned) — identity, not count,
and a ratchet DOWN toward zero.
- assertTightCeiling: a size/length budget whose ceiling must stay within a
grace band of the high-water mark, so budgets may only tighten, never creep.
Ratchets converted onto the primitive
- windows-test-parity-guard.test.cjs: rmSync rule deleted (now ESLint-enforced);
the remaining six patterns moved from integer baselines to named-set
allowlists with ratchet-down.
- scripts/lint-test-file-count.{cjs,allowlist.json}: per-module integer counts →
named filename sets (closes the swap-a-file-keep-the-count blind spot); a
module dropping under cap now FAILS to force pruning its allowlist entry.
- enh-2790 skill-count `<= 63` → named skill allowlist (ratchets toward ~58).
Size budgets hardened (tighten-only)
- agent-size / workflow-size / feat-3039 help-tiered: ceilings lowered to the
current high-water mark and an assertTightCeiling anti-creep check added per
tier. Fixed external-contract limits (description ≤100 chars, agent ≤100 KB)
are intentionally left as-is — they are not grandfathered creeping budgets.
No user-facing behavior change (tests + tooling only); no USER_FACING_PREFIXES
touched, so no changeset fragment is required.
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
112 lines
4.2 KiB
JavaScript
112 lines
4.2 KiB
JavaScript
// allow-test-rule: source-text-is-the-product
|
|
// Workflow .md / agent .md / command .md / reference .md files — their text
|
|
// IS what the runtime loads. Testing text content tests the deployed contract.
|
|
// Per CONTRIBUTING.md exception matrix.
|
|
|
|
/**
|
|
* Regression: issue #2639 — Codex install generated agent TOMLs with stale
|
|
* Claude-specific references (CLAUDE.md, .claude/skills/, .claudeignore).
|
|
*
|
|
* RCA: `installCodexConfig()` applied a narrow path-only regex pass before
|
|
* calling `generateCodexAgentToml()`, bypassing the full
|
|
* `convertClaudeToCodexMarkdown()` + `neutralizeAgentReferences(..., 'AGENTS.md')`
|
|
* pipeline used on the .md emit path. Fix routes the TOML path through the
|
|
* same pipeline and extends the pipeline to cover bare `.claude/skills/`,
|
|
* `.claude/commands/`, `.claude/agents/`, and `.claudeignore`.
|
|
*/
|
|
|
|
process.env.GSD_TEST_MODE = '1';
|
|
|
|
const { test, describe, beforeEach, afterEach } = require('node:test');
|
|
const assert = require('node:assert/strict');
|
|
const fs = require('fs');
|
|
const path = require('path');
|
|
const os = require('os');
|
|
|
|
const { installCodexConfig } = require('../bin/install.js');
|
|
const { cleanup } = require('./helpers.cjs');
|
|
|
|
function makeTempDir() {
|
|
return fs.mkdtempSync(path.join(os.tmpdir(), 'gsd-2639-'));
|
|
}
|
|
|
|
function writeAgentFixture(agentsSrc, name, body) {
|
|
const content = `---
|
|
name: ${name}
|
|
description: Test agent for #2639
|
|
---
|
|
|
|
${body}
|
|
`;
|
|
fs.writeFileSync(path.join(agentsSrc, `${name}.md`), content);
|
|
}
|
|
|
|
describe('#2639 — Codex TOML emit routes through full neutralization pipeline', () => {
|
|
let tmpDir;
|
|
let agentsSrc;
|
|
let targetDir;
|
|
|
|
beforeEach(() => {
|
|
tmpDir = makeTempDir();
|
|
agentsSrc = path.join(tmpDir, 'agents');
|
|
targetDir = path.join(tmpDir, 'codex');
|
|
fs.mkdirSync(agentsSrc, { recursive: true });
|
|
fs.mkdirSync(targetDir, { recursive: true });
|
|
});
|
|
|
|
afterEach(() => {
|
|
cleanup(tmpDir);
|
|
});
|
|
|
|
test('strips CLAUDE.md, .claude/skills/, .claude/commands/, .claude/agents/, and .claudeignore from emitted TOML', () => {
|
|
writeAgentFixture(agentsSrc, 'gsd-code-reviewer', [
|
|
'**Project instructions:** Read `./CLAUDE.md` if it exists.',
|
|
'',
|
|
'**CLAUDE.md enforcement:** If `./CLAUDE.md` exists, treat it as hard constraints.',
|
|
'',
|
|
'**Project skills:** Check `.claude/skills/` or `.agents/skills/` directory.',
|
|
'',
|
|
'Also check `.claude/commands/` and `.claude/agents/` for definitions.',
|
|
'',
|
|
'DO respect .gitignore and .claudeignore. Do not review ignored files.',
|
|
'',
|
|
'Claude will refuse the task if policy violated.',
|
|
].join('\n'));
|
|
|
|
installCodexConfig(targetDir, agentsSrc);
|
|
|
|
const tomlPath = path.join(targetDir, 'agents', 'gsd-code-reviewer.toml');
|
|
assert.ok(fs.existsSync(tomlPath), 'per-agent TOML written');
|
|
const toml = fs.readFileSync(tomlPath, 'utf8');
|
|
|
|
assert.ok(!toml.includes('CLAUDE.md'), 'no CLAUDE.md references remain in TOML');
|
|
assert.ok(!toml.includes('.claude/skills/'), 'no .claude/skills/ references remain');
|
|
assert.ok(!toml.includes('.claude/commands/'), 'no .claude/commands/ references remain');
|
|
assert.ok(!toml.includes('.claude/agents/'), 'no .claude/agents/ references remain');
|
|
assert.ok(!toml.includes('.claudeignore'), 'no .claudeignore references remain');
|
|
|
|
assert.ok(toml.includes('AGENTS.md'), 'AGENTS.md substituted for CLAUDE.md');
|
|
assert.ok(
|
|
toml.includes('.codex/skills/') || toml.includes('.agents/skills/'),
|
|
'skills path neutralized'
|
|
);
|
|
|
|
// Standalone "Claude" agent-name references replaced
|
|
assert.ok(!/\bClaude\b(?! Code| Opus| Sonnet| Haiku| native| based)/.test(toml),
|
|
'standalone Claude agent-name references replaced');
|
|
});
|
|
|
|
test('preserves Claude product/model names (Claude Code, Claude Opus) in TOML', () => {
|
|
writeAgentFixture(agentsSrc, 'gsd-executor', [
|
|
'This agent runs under Claude Code with the Claude Opus 4 model.',
|
|
'Do not confuse with Claude Sonnet or Claude Haiku.',
|
|
].join('\n'));
|
|
|
|
installCodexConfig(targetDir, agentsSrc);
|
|
const toml = fs.readFileSync(path.join(targetDir, 'agents', 'gsd-executor.toml'), 'utf8');
|
|
|
|
assert.ok(toml.includes('Claude Code'), 'Claude Code product name preserved');
|
|
assert.ok(toml.includes('Claude Opus'), 'Claude Opus model name preserved');
|
|
});
|
|
});
|