config-audit/tests/scanners/posture-humanizer.test.mjs
Kjell Tore Guttormsen e3b044a476 feat(ost): v5.6 C — output-style scanner (CA-OST, count 13→14)
New orchestrated scanner output-style-scanner.mjs — first new family since
SKL. Three findings, each pinned to a CONFIRMED V-row of the steering-model
plan + re-verified against code.claude.com/docs/en/output-styles:

- CA-OST-001 (medium, V10): user/project custom style missing
  keep-coding-instructions:true (default false) → silently strips built-in
  software-engineering instructions when active. Scoped to user/project.
- CA-OST-002 (low, V11): plugin style with force-for-plugin:true overrides the
  user's selected outputStyle. Verifiseringsplikt correction — the plan bullet
  said "project/user style," but force-for-plugin is plugin-styles-only per the
  docs, so the check keys on source==='plugin'.
- CA-OST-003 (medium): settings outputStyle matching no built-in
  (Default/Explanatory/Learning/Proactive, case-insensitive) nor discovered
  custom style → dead config.

Byte-stability — a scanner addition, not a field addition. Growing the
scanners array + scanners_ok cannot be hidden by a field strip, but re-seeding
frozen v5.0.0 (the SKL precedent) would now bake in B2's hotspot triple +
claudeMd drift. So, per the B2 lesson, frozen v5.0.0 snapshots are PRESERVED
and the OST entry is stripped at compare time via new
tests/helpers/strip-added-scanner.mjs (wired into json/raw-backcompat + the
Step 5/6 humanizer tests); only SC-5 default-output is regenerated (additive
OST entry, diff reviewed). OST is fixture-gated (no output styles on
marketplace-medium / hermetic HOME → silent).

Wiring: orchestrator; humanizer (OST→Configuration mistake) + humanizer-data
OST family (title-coupled); scoring (OST→Settings, keeps 10 areas). Suite
1012→1023 (+11). Badges: scanners 14, tests 1023, TRANSLATIONS families 15.
Lore swept: README, CLAUDE.md, scanner-internals, humanizer.md. self-audit A/A.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-20 21:02:44 +02:00

144 lines
6.1 KiB
JavaScript

import { describe, it } from 'node:test';
import assert from 'node:assert/strict';
import { resolve, dirname } from 'node:path';
import { fileURLToPath } from 'node:url';
import { execFile } from 'node:child_process';
import { promisify } from 'node:util';
import { readFile, unlink } from 'node:fs/promises';
import { hermeticEnv } from '../helpers/hermetic-home.mjs';
import { stripHotspotLoadPattern } from '../helpers/strip-hotspot-load-pattern.mjs';
import { stripAddedScanners, stripAddedScannerStderr } from '../helpers/strip-added-scanner.mjs';
const exec = promisify(execFile);
const __dirname = dirname(fileURLToPath(import.meta.url));
const REPO = resolve(__dirname, '../..');
const CLI = resolve(REPO, 'scanners/posture.mjs');
const FIXTURE = resolve(REPO, 'tests/fixtures/marketplace-medium');
const POSTURE_JSON_SNAPSHOT = resolve(REPO, 'tests/snapshots/v5.0.0/posture.json');
const POSTURE_STDERR_SNAPSHOT = resolve(REPO, 'tests/snapshots/v5.0.0-stderr/posture.txt');
/**
* Normalize a runPosture result for snapshot comparison by zeroing out
* time-varying fields, machine-specific paths, and ancestor-cascade-derived
* counts. `claudeMdEstimatedTokens` reflects walkClaudeMdCascade walking
* upward from the fixture; any docs edit to this plugin's own CLAUDE.md
* ripples into it even though scanner behavior is unchanged.
*/
function normalizePosture(p) {
const out = JSON.parse(JSON.stringify(p));
if (out.scannerEnvelope) {
if (out.scannerEnvelope.meta) {
out.scannerEnvelope.meta.target = '<TARGET>';
out.scannerEnvelope.meta.timestamp = '<TIMESTAMP>';
}
if (Array.isArray(out.scannerEnvelope.scanners)) {
for (const s of out.scannerEnvelope.scanners) {
s.duration_ms = 0;
if (s.activeConfig && 'claudeMdEstimatedTokens' in s.activeConfig) {
s.activeConfig.claudeMdEstimatedTokens = '<ANCESTOR_DERIVED>';
}
}
}
}
return stripAddedScanners(stripHotspotLoadPattern(out));
}
/**
* Strip time-varying durations (Xms) so progress lines compare verbatim across
* runs, and drop the additive [OST] progress line (v5.6 C) so a 14-scanner live
* run matches the frozen 13-scanner v5.0.0 stderr scorecard.
*/
function normalizeStderr(s) {
return stripAddedScannerStderr(s).replace(/\(\d+ms\)/g, '(0ms)');
}
async function runPosture(flags) {
const proc = await exec('node', [CLI, FIXTURE, ...flags], {
timeout: 60000,
cwd: REPO,
env: hermeticEnv(),
}).catch(err => err); // posture exits non-zero on findings — capture either way
return {
stdout: proc.stdout || '',
stderr: proc.stderr || '',
};
}
describe('posture humanizer wiring (Step 6)', () => {
describe('--json mode (SC-6: byte-equal stdout)', () => {
it('stdout JSON deepEquals v5.0.0 snapshot', async () => {
const { stdout } = await runPosture(['--json']);
const actual = JSON.parse(stdout);
const expected = JSON.parse(await readFile(POSTURE_JSON_SNAPSHOT, 'utf-8'));
assert.deepStrictEqual(normalizePosture(actual), normalizePosture(expected));
});
it('does NOT write a scorecard to stderr (suppressed)', async () => {
const { stderr } = await runPosture(['--json']);
assert.ok(!stderr.includes('Config-Audit Health Score'),
'stderr must NOT contain scorecard in --json mode');
assert.ok(!stderr.includes('Configuration health'),
'stderr must NOT contain humanized scorecard in --json mode');
});
it('preserves v5.0.0 finding shape (no humanizer fields in scannerEnvelope)', async () => {
const { stdout } = await runPosture(['--json']);
const actual = JSON.parse(stdout);
for (const s of actual.scannerEnvelope.scanners) {
for (const f of s.findings) {
assert.equal(f.userImpactCategory, undefined,
`${f.id}: --json findings must not have userImpactCategory`);
}
}
});
});
describe('--raw mode (SC-7: byte-equal stdout + verbatim stderr)', () => {
it('stdout JSON deepEquals v5.0.0 snapshot', async () => {
const { stdout } = await runPosture(['--raw']);
const actual = JSON.parse(stdout);
const expected = JSON.parse(await readFile(POSTURE_JSON_SNAPSHOT, 'utf-8'));
assert.deepStrictEqual(normalizePosture(actual), normalizePosture(expected));
});
it('stderr scorecard verbatim matches v5.0.0 stderr snapshot', async () => {
const { stderr } = await runPosture(['--raw']);
const expected = await readFile(POSTURE_STDERR_SNAPSHOT, 'utf-8');
// Compare the scorecard portion verbatim (modulo timing in scanner progress lines)
assert.equal(normalizeStderr(stderr).trim(), normalizeStderr(expected).trim());
});
it('preserves v5.0.0 finding shape in stdout', async () => {
const { stdout } = await runPosture(['--raw']);
const actual = JSON.parse(stdout);
for (const s of actual.scannerEnvelope.scanners) {
for (const f of s.findings) {
assert.equal(f.userImpactCategory, undefined,
`${f.id}: --raw findings must not have userImpactCategory`);
}
}
});
});
describe('default mode (humanized scorecard)', () => {
it('writes humanized scorecard to stderr', async () => {
const { stderr } = await runPosture([]);
// Humanized scorecard must contain at least one user-friendly cue not in raw v5.0.0
const hasGradeContext = /healthy|good shape|attention|polish|setup/i.test(stderr);
assert.ok(hasGradeContext,
`humanized stderr scorecard must contain user-friendly phrasing, got:\n${stderr}`);
});
it('does NOT write JSON to stdout in default mode', async () => {
const { stdout } = await runPosture([]);
assert.equal(stdout.trim(), '', 'default mode must not write JSON to stdout');
});
it('humanized scorecard differs byte-wise from v5.0.0 stderr', async () => {
const { stderr } = await runPosture([]);
const expected = await readFile(POSTURE_STDERR_SNAPSHOT, 'utf-8');
assert.notEqual(normalizeStderr(stderr).trim(), normalizeStderr(expected).trim(),
'humanized stderr must differ from v5.0.0 verbatim stderr');
});
});
});