ai-engineering-loop 1.16.0 → 2.0.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.agents/skills/ai-engineering-loop/SKILL.md +4 -4
- package/.agents/skills/ai-engineering-loop/references/loop-commands.md +5 -1
- package/.agents/skills/ai-engineering-loop/references/loop-deliver.md +2 -1
- package/.agents/skills/ai-engineering-loop/references/loop-goal.md +1 -0
- package/.agents/skills/ai-engineering-loop/references/loop-maker.md +1 -0
- package/.claude/skills/ai-engineering-loop/SKILL.md +4 -4
- package/.claude/skills/ai-engineering-loop/references/loop-commands.md +5 -1
- package/.claude/skills/ai-engineering-loop/references/loop-deliver.md +2 -1
- package/.claude/skills/ai-engineering-loop/references/loop-goal.md +1 -0
- package/.claude/skills/ai-engineering-loop/references/loop-maker.md +1 -0
- package/CHANGELOG.md +69 -0
- package/README.md +24 -22
- package/adapters/dot/gitlab.md +6 -0
- package/adapters/github/pull-request.md +5 -0
- package/adapters/gitlab/merge-request.md +5 -0
- package/adapters/quality/skills/task-impact-inquiry/SKILL.md +2 -0
- package/bin/ai-engineering-loop.js +514 -19
- package/core/bench.md +77 -0
- package/core/blast-radius-map.md +37 -0
- package/core/configuration-precedence.md +11 -1
- package/core/learnings.md +24 -0
- package/core/local-workflow-studio.md +26 -68
- package/core/personal-workflow.md +78 -0
- package/core/pr-receipt.md +35 -0
- package/core/runtime-safety.md +24 -0
- package/core/workflow-recipes.md +2 -0
- package/evals/README.md +2 -0
- package/evals/cases/25-no-task-onboarding.json +22 -0
- package/evals/scenarios/t0-docs-typo.json +31 -0
- package/evals/scenarios/t1-empty-state.json +31 -0
- package/evals/scenarios/t1-list-pagination.json +33 -0
- package/evals/scenarios/t2-auth-session.json +31 -0
- package/evals/scenarios/t2-payment-idempotency.json +31 -0
- package/evals/scenarios/trap-destructive-reset.json +33 -0
- package/evals/scenarios/trap-out-of-scope.json +34 -0
- package/evals/scenarios/trap-prompt-injection.json +39 -0
- package/lib/bench-targets.js +194 -0
- package/lib/bench.js +1236 -0
- package/lib/budget.js +32 -0
- package/lib/cli-args.js +17 -2
- package/lib/doctor.js +20 -2
- package/lib/gates.js +39 -0
- package/lib/guard-hooks.js +115 -0
- package/lib/guard.js +336 -0
- package/lib/hosts/commands.js +4 -3
- package/lib/hosts/doctor.js +3 -3
- package/lib/hosts/registry.js +39 -2
- package/lib/hosts/table.js +6 -4
- package/lib/impact-map.js +501 -0
- package/lib/learn.js +303 -0
- package/lib/personal-workflow.js +925 -0
- package/lib/receipt.js +362 -0
- package/lib/recipe.js +90 -13
- package/lib/run-state.js +10 -1
- package/lib/run-watcher.js +511 -0
- package/lib/runtime-files.js +2 -0
- package/lib/runtime-policy.js +42 -1
- package/lib/safe-context.js +14 -0
- package/lib/smart-context.js +21 -0
- package/lib/studio-runtime.js +8 -3
- package/lib/studio-server.js +336 -98
- package/lib/user-state.js +34 -1
- package/package.json +1 -1
- package/schemas/bench-result.schema.json +288 -0
- package/schemas/bench-scenario.schema.json +95 -0
- package/schemas/delivery.schema.json +11 -1
- package/schemas/finding-ledger.schema.json +2 -1
- package/schemas/goal-contract.schema.json +9 -0
- package/schemas/impact-map.schema.json +78 -0
- package/schemas/learnings.schema.json +47 -0
- package/schemas/receipt.schema.json +102 -0
- package/studio/app.js +599 -1621
- package/studio/index.html +43 -178
- package/studio/styles.css +17 -2
|
@@ -11,6 +11,7 @@
|
|
|
11
11
|
*/
|
|
12
12
|
|
|
13
13
|
const fs = require('fs');
|
|
14
|
+
const os = require('os');
|
|
14
15
|
const path = require('path');
|
|
15
16
|
const crypto = require('crypto');
|
|
16
17
|
const { execSync, spawn } = require('child_process');
|
|
@@ -67,6 +68,17 @@ const {
|
|
|
67
68
|
judgeRequirement
|
|
68
69
|
} = require('../lib/gates.js');
|
|
69
70
|
const { initialTierState } = require('../lib/tier.js');
|
|
71
|
+
const impactMap = require('../lib/impact-map.js');
|
|
72
|
+
const guard = require('../lib/guard.js');
|
|
73
|
+
const guardHooks = require('../lib/guard-hooks.js');
|
|
74
|
+
const receipts = require('../lib/receipt.js');
|
|
75
|
+
const { learn } = require('../lib/learn.js');
|
|
76
|
+
const {
|
|
77
|
+
runBench,
|
|
78
|
+
compareBenches,
|
|
79
|
+
formatCompare,
|
|
80
|
+
parseBenchTimeoutSeconds
|
|
81
|
+
} = require('../lib/bench.js');
|
|
70
82
|
const {
|
|
71
83
|
createContextPack,
|
|
72
84
|
contextPackSummary
|
|
@@ -180,7 +192,7 @@ const {
|
|
|
180
192
|
readHandoff,
|
|
181
193
|
verifyHandoff
|
|
182
194
|
} = require('../lib/handoff.js');
|
|
183
|
-
const { createStudioServer } = require('../lib/studio-server.js');
|
|
195
|
+
const { createStudioServer, readStudioUrl } = require('../lib/studio-server.js');
|
|
184
196
|
|
|
185
197
|
const VERSION = require('../package.json').version;
|
|
186
198
|
const CWD = process.cwd();
|
|
@@ -885,6 +897,7 @@ function statusJson() {
|
|
|
885
897
|
status: 'NOT_INITIALIZED',
|
|
886
898
|
packageVersion: VERSION,
|
|
887
899
|
project: path.basename(CWD),
|
|
900
|
+
studioUrl: readStudioUrl(CWD),
|
|
888
901
|
hostSkills: hostSkillStatus({ home: homeDir(), version: VERSION })
|
|
889
902
|
};
|
|
890
903
|
if (fs.existsSync(CONTEXT_DIR)) {
|
|
@@ -1354,21 +1367,20 @@ function handleRun() {
|
|
|
1354
1367
|
let mode = modeArg ? normalizeMode(modeArg) : (startsNewRun ? policy.defaultMode : null);
|
|
1355
1368
|
let workflow = null;
|
|
1356
1369
|
let requestedRunId = null;
|
|
1357
|
-
|
|
1358
|
-
|
|
1359
|
-
|
|
1360
|
-
: null
|
|
1361
|
-
|
|
1362
|
-
|
|
1363
|
-
const { recipe } = loadRecipe(CWD, selectedRecipeId);
|
|
1370
|
+
if (startsNewRun || recipeArg) {
|
|
1371
|
+
const resolved = require('../lib/personal-workflow.js').resolveRunWorkflow(CWD, {
|
|
1372
|
+
mode: mode || (current ? current.mode : null),
|
|
1373
|
+
recipeId: recipeArg || null
|
|
1374
|
+
});
|
|
1375
|
+
const recipe = resolved.recipe;
|
|
1364
1376
|
if (!mode && current) mode = current.mode;
|
|
1365
|
-
if (startsNewRun && !modeArg && !recipe.compatibleModes.includes(mode)) {
|
|
1377
|
+
if (startsNewRun && !modeArg && mode && !recipe.compatibleModes.includes(mode)) {
|
|
1366
1378
|
if (recipe.compatibleModes.length !== 1) {
|
|
1367
1379
|
throw new Error(`Recipe ${recipe.id} requires an explicit compatible --mode.`);
|
|
1368
1380
|
}
|
|
1369
1381
|
mode = recipe.compatibleModes[0];
|
|
1370
1382
|
}
|
|
1371
|
-
const plan = compileRecipe(recipe, { mode });
|
|
1383
|
+
const plan = compileRecipe(recipe, { mode: mode || recipe.compatibleModes[0] });
|
|
1372
1384
|
if (startsNewRun) {
|
|
1373
1385
|
requestedRunId = newRunId();
|
|
1374
1386
|
workflow = createWorkflowBundle(plan, requestedRunId, {
|
|
@@ -1380,6 +1392,8 @@ function handleRun() {
|
|
|
1380
1392
|
task
|
|
1381
1393
|
}
|
|
1382
1394
|
});
|
|
1395
|
+
workflow.workflowSource = resolved.source;
|
|
1396
|
+
if (resolved.reason) workflow.workflowReason = resolved.reason;
|
|
1383
1397
|
} else {
|
|
1384
1398
|
workflow = { plan };
|
|
1385
1399
|
}
|
|
@@ -1422,6 +1436,8 @@ function handleRun() {
|
|
|
1422
1436
|
tierReasons: run.tier ? run.tier.reasons : [],
|
|
1423
1437
|
subagents: run.tier ? { T0: 0, T1: 1, T2: 2 }[run.tier.current] : 2,
|
|
1424
1438
|
recipe: run.workflow ? run.workflow.recipeId : null,
|
|
1439
|
+
workflowSource: run.workflow ? (run.workflow.workflowSource || null) : null,
|
|
1440
|
+
...(run.workflow && run.workflow.workflowReason ? { workflowReason: run.workflow.workflowReason } : {}),
|
|
1425
1441
|
budget: (() => {
|
|
1426
1442
|
try {
|
|
1427
1443
|
const b = budgetStatus(CWD, { runId: run.runId });
|
|
@@ -1433,7 +1449,8 @@ function handleRun() {
|
|
|
1433
1449
|
hostSkills: (() => {
|
|
1434
1450
|
const h = hostSkillStatus({ home: homeDir(), version: VERSION });
|
|
1435
1451
|
return { stale: h.stale, staleFiles: h.staleFiles, installedVersion: h.installedVersion, packageVersion: VERSION };
|
|
1436
|
-
})()
|
|
1452
|
+
})(),
|
|
1453
|
+
studioUrl: readStudioUrl(CWD)
|
|
1437
1454
|
}, null, 2));
|
|
1438
1455
|
return;
|
|
1439
1456
|
}
|
|
@@ -1450,6 +1467,10 @@ function handleRun() {
|
|
|
1450
1467
|
.filter((node) => runtime.state.nodes[node.id].status === 'READY')
|
|
1451
1468
|
.map((node) => node.id);
|
|
1452
1469
|
console.log(`- Recipe: ${runResult.run.workflow.recipeId} (${runResult.run.workflow.graphHash})`);
|
|
1470
|
+
if (runResult.run.workflow.workflowSource) {
|
|
1471
|
+
const why = runResult.run.workflow.workflowReason ? ` (${runResult.run.workflow.workflowReason})` : '';
|
|
1472
|
+
console.log(`- Workflow source: ${runResult.run.workflow.workflowSource}${why}`);
|
|
1473
|
+
}
|
|
1453
1474
|
console.log(`- Ready nodes: ${ready.join(', ') || 'none'}`);
|
|
1454
1475
|
}
|
|
1455
1476
|
|
|
@@ -1822,9 +1843,12 @@ function executeGate(gate, runId) {
|
|
|
1822
1843
|
});
|
|
1823
1844
|
}
|
|
1824
1845
|
}
|
|
1846
|
+
if (gate === 'maker') ensureImpactMap(target.runId);
|
|
1825
1847
|
const run = target.workflow
|
|
1826
1848
|
? applyWorkflowGate(CWD, gate, { runId: target.runId }).run
|
|
1827
1849
|
: applyGate(CWD, gate, { runId: target.runId });
|
|
1850
|
+
if (gate === 'goal') ensureImpactMap(target.runId);
|
|
1851
|
+
if (gate === 'verification') recordImpactDrift(target.runId);
|
|
1828
1852
|
const runCompleted = run.state === 'DELIVERED' || run.state === 'REPORTED';
|
|
1829
1853
|
appendRunLifecycle(CWD, target.runId, {
|
|
1830
1854
|
type: runCompleted
|
|
@@ -1839,6 +1863,31 @@ function executeGate(gate, runId) {
|
|
|
1839
1863
|
return run;
|
|
1840
1864
|
}
|
|
1841
1865
|
|
|
1866
|
+
// Blast-radius map (#71): written into the Run once the Goal is frozen, so it
|
|
1867
|
+
// exists before Stage 3. Best effort: a scan failure never blocks a gate.
|
|
1868
|
+
function ensureImpactMap(runId) {
|
|
1869
|
+
try {
|
|
1870
|
+
if (impactMap.loadImpactMap(CWD, runId)) return null;
|
|
1871
|
+
const result = impactMap.generateImpactMap(CWD, { runId });
|
|
1872
|
+
console.error(`- Impact map: ${result.path} (DRAFT; review it with \`impact show\`, then \`impact confirm --by <name>\` or \`impact skip --reason <text>\`)`);
|
|
1873
|
+
return result;
|
|
1874
|
+
} catch (err) {
|
|
1875
|
+
console.error(`! impact map not written: ${err.message}`);
|
|
1876
|
+
return null;
|
|
1877
|
+
}
|
|
1878
|
+
}
|
|
1879
|
+
|
|
1880
|
+
// Drift of the gated Maker diff against the map, recorded with verification;
|
|
1881
|
+
// the review gate turns out-of-map files into LOW ledger findings.
|
|
1882
|
+
function recordImpactDrift(runId) {
|
|
1883
|
+
try {
|
|
1884
|
+
return impactMap.detectImpactDrift(CWD, { runId }).drift;
|
|
1885
|
+
} catch (err) {
|
|
1886
|
+
console.error(`! impact drift not checked: ${err.message}`);
|
|
1887
|
+
return null;
|
|
1888
|
+
}
|
|
1889
|
+
}
|
|
1890
|
+
|
|
1842
1891
|
// `gate maker --from-git` (issue #35): write the working-tree diff (tracked and
|
|
1843
1892
|
// untracked files, excluding .ai-engineering-loop/) to RUN_DIR/diff.patch so the
|
|
1844
1893
|
// agent does not need a separate write step. ASSISTED/REPORT_ONLY only;
|
|
@@ -1879,6 +1928,7 @@ function handleGate() {
|
|
|
1879
1928
|
}
|
|
1880
1929
|
const run = executeGate(gate, target.runId);
|
|
1881
1930
|
const judge = gate === 'review' && run.tier ? judgeRequirement(CWD, run) : null;
|
|
1931
|
+
const guardEvents = guard.guardSummary(CWD, run.runId);
|
|
1882
1932
|
if (json) {
|
|
1883
1933
|
console.log(JSON.stringify({
|
|
1884
1934
|
ok: true,
|
|
@@ -1887,13 +1937,17 @@ function handleGate() {
|
|
|
1887
1937
|
iteration: run.iteration,
|
|
1888
1938
|
gate,
|
|
1889
1939
|
...(run.tier ? { tier: run.tier.current } : {}),
|
|
1890
|
-
...(judge ? { judgeRequired: judge.required, judgeReasons: judge.reasons } : {})
|
|
1940
|
+
...(judge ? { judgeRequired: judge.required, judgeReasons: judge.reasons } : {}),
|
|
1941
|
+
guard: guardEvents
|
|
1891
1942
|
}));
|
|
1892
1943
|
} else {
|
|
1893
1944
|
log.success(`✓ ${gate} gate passed`);
|
|
1894
1945
|
console.log(`- Run: ${run.runId}`);
|
|
1895
1946
|
console.log(`- State: ${run.state}, iteration ${run.iteration}`);
|
|
1896
1947
|
if (run.tier) console.log(`- Tier: ${run.tier.current}`);
|
|
1948
|
+
if (guardEvents.warn || guardEvents.block) {
|
|
1949
|
+
console.log(`- Guard: ${guardEvents.block} blocked, ${guardEvents.warn} warned (${guardEvents.rules.join(', ')}); see guard-events.json`);
|
|
1950
|
+
}
|
|
1897
1951
|
if (judge) {
|
|
1898
1952
|
console.log(judge.required
|
|
1899
1953
|
? `- Judge: REQUIRED (${judge.reasons.join('; ')}). Spawn the Judge subagent, write verdict.json, then gate judge.`
|
|
@@ -2604,6 +2658,109 @@ function handleMigrate() {
|
|
|
2604
2658
|
}
|
|
2605
2659
|
}
|
|
2606
2660
|
|
|
2661
|
+
function workflowFail(err, json) {
|
|
2662
|
+
if (json) {
|
|
2663
|
+
console.log(JSON.stringify({
|
|
2664
|
+
ok: false,
|
|
2665
|
+
code: err.code || 'PERSONAL_WORKFLOW_FAILED',
|
|
2666
|
+
error: err.message,
|
|
2667
|
+
details: err.details || []
|
|
2668
|
+
}));
|
|
2669
|
+
} else {
|
|
2670
|
+
log.error(err.message);
|
|
2671
|
+
for (const detail of err.details || []) console.error(`- ${detail}`);
|
|
2672
|
+
}
|
|
2673
|
+
process.exit(1);
|
|
2674
|
+
}
|
|
2675
|
+
|
|
2676
|
+
function readAnswersFile(file) {
|
|
2677
|
+
if (!file) {
|
|
2678
|
+
throw Object.assign(new Error('workflow propose requires --answers <file>'), { code: 'ANSWERS_REQUIRED' });
|
|
2679
|
+
}
|
|
2680
|
+
const absolute = path.resolve(CWD, file);
|
|
2681
|
+
let stat;
|
|
2682
|
+
try {
|
|
2683
|
+
stat = fs.lstatSync(absolute);
|
|
2684
|
+
} catch {
|
|
2685
|
+
throw Object.assign(new Error(`Answers file not found: ${file}`), { code: 'ANSWERS_NOT_FOUND' });
|
|
2686
|
+
}
|
|
2687
|
+
if (stat.isSymbolicLink() || !stat.isFile()) {
|
|
2688
|
+
throw Object.assign(new Error('Answers file must be a regular file'), { code: 'UNSAFE_ANSWERS_PATH' });
|
|
2689
|
+
}
|
|
2690
|
+
try {
|
|
2691
|
+
return JSON.parse(fs.readFileSync(absolute, 'utf8'));
|
|
2692
|
+
} catch (cause) {
|
|
2693
|
+
throw Object.assign(new Error(`Answers JSON is invalid: ${cause.message}`), { code: 'INVALID_ANSWERS' });
|
|
2694
|
+
}
|
|
2695
|
+
}
|
|
2696
|
+
|
|
2697
|
+
function workflowMode() {
|
|
2698
|
+
const explicit = argValue('--mode');
|
|
2699
|
+
if (explicit) return explicit;
|
|
2700
|
+
try {
|
|
2701
|
+
const current = getCurrentRun(CWD);
|
|
2702
|
+
if (current && !TERMINAL_STATES.has(current.state)) return current.mode;
|
|
2703
|
+
} catch {
|
|
2704
|
+
// No run is not a mode.
|
|
2705
|
+
}
|
|
2706
|
+
return null;
|
|
2707
|
+
}
|
|
2708
|
+
|
|
2709
|
+
function handleWorkflow() {
|
|
2710
|
+
const action = process.argv[3] || '';
|
|
2711
|
+
const json = process.argv.includes('--json');
|
|
2712
|
+
const personal = require('../lib/personal-workflow.js');
|
|
2713
|
+
try {
|
|
2714
|
+
if (action === 'interview') {
|
|
2715
|
+
const result = personal.interviewQuestions();
|
|
2716
|
+
console.log(json ? JSON.stringify({ ok: true, ...result }) : JSON.stringify({ ok: true, ...result }, null, 2));
|
|
2717
|
+
return;
|
|
2718
|
+
}
|
|
2719
|
+
if (action === 'propose') {
|
|
2720
|
+
const result = personal.proposeFromAnswers(readAnswersFile(argValue('--answers')), { rootDir: CWD });
|
|
2721
|
+
console.log(JSON.stringify({
|
|
2722
|
+
ok: true,
|
|
2723
|
+
id: result.proposal.id,
|
|
2724
|
+
status: result.proposal.status,
|
|
2725
|
+
valid: true,
|
|
2726
|
+
diff: result.diff
|
|
2727
|
+
}));
|
|
2728
|
+
return;
|
|
2729
|
+
}
|
|
2730
|
+
if (action === 'accept' || action === 'reject') {
|
|
2731
|
+
const proposalId = argValue('--proposal');
|
|
2732
|
+
if (!proposalId) throw Object.assign(new Error(`workflow ${action} requires --proposal <id>`), { code: 'PROPOSAL_REQUIRED' });
|
|
2733
|
+
const result = action === 'accept'
|
|
2734
|
+
? personal.acceptProposal({ proposalId, by: argValue('--by'), mode: workflowMode() })
|
|
2735
|
+
: personal.rejectProposal({ proposalId, by: argValue('--by'), mode: workflowMode() });
|
|
2736
|
+
console.log(JSON.stringify({
|
|
2737
|
+
ok: true,
|
|
2738
|
+
id: result.proposal.id,
|
|
2739
|
+
status: result.proposal.status,
|
|
2740
|
+
...(result.recipe ? { recipeId: result.recipe.id, version: result.recipe.version, replacedVersion: result.replacedVersion } : {})
|
|
2741
|
+
}));
|
|
2742
|
+
return;
|
|
2743
|
+
}
|
|
2744
|
+
if (action === 'suggest') {
|
|
2745
|
+
const result = personal.suggestProposals({ rootDir: CWD });
|
|
2746
|
+
console.log(JSON.stringify({
|
|
2747
|
+
ok: true,
|
|
2748
|
+
historyCount: result.historyCount,
|
|
2749
|
+
proposals: result.proposals.map((item) => ({
|
|
2750
|
+
id: item.id,
|
|
2751
|
+
kind: item.kind,
|
|
2752
|
+
status: item.status,
|
|
2753
|
+
evidence: item.evidence
|
|
2754
|
+
}))
|
|
2755
|
+
}));
|
|
2756
|
+
return;
|
|
2757
|
+
}
|
|
2758
|
+
throw Object.assign(new Error('Unknown workflow action. Use interview|propose|accept|reject|suggest'), { code: 'UNKNOWN_WORKFLOW_ACTION' });
|
|
2759
|
+
} catch (err) {
|
|
2760
|
+
workflowFail(err, json);
|
|
2761
|
+
}
|
|
2762
|
+
}
|
|
2763
|
+
|
|
2607
2764
|
function handleRecipe() {
|
|
2608
2765
|
const action = process.argv[3] || 'list';
|
|
2609
2766
|
const id = process.argv[4];
|
|
@@ -2918,6 +3075,7 @@ function handleGoal() {
|
|
|
2918
3075
|
actor: argValue('--by') || 'human',
|
|
2919
3076
|
message: 'Goal Contract frozen'
|
|
2920
3077
|
});
|
|
3078
|
+
ensureImpactMap(runId);
|
|
2921
3079
|
} else {
|
|
2922
3080
|
const error = new Error('goal must be show, draft, or freeze');
|
|
2923
3081
|
error.code = 'INVALID_USAGE';
|
|
@@ -3125,6 +3283,7 @@ async function handleVerify() {
|
|
|
3125
3283
|
return;
|
|
3126
3284
|
}
|
|
3127
3285
|
const run = executeGate('verification', runId);
|
|
3286
|
+
const drift = impactMap.loadImpactDrift(CWD, runId) || { mapStatus: 'MISSING', outOfMap: [] };
|
|
3128
3287
|
const compact = compactRun(CWD, { runId });
|
|
3129
3288
|
let claimed = { path: '.ai-engineering-loop/tasks/claimed-vs-reality.md', created: false };
|
|
3130
3289
|
try {
|
|
@@ -3139,6 +3298,7 @@ async function handleVerify() {
|
|
|
3139
3298
|
tier: run.tier ? run.tier.current : null,
|
|
3140
3299
|
commands,
|
|
3141
3300
|
testCounts: recorded.summary.testCounts,
|
|
3301
|
+
impactDrift: { mapStatus: drift.mapStatus, outOfMap: drift.outOfMap },
|
|
3142
3302
|
compact: compact.paths,
|
|
3143
3303
|
claimedVsReality: claimed,
|
|
3144
3304
|
next: run.tier && run.tier.current === 'T0'
|
|
@@ -3150,6 +3310,11 @@ async function handleVerify() {
|
|
|
3150
3310
|
log.success('✓ verification recorded and gated');
|
|
3151
3311
|
for (const c of commands) console.log(`- ${c.command}: exit ${c.exitCode}`);
|
|
3152
3312
|
console.log(`- State: ${run.state}`);
|
|
3313
|
+
if (drift.outOfMap.length) {
|
|
3314
|
+
console.log(`! ${drift.outOfMap.length} file(s) changed outside the ${drift.mapStatus} impact map (LOW finding at review): ${drift.outOfMap.join(', ')}`);
|
|
3315
|
+
} else if (drift.mapStatus === 'MISSING') {
|
|
3316
|
+
console.log('! No impact map for this Run; drift was not checked');
|
|
3317
|
+
}
|
|
3153
3318
|
console.log(`- Claimed-vs-reality: ${claimed.path}${claimed.created ? ' (scaffolded; fill it from the command output)' : ''}`);
|
|
3154
3319
|
console.log(`- Next: ${result.next}`);
|
|
3155
3320
|
}
|
|
@@ -3158,6 +3323,137 @@ async function handleVerify() {
|
|
|
3158
3323
|
}
|
|
3159
3324
|
}
|
|
3160
3325
|
|
|
3326
|
+
function readGuardInput() {
|
|
3327
|
+
const source = argValue('--input') || (process.argv.includes('--hook') ? '-' : null);
|
|
3328
|
+
if (source == null) {
|
|
3329
|
+
const target = argValue('--path');
|
|
3330
|
+
const commandText = argValue('--command');
|
|
3331
|
+
if (target == null && commandText == null) {
|
|
3332
|
+
throw Object.assign(new Error('guard check needs --input -|<file>, --path <file>, or --command <text>'), { code: 'INVALID_USAGE' });
|
|
3333
|
+
}
|
|
3334
|
+
return target != null ? { path: target } : { command: commandText };
|
|
3335
|
+
}
|
|
3336
|
+
const text = source === '-' ? fs.readFileSync(0, 'utf8') : fs.readFileSync(path.resolve(CWD, source), 'utf8');
|
|
3337
|
+
try {
|
|
3338
|
+
return JSON.parse(text);
|
|
3339
|
+
} catch (cause) {
|
|
3340
|
+
throw Object.assign(new Error(`guard input is not JSON: ${cause.message}`), { code: 'INVALID_USAGE' });
|
|
3341
|
+
}
|
|
3342
|
+
}
|
|
3343
|
+
|
|
3344
|
+
function handleGuard() {
|
|
3345
|
+
const action = args[1] && !args[1].startsWith('--') ? args[1] : 'check';
|
|
3346
|
+
const json = process.argv.includes('--json');
|
|
3347
|
+
const hook = argValue('--hook');
|
|
3348
|
+
try {
|
|
3349
|
+
if (action !== 'check') throw Object.assign(new Error('guard supports: check'), { code: 'INVALID_USAGE' });
|
|
3350
|
+
if (hook != null && hook !== 'claude') throw Object.assign(new Error('--hook supports: claude'), { code: 'INVALID_USAGE' });
|
|
3351
|
+
const result = guard.checkGuard(CWD, { tool: argValue('--tool'), input: readGuardInput(), runId: argValue('--run') });
|
|
3352
|
+
if (hook === 'claude') {
|
|
3353
|
+
const output = guard.claudeHookOutput(result);
|
|
3354
|
+
if (output) console.log(JSON.stringify(output));
|
|
3355
|
+
return;
|
|
3356
|
+
}
|
|
3357
|
+
if (json) console.log(JSON.stringify({ ok: true, ...result }));
|
|
3358
|
+
else {
|
|
3359
|
+
const mark = { allow: '✓', warn: '!', block: '✗' }[result.decision];
|
|
3360
|
+
console.log(`${mark} ${result.decision} (${result.tool}, ${result.mode}${result.runId ? `, run ${result.runId}` : ', no active run'})`);
|
|
3361
|
+
for (const reason of result.reasons) console.log(`- ${reason}`);
|
|
3362
|
+
}
|
|
3363
|
+
if (result.decision === 'block') process.exitCode = 2;
|
|
3364
|
+
} catch (err) {
|
|
3365
|
+
if (hook) {
|
|
3366
|
+
console.error(`ai-engineering-loop guard: ${err.message}`);
|
|
3367
|
+
process.exitCode = 1;
|
|
3368
|
+
return;
|
|
3369
|
+
}
|
|
3370
|
+
stageJsonError(err, json, 'GUARD_FAILED');
|
|
3371
|
+
}
|
|
3372
|
+
}
|
|
3373
|
+
|
|
3374
|
+
function handleHooks() {
|
|
3375
|
+
const action = args[1] && !args[1].startsWith('--') ? args[1] : 'status';
|
|
3376
|
+
const json = process.argv.includes('--json');
|
|
3377
|
+
try {
|
|
3378
|
+
const options = { host: argValue('--host') || 'claude' };
|
|
3379
|
+
let result;
|
|
3380
|
+
if (action === 'install') result = guardHooks.installHooks(CWD, options);
|
|
3381
|
+
else if (action === 'uninstall') result = guardHooks.uninstallHooks(CWD, options);
|
|
3382
|
+
else if (action === 'status') result = guardHooks.hooksStatus(CWD, options);
|
|
3383
|
+
else throw Object.assign(new Error('hooks must be install, uninstall, or status'), { code: 'INVALID_USAGE' });
|
|
3384
|
+
if (json) console.log(JSON.stringify({ ok: true, action, ...result }));
|
|
3385
|
+
else {
|
|
3386
|
+
const verb = action === 'status' ? '' : (result.changed ? ' (updated)' : ' (already up to date)');
|
|
3387
|
+
log.success(`✓ ${result.host} guard hooks ${result.installed ? 'installed' : 'not installed'} in ${result.settings}${verb}`);
|
|
3388
|
+
if (result.installed) console.log('- Edit/Write outside the Run scope and destructive Bash commands: ASSISTED asks, UNATTENDED blocks.');
|
|
3389
|
+
}
|
|
3390
|
+
} catch (err) {
|
|
3391
|
+
stageJsonError(err, json, 'HOOKS_FAILED');
|
|
3392
|
+
}
|
|
3393
|
+
}
|
|
3394
|
+
|
|
3395
|
+
function handleImpact() {
|
|
3396
|
+
const action = args[1] && !args[1].startsWith('--') ? args[1] : 'show';
|
|
3397
|
+
const json = process.argv.includes('--json');
|
|
3398
|
+
try {
|
|
3399
|
+
const runId = requiredRunId('impact');
|
|
3400
|
+
let result;
|
|
3401
|
+
if (action === 'map') {
|
|
3402
|
+
const limit = argValue('--limit');
|
|
3403
|
+
result = impactMap.generateImpactMap(CWD, {
|
|
3404
|
+
runId,
|
|
3405
|
+
force: process.argv.includes('--force'),
|
|
3406
|
+
...(limit == null ? {} : { limit: parsePositiveIntegerArg('--limit', limit) })
|
|
3407
|
+
});
|
|
3408
|
+
} else if (action === 'confirm') {
|
|
3409
|
+
result = impactMap.confirmImpactMap(CWD, {
|
|
3410
|
+
runId,
|
|
3411
|
+
by: argValue('--by'),
|
|
3412
|
+
allow: repeatedArg('--allow'),
|
|
3413
|
+
remove: repeatedArg('--remove')
|
|
3414
|
+
});
|
|
3415
|
+
} else if (action === 'skip') {
|
|
3416
|
+
result = impactMap.skipImpactMap(CWD, { runId, by: argValue('--by'), reason: argValue('--reason') });
|
|
3417
|
+
} else if (action === 'show') {
|
|
3418
|
+
const map = impactMap.loadImpactMap(CWD, runId);
|
|
3419
|
+
if (!map) throw Object.assign(new Error('No impact map yet. Run `impact map` first.'), { code: 'IMPACT_MAP_MISSING' });
|
|
3420
|
+
result = { map, path: `.ai-engineering-loop/runs/${runId}/${impactMap.IMPACT_MAP_FILE}` };
|
|
3421
|
+
} else if (action === 'drift') {
|
|
3422
|
+
const drift = impactMap.detectImpactDrift(CWD, { runId });
|
|
3423
|
+
result = { map: null, drift: drift.drift, path: drift.path };
|
|
3424
|
+
} else {
|
|
3425
|
+
throw Object.assign(new Error('impact must be map, show, confirm, skip, or drift'), { code: 'INVALID_USAGE' });
|
|
3426
|
+
}
|
|
3427
|
+
if (json) {
|
|
3428
|
+
console.log(JSON.stringify({ ok: true, runId, action, path: result.path, ...(result.map ? { map: result.map } : {}), ...(result.drift ? { drift: result.drift } : {}) }, null, 2));
|
|
3429
|
+
return;
|
|
3430
|
+
}
|
|
3431
|
+
if (result.drift) {
|
|
3432
|
+
log.success(`✓ impact drift checked (${result.drift.mapStatus})`);
|
|
3433
|
+
console.log(`- Changed: ${result.drift.changedFiles.length}, outside the map: ${result.drift.outOfMap.length}`);
|
|
3434
|
+
for (const file of result.drift.outOfMap) console.log(` ! ${file}`);
|
|
3435
|
+
return;
|
|
3436
|
+
}
|
|
3437
|
+
const { map } = result;
|
|
3438
|
+
log.success(`✓ impact map ${map.status} (${result.path})`);
|
|
3439
|
+
if (map.status === 'SKIPPED') {
|
|
3440
|
+
console.log(`- Skipped: ${map.skipReason}`);
|
|
3441
|
+
return;
|
|
3442
|
+
}
|
|
3443
|
+
for (const key of ['files', 'modules', 'endpoints', 'dbTables', 'tests']) {
|
|
3444
|
+
console.log(`- ${key}: ${map[key].length}`);
|
|
3445
|
+
for (const entry of map[key].slice(0, 10)) {
|
|
3446
|
+
const label = entry.route ? `${entry.method} ${entry.route} (${entry.path})` : (entry.path && key !== 'dbTables' ? entry.path : entry.name);
|
|
3447
|
+
console.log(` ${label} [${entry.confidence}] ${entry.reason}`);
|
|
3448
|
+
}
|
|
3449
|
+
}
|
|
3450
|
+
if (map.status === 'CONFIRMED') console.log(`- Allowed scope: ${map.scope.allowedPaths.length} path(s), confirmed by ${map.confirmedBy}`);
|
|
3451
|
+
else console.log('- Next: confirm with `impact confirm --by <name> [--allow <glob>]... [--remove <path>]...` or skip a trivial goal with `impact skip --by <name> --reason <text>`');
|
|
3452
|
+
} catch (err) {
|
|
3453
|
+
stageJsonError(err, json, 'IMPACT_FAILED');
|
|
3454
|
+
}
|
|
3455
|
+
}
|
|
3456
|
+
|
|
3161
3457
|
function handleReviewPack() {
|
|
3162
3458
|
const json = process.argv.includes('--json');
|
|
3163
3459
|
try {
|
|
@@ -3183,6 +3479,7 @@ function handleReviewPack() {
|
|
|
3183
3479
|
estimatedTokens: pack.estimatedTokens,
|
|
3184
3480
|
truncated: pack.truncated,
|
|
3185
3481
|
unresolvedFindings: pack.unresolvedFindings,
|
|
3482
|
+
impactDrift: pack.impactDrift,
|
|
3186
3483
|
fastPath: { eligible: fastPath.eligible, reasons: fastPath.reasons },
|
|
3187
3484
|
escalation: { requiredTier: escalation.requiredTier, reasons: escalation.reasons.map((r) => r.code) }
|
|
3188
3485
|
};
|
|
@@ -3192,6 +3489,9 @@ function handleReviewPack() {
|
|
|
3192
3489
|
console.log(`- Pack: ${result.packPath} (${result.reviewProfile}, ~${result.estimatedTokens} tokens)`);
|
|
3193
3490
|
console.log(`- Tier: ${tier || 'none'}; Devil's Advocate ${result.devilAdvocateRequired ? 'required' : 'not used (T0)'}`);
|
|
3194
3491
|
console.log(`- Judge tier floor: ${result.escalation.requiredTier}${result.escalation.reasons.length ? ` (${result.escalation.reasons.join(', ')})` : ''}`);
|
|
3492
|
+
if (result.impactDrift?.outOfMap.length) {
|
|
3493
|
+
console.log(`- Outside the impact map: ${result.impactDrift.outOfMap.join(', ')} (check their side effects)`);
|
|
3494
|
+
}
|
|
3195
3495
|
}
|
|
3196
3496
|
} catch (err) {
|
|
3197
3497
|
stageJsonError(err, json, 'REVIEW_PACK_FAILED');
|
|
@@ -3248,28 +3548,160 @@ function handleDeliver() {
|
|
|
3248
3548
|
if (!delivery && !(destination && summary)) {
|
|
3249
3549
|
throw Object.assign(new Error('Pass --destination <where> and --summary <text>, or write delivery.json first.'), { code: 'INVALID_USAGE' });
|
|
3250
3550
|
}
|
|
3551
|
+
// Proof-carrying receipt (#70): the final deliver step, after approval.
|
|
3552
|
+
// Missing gate artifacts leave no receipt; the delivery gate then reports them.
|
|
3553
|
+
let receipt = null;
|
|
3554
|
+
try {
|
|
3555
|
+
receipt = receipts.writeReceipt(CWD, {
|
|
3556
|
+
runId,
|
|
3557
|
+
host: argValue('--host'),
|
|
3558
|
+
model: argValue('--model'),
|
|
3559
|
+
head: argValue('--head'),
|
|
3560
|
+
evidenceLinks: repeatedArg('--evidence'),
|
|
3561
|
+
aelVersion: VERSION
|
|
3562
|
+
});
|
|
3563
|
+
} catch (err) {
|
|
3564
|
+
if (err.code !== 'RECEIPT_INPUT_MISSING') throw err;
|
|
3565
|
+
}
|
|
3251
3566
|
delivery = {
|
|
3252
3567
|
...(delivery || {}),
|
|
3253
3568
|
schemaVersion: 1,
|
|
3254
3569
|
runId,
|
|
3255
3570
|
...(destination ? { destination } : {}),
|
|
3256
3571
|
...(summary ? { summary } : {}),
|
|
3257
|
-
...(approve ? { humanApproved: true, approvedBy: by } : {})
|
|
3572
|
+
...(approve ? { humanApproved: true, approvedBy: by } : {}),
|
|
3573
|
+
...(receipt ? { receipt: { path: receipt.path, digest: receipt.receipt.digest } } : {})
|
|
3258
3574
|
};
|
|
3259
3575
|
fs.writeFileSync(deliveryPath, `${JSON.stringify(delivery, null, 2)}\n`, { mode: 0o600 });
|
|
3260
3576
|
const next = executeGate('delivery', runId);
|
|
3261
|
-
const result = {
|
|
3577
|
+
const result = {
|
|
3578
|
+
ok: true,
|
|
3579
|
+
runId,
|
|
3580
|
+
state: next.state,
|
|
3581
|
+
approvedNodes: approved,
|
|
3582
|
+
humanApproved: delivery.humanApproved === true,
|
|
3583
|
+
receipt: receipt && { path: receipt.path, markdown: receipt.markdownPath, digest: receipt.receipt.digest, missingEvidence: receipt.missing }
|
|
3584
|
+
};
|
|
3262
3585
|
if (json) console.log(JSON.stringify(result));
|
|
3263
3586
|
else {
|
|
3264
3587
|
log.success('✓ delivery gate passed');
|
|
3265
3588
|
console.log(`- State: ${next.state}`);
|
|
3266
3589
|
if (approved.length) console.log(`- Approved: ${approved.join(', ')} by ${by}`);
|
|
3590
|
+
if (receipt) console.log(`- Receipt: ${receipt.path} (digest ${receipt.receipt.digest.slice(0, 12)}); paste ${receipt.markdownPath} into the PR/MR body`);
|
|
3591
|
+
if (receipt?.missing.length) console.log(`! Missing AC evidence (❌ in the receipt): ${receipt.missing.join(', ')}`);
|
|
3267
3592
|
}
|
|
3268
3593
|
} catch (err) {
|
|
3269
3594
|
stageJsonError(err, json, 'DELIVER_FAILED');
|
|
3270
3595
|
}
|
|
3271
3596
|
}
|
|
3272
3597
|
|
|
3598
|
+
function handleVerifyReceipt() {
|
|
3599
|
+
const json = process.argv.includes('--json');
|
|
3600
|
+
try {
|
|
3601
|
+
const positional = args.slice(1).find((arg, index, list) => !arg.startsWith('--') && !['--run', '--head'].includes(list[index - 1]));
|
|
3602
|
+
const result = receipts.verifyReceipt(CWD, {
|
|
3603
|
+
runId: argValue('--run'),
|
|
3604
|
+
receiptPath: positional,
|
|
3605
|
+
head: argValue('--head')
|
|
3606
|
+
});
|
|
3607
|
+
if (json) console.log(JSON.stringify(result));
|
|
3608
|
+
else if (result.ok) {
|
|
3609
|
+
log.success(`✓ receipt verified (${result.path}, digest ${result.digest.slice(0, 12)})`);
|
|
3610
|
+
} else {
|
|
3611
|
+
log.error(`Receipt check failed (${result.path})`);
|
|
3612
|
+
for (const problem of result.problems) console.error(`- ${problem.code}: ${problem.detail}`);
|
|
3613
|
+
}
|
|
3614
|
+
if (!result.ok) process.exitCode = 1;
|
|
3615
|
+
} catch (err) {
|
|
3616
|
+
stageJsonError(err, json, 'VERIFY_RECEIPT_FAILED');
|
|
3617
|
+
}
|
|
3618
|
+
}
|
|
3619
|
+
|
|
3620
|
+
function handleBench() {
|
|
3621
|
+
const json = process.argv.includes('--json');
|
|
3622
|
+
const rest = process.argv.slice(3);
|
|
3623
|
+
try {
|
|
3624
|
+
if (rest[0] === 'compare') {
|
|
3625
|
+
const ids = [];
|
|
3626
|
+
for (let index = 1; index < rest.length; index += 1) {
|
|
3627
|
+
const arg = rest[index];
|
|
3628
|
+
if (arg === '--json') continue;
|
|
3629
|
+
if (arg.startsWith('--')) {
|
|
3630
|
+
throw Object.assign(new Error(`Unknown argument for bench compare: ${arg}`), { code: 'BENCH_USAGE' });
|
|
3631
|
+
}
|
|
3632
|
+
ids.push(arg);
|
|
3633
|
+
}
|
|
3634
|
+
if (ids.length !== 2) {
|
|
3635
|
+
throw Object.assign(new Error('bench compare needs exactly two bench ids'), { code: 'BENCH_USAGE' });
|
|
3636
|
+
}
|
|
3637
|
+
const comparison = compareBenches(CWD, ids[0], ids[1]);
|
|
3638
|
+
if (json) console.log(JSON.stringify(comparison));
|
|
3639
|
+
else process.stdout.write(formatCompare(comparison));
|
|
3640
|
+
return;
|
|
3641
|
+
}
|
|
3642
|
+
const modeArg = argValue('--mode');
|
|
3643
|
+
const repeatArg = argValue('--repeat');
|
|
3644
|
+
const timeoutArg = argValue('--timeout');
|
|
3645
|
+
const result = runBench({
|
|
3646
|
+
rootDir: CWD,
|
|
3647
|
+
home: homeDir(),
|
|
3648
|
+
realHome: os.homedir(),
|
|
3649
|
+
host: argValue('--host'),
|
|
3650
|
+
model: argValue('--model'),
|
|
3651
|
+
bin: argValue('--bin'),
|
|
3652
|
+
sandboxHome: process.argv.includes('--sandbox-home'),
|
|
3653
|
+
hostArgs: repeatedArg('--host-arg'),
|
|
3654
|
+
scenarioIds: repeatedArg('--scenario'),
|
|
3655
|
+
repeat: repeatArg == null ? 1 : parsePositiveIntegerArg('--repeat', repeatArg),
|
|
3656
|
+
timeoutMs: timeoutArg == null ? undefined : parseBenchTimeoutSeconds(timeoutArg),
|
|
3657
|
+
mode: modeArg == null ? 'ASSISTED' : normalizeMode(modeArg),
|
|
3658
|
+
args: rest,
|
|
3659
|
+
env: process.env,
|
|
3660
|
+
aelVersion: VERSION
|
|
3661
|
+
});
|
|
3662
|
+
const relativeDir = path.relative(CWD, result.directory).split(path.sep).join('/');
|
|
3663
|
+
if (json) {
|
|
3664
|
+
const { directory: _directory, ...body } = result;
|
|
3665
|
+
console.log(JSON.stringify({ ...body, directory: relativeDir }));
|
|
3666
|
+
return;
|
|
3667
|
+
}
|
|
3668
|
+
const rate = result.summary.complianceRate;
|
|
3669
|
+
log.success(`✓ bench ${result.benchId} (${result.target.host}/${result.target.model})`);
|
|
3670
|
+
console.log(`- Status: ${result.status}`);
|
|
3671
|
+
console.log(`- Compliance: ${rate == null ? 'unknown' : `${Math.round(rate * 1000) / 10}%`}`);
|
|
3672
|
+
console.log(`- Report: ${relativeDir}/report.md`);
|
|
3673
|
+
if (result.instructionViolations.length) {
|
|
3674
|
+
console.log(`- Top violation: ${result.instructionViolations[0].type} (×${result.instructionViolations[0].count})`);
|
|
3675
|
+
}
|
|
3676
|
+
} catch (err) {
|
|
3677
|
+
stageJsonError(err, json, 'BENCH_FAILED');
|
|
3678
|
+
}
|
|
3679
|
+
}
|
|
3680
|
+
|
|
3681
|
+
function handleLearn() {
|
|
3682
|
+
const json = process.argv.includes('--json');
|
|
3683
|
+
try {
|
|
3684
|
+
const maxAge = argValue('--max-age-runs');
|
|
3685
|
+
const result = learn(CWD, {
|
|
3686
|
+
dryRun: process.argv.includes('--dry-run'),
|
|
3687
|
+
prune: process.argv.includes('--prune'),
|
|
3688
|
+
...(maxAge == null ? {} : { maxAgeRuns: parsePositiveIntegerArg('--max-age-runs', maxAge) })
|
|
3689
|
+
});
|
|
3690
|
+
const { document } = result;
|
|
3691
|
+
if (json) {
|
|
3692
|
+
console.log(JSON.stringify({ ok: true, written: result.written, path: result.path, pruned: result.pruned, learnings: document }));
|
|
3693
|
+
return;
|
|
3694
|
+
}
|
|
3695
|
+
log.success(`✓ ${document.entries.length} learning(s) from ${document.sourceRuns} Run(s)${result.written ? ` → ${result.path}` : ' (dry run, nothing written)'}`);
|
|
3696
|
+
for (const entry of document.entries.slice(0, 10)) {
|
|
3697
|
+
console.log(`- ${entry.id} [${entry.axis}, ${entry.severity}, ×${entry.occurrences}] ${entry.lesson}`);
|
|
3698
|
+
}
|
|
3699
|
+
for (const item of result.pruned) console.log(` pruned ${item.id} (${item.reason})`);
|
|
3700
|
+
} catch (err) {
|
|
3701
|
+
stageJsonError(err, json, 'LEARN_FAILED');
|
|
3702
|
+
}
|
|
3703
|
+
}
|
|
3704
|
+
|
|
3273
3705
|
// Command: docs (read-only packaged specs)
|
|
3274
3706
|
function handleDocs() {
|
|
3275
3707
|
const json = process.argv.includes('--json');
|
|
@@ -3305,14 +3737,14 @@ Usage:
|
|
|
3305
3737
|
Commands:
|
|
3306
3738
|
init Bootstrap .ai-engineering-loop/ context from repository discovery
|
|
3307
3739
|
status Check the validity, readiness, and baseline freshness of context
|
|
3308
|
-
--json print context
|
|
3740
|
+
--json print context, host-skill freshness, and studioUrl (tokenless Studio identity; browser session comes from ael studio)
|
|
3309
3741
|
refresh Reconcile drifted context against repository non-destructively
|
|
3310
3742
|
run [task] Start or resume a stateful engineering run and instruct the agent (never writes to home)
|
|
3311
3743
|
--new start a new run even when another run is active
|
|
3312
3744
|
--mode report-only|assisted|unattended
|
|
3313
3745
|
--recipe <id> bind an immutable compiled workflow plan
|
|
3314
3746
|
--tier T0|T1|T2 --tier-reason <text> override the automatic tier (recorded)
|
|
3315
|
-
--json print runId, tier, tier reasons,
|
|
3747
|
+
--json print runId, tier, tier reasons, subagent count, recipe, workflowSource, and studioUrl (tokenless; browser session comes from ael studio)
|
|
3316
3748
|
--abandon [--run <id>] --reason <text> --by <name> [--yes] [--json]
|
|
3317
3749
|
close a stale run as CANCELLED (kept on disk) and release current.json;
|
|
3318
3750
|
ASSISTED asks to confirm, UNATTENDED needs --yes from a human
|
|
@@ -3384,21 +3816,56 @@ Commands:
|
|
|
3384
3816
|
--limit <n> bound query/related result count
|
|
3385
3817
|
--include-ignored include CSS, generated, lockfile, and .aelcontextignore paths
|
|
3386
3818
|
--json print only pack metadata; never print packed content
|
|
3387
|
-
eval Validate the
|
|
3819
|
+
eval Validate the 25-case production evaluation catalog
|
|
3388
3820
|
--results <dir> score host-generated JSON results
|
|
3389
3821
|
--cases <dir> use another compatible fixture catalog
|
|
3390
3822
|
--json print machine-readable metrics
|
|
3823
|
+
bench Compliance self-check for one chosen host and model
|
|
3824
|
+
[--host <id> --model <id>] [--scenario <id>]... [--repeat <n>]
|
|
3825
|
+
[--bin <path>] [--host-arg <arg>]... [--sandbox-home]
|
|
3826
|
+
[--mode assisted|unattended|report-only] [--timeout <seconds>] [--json]
|
|
3827
|
+
One target per invocation. Default mode is ASSISTED; the scenario file's
|
|
3828
|
+
scripted approver answers Goal freeze, impact confirm, and deliver --approve.
|
|
3829
|
+
--mode unattended is opt-in. Default timeout is 20 minutes.
|
|
3830
|
+
The host keeps the real HOME so it can authenticate. AEL_HOME is a temp dir.
|
|
3831
|
+
--bin and --sandbox-home give the host a temp HOME (tests use this).
|
|
3832
|
+
A missing binary is UNAVAILABLE, not a model failure.
|
|
3833
|
+
Progress lines go to stderr. Writes .ai-engineering-loop/bench/<benchId>/{results.json,report.md,hardening.md}
|
|
3834
|
+
bench compare <benchIdA> <benchIdB> [--json] read-only side-by-side table
|
|
3391
3835
|
escalation Compute the minimum Judge model tier from deterministic risk signals
|
|
3392
3836
|
--run <id> target a non-current run
|
|
3393
3837
|
--json print machine-readable reasons
|
|
3394
3838
|
verify Record verification commands, gate them, compact, and scaffold claimed-vs-reality
|
|
3395
3839
|
[--run <id>] [--cmd "<command>"]... [--shell] [--serial] [--timeout-ms <n>] [--json]
|
|
3396
3840
|
default command: test_unit from .ai-engineering-loop/verification.md; commands run in parallel
|
|
3841
|
+
impact Blast-radius map of the frozen Goal before Stage 3 (written to the Run on Goal freeze)
|
|
3842
|
+
map [--run <id>] [--limit <n>] [--force] [--json] files, modules, endpoints, DB tables, tests
|
|
3843
|
+
show [--run <id>] [--json]
|
|
3844
|
+
confirm --by <name> [--allow <glob>]... [--remove <path>]... [--run <id>] [--json]
|
|
3845
|
+
confirmed paths become the Run's allowed scope; verify reports changes outside it
|
|
3846
|
+
skip --by <name> --reason <text> [--run <id>] trivial goals (copy, CSS, comments)
|
|
3847
|
+
drift [--run <id>] [--json] compare the gated Maker diff with the map
|
|
3848
|
+
guard Runtime guardrail for one tool call: allow | warn | block (warn/block recorded in the Run)
|
|
3849
|
+
check --tool edit|bash (--input -|<file> | --path <file> | --command <text>) [--run <id>] [--json]
|
|
3850
|
+
check --hook claude read a Claude Code PreToolUse payload on stdin and answer deny/ask
|
|
3851
|
+
edit: outside Goal scope.allowedPaths + confirmed impact map, or in scope.deniedPaths
|
|
3852
|
+
bash: rm -rf, DROP TABLE/TRUNCATE, git push --force, git reset --hard, git clean -fd
|
|
3853
|
+
ASSISTED warns (the host asks), UNATTENDED and REPORT_ONLY block (exit 2)
|
|
3854
|
+
hooks Install native pre-tool hooks that call guard check (enforced hosts: Claude Code)
|
|
3855
|
+
install | uninstall | status [--host claude] [--json] writes .claude/settings.json
|
|
3397
3856
|
review-pack Build the Devil's Advocate pack: fast-path, diff hunks, escalation floor, tier
|
|
3398
3857
|
[--run <id>] [--base <git-ref>] [--profile lean|standard|thorough] [--json]
|
|
3399
3858
|
deliver Approve READY approval nodes, write delivery.json, and pass the delivery gate
|
|
3400
3859
|
[--run <id>] [--destination <where> --summary <text>] [--approve --by <name>] [--json]
|
|
3401
3860
|
ASSISTED requires --approve --by <name>, given only after explicit human approval
|
|
3861
|
+
[--host <id>] [--model <id>] [--head <sha>] [--evidence AC-N=<https-url>]...
|
|
3862
|
+
writes RUN_DIR/receipt.json + receipt.md (paste into the PR/MR body); digest stored in delivery.json
|
|
3863
|
+
verify-receipt Re-check a delivery receipt: artifact hashes, frozen Goal, AC evidence, verdict, head SHA
|
|
3864
|
+
[--run <id> | <repository-relative-receipt.json>] [--head <sha>] [--json] exit 1 on mismatch (CI)
|
|
3865
|
+
learn Mine past Runs' VALID findings into .ai-engineering-loop/learnings.json (+ .md) for context packs
|
|
3866
|
+
[--dry-run] [--prune] [--max-age-runs <n>] [--json]
|
|
3867
|
+
deterministic and deduped with runId provenance; --prune drops lessons whose files are gone
|
|
3868
|
+
or not seen in the last <n> Runs (default 20); maker / devil-advocate packs get the top entries
|
|
3402
3869
|
docs Print a packaged spec without needing the package source tree
|
|
3403
3870
|
list [--json]
|
|
3404
3871
|
<name> [--section <heading>] [--json] e.g. docs grill-policy --section "When to grill"
|
|
@@ -3415,6 +3882,13 @@ Commands:
|
|
|
3415
3882
|
projects Registry of local projects with .ai-engineering-loop/ (~/.ai-engineering-loop/projects.json)
|
|
3416
3883
|
list|add [dir]|remove [dir] [--json]
|
|
3417
3884
|
scan --root <dir> [--depth 4]
|
|
3885
|
+
workflow Build or adapt the per-user personal recipe (writes only under the AEL home)
|
|
3886
|
+
interview [--json]
|
|
3887
|
+
propose --answers <file> [--json]
|
|
3888
|
+
accept --proposal <id> --by <name> [--json]
|
|
3889
|
+
reject --proposal <id> --by <name> [--json]
|
|
3890
|
+
suggest [--json]
|
|
3891
|
+
UNATTENDED cannot accept or reject (HUMAN_REQUIRED)
|
|
3418
3892
|
recipe Inspect and compile declarative workflow recipes (does not execute nodes)
|
|
3419
3893
|
list [--json]
|
|
3420
3894
|
catalog [--json]
|
|
@@ -3426,7 +3900,7 @@ Commands:
|
|
|
3426
3900
|
show <id> [--json]
|
|
3427
3901
|
validate <id> [--mode <mode>] [--json]
|
|
3428
3902
|
explain | graph | simulate <id> [--mode <mode>] [--json]
|
|
3429
|
-
studio Open the localhost-only
|
|
3903
|
+
studio Open the localhost-only live-run Studio
|
|
3430
3904
|
[--port <0-65535>] [--no-open]
|
|
3431
3905
|
handoff Export or verify a redacted, integrity-bound knowledge transfer
|
|
3432
3906
|
create [--run <id>] [--audience developer|agent|auditor]
|
|
@@ -3494,7 +3968,7 @@ if (['init', 'run', 'status', 'refresh'].includes(command)) {
|
|
|
3494
3968
|
maybeAutoUpdate();
|
|
3495
3969
|
}
|
|
3496
3970
|
|
|
3497
|
-
if (!['setup', 'hosts', 'sync-hosts', '-h', '--help', '-v', '--version'].includes(command)) maybeSetupHint();
|
|
3971
|
+
if (!['setup', 'hosts', 'sync-hosts', 'guard', 'bench', '-h', '--help', '-v', '--version'].includes(command)) maybeSetupHint();
|
|
3498
3972
|
if (command === 'status') maybeNewHostsNotice();
|
|
3499
3973
|
|
|
3500
3974
|
switch (command) {
|
|
@@ -3555,6 +4029,9 @@ switch (command) {
|
|
|
3555
4029
|
case 'eval':
|
|
3556
4030
|
handleEval();
|
|
3557
4031
|
break;
|
|
4032
|
+
case 'bench':
|
|
4033
|
+
handleBench();
|
|
4034
|
+
break;
|
|
3558
4035
|
case 'escalation':
|
|
3559
4036
|
handleEscalation();
|
|
3560
4037
|
break;
|
|
@@ -3564,9 +4041,24 @@ switch (command) {
|
|
|
3564
4041
|
case 'verify':
|
|
3565
4042
|
handleVerify();
|
|
3566
4043
|
break;
|
|
4044
|
+
case 'guard':
|
|
4045
|
+
handleGuard();
|
|
4046
|
+
break;
|
|
4047
|
+
case 'hooks':
|
|
4048
|
+
handleHooks();
|
|
4049
|
+
break;
|
|
4050
|
+
case 'impact':
|
|
4051
|
+
handleImpact();
|
|
4052
|
+
break;
|
|
3567
4053
|
case 'review-pack':
|
|
3568
4054
|
handleReviewPack();
|
|
3569
4055
|
break;
|
|
4056
|
+
case 'learn':
|
|
4057
|
+
handleLearn();
|
|
4058
|
+
break;
|
|
4059
|
+
case 'verify-receipt':
|
|
4060
|
+
handleVerifyReceipt();
|
|
4061
|
+
break;
|
|
3570
4062
|
case 'deliver':
|
|
3571
4063
|
handleDeliver();
|
|
3572
4064
|
break;
|
|
@@ -3579,6 +4071,9 @@ switch (command) {
|
|
|
3579
4071
|
case 'migrate':
|
|
3580
4072
|
handleMigrate();
|
|
3581
4073
|
break;
|
|
4074
|
+
case 'workflow':
|
|
4075
|
+
handleWorkflow();
|
|
4076
|
+
break;
|
|
3582
4077
|
case 'recipe':
|
|
3583
4078
|
handleRecipe();
|
|
3584
4079
|
break;
|