Compare commits
| Author | SHA1 | Date | |
|---|---|---|---|
 jason.woltje
|
3100dfd66d |
@@ -22,9 +22,9 @@ steps:
|
||||
image: gcr.io/kaniko-project/executor:debug
|
||||
environment:
|
||||
REGISTRY_USER:
|
||||
from_secret: REGISTRY_USERNAME
|
||||
from_secret: gitea_username
|
||||
REGISTRY_PASS:
|
||||
from_secret: REGISTRY_PASSWORD
|
||||
from_secret: gitea_password
|
||||
CI_COMMIT_BRANCH: ${CI_COMMIT_BRANCH}
|
||||
CI_COMMIT_TAG: ${CI_COMMIT_TAG}
|
||||
CI_COMMIT_SHA: ${CI_COMMIT_SHA}
|
||||
|
||||
@@ -270,9 +270,9 @@ steps:
|
||||
when: *image_build_when
|
||||
environment:
|
||||
REGISTRY_USER:
|
||||
from_secret: REGISTRY_USERNAME
|
||||
from_secret: gitea_username
|
||||
REGISTRY_PASS:
|
||||
from_secret: REGISTRY_PASSWORD
|
||||
from_secret: gitea_password
|
||||
CI_COMMIT_BRANCH: ${CI_COMMIT_BRANCH}
|
||||
CI_COMMIT_TAG: ${CI_COMMIT_TAG}
|
||||
CI_COMMIT_SHA: ${CI_COMMIT_SHA}
|
||||
@@ -306,9 +306,9 @@ steps:
|
||||
when: *main_image_build_when
|
||||
environment:
|
||||
REGISTRY_USER:
|
||||
from_secret: REGISTRY_USERNAME
|
||||
from_secret: gitea_username
|
||||
REGISTRY_PASS:
|
||||
from_secret: REGISTRY_PASSWORD
|
||||
from_secret: gitea_password
|
||||
CI_COMMIT_BRANCH: ${CI_COMMIT_BRANCH}
|
||||
CI_COMMIT_TAG: ${CI_COMMIT_TAG}
|
||||
CI_COMMIT_SHA: ${CI_COMMIT_SHA}
|
||||
@@ -333,9 +333,9 @@ steps:
|
||||
when: *main_image_build_when
|
||||
environment:
|
||||
REGISTRY_USER:
|
||||
from_secret: REGISTRY_USERNAME
|
||||
from_secret: gitea_username
|
||||
REGISTRY_PASS:
|
||||
from_secret: REGISTRY_PASSWORD
|
||||
from_secret: gitea_password
|
||||
CI_COMMIT_BRANCH: ${CI_COMMIT_BRANCH}
|
||||
CI_COMMIT_TAG: ${CI_COMMIT_TAG}
|
||||
CI_COMMIT_SHA: ${CI_COMMIT_SHA}
|
||||
|
||||
@@ -138,9 +138,9 @@ mosaic brain tasks
|
||||
mosaic brain conversations
|
||||
|
||||
# Agent forge pipeline
|
||||
mosaic forge run [--simulate] # fails closed (FORGE_NO_EXECUTOR) with no executor wired; --simulate for typed simulated runs
|
||||
mosaic forge run
|
||||
mosaic forge status
|
||||
mosaic forge resume [--simulate] # same fail-closed rule as forge run
|
||||
mosaic forge resume
|
||||
mosaic forge personas
|
||||
|
||||
# Structured logging
|
||||
|
||||
@@ -190,13 +190,7 @@ beforeEach((ctx) => {
|
||||
});
|
||||
|
||||
afterAll(async () => {
|
||||
// Cleanup only when the fixture actually installed rows. `handle` is set
|
||||
// before the first query (createDb connects lazily), so on an unreachable
|
||||
// database `handle` is truthy while nothing was inserted — cleanup must
|
||||
// honor `dbAvailable` or the skip path fails the file with ECONNREFUSED in
|
||||
// afterAll (caught live by the publish pipeline's no-DATABASE_URL verify
|
||||
// step, pipeline 2486).
|
||||
if (!handle || !dbAvailable) return;
|
||||
if (!handle) return;
|
||||
const db = handle.db;
|
||||
|
||||
// Delete in dependency order (FK constraints)
|
||||
|
||||
@@ -15,18 +15,18 @@
|
||||
> `done` requires: repo quality gates green, independent review recorded, terminal-green CI on
|
||||
> the PR head, squash merge to `next`, and acceptance evidence in notes.
|
||||
|
||||
| id | status | description | issue | agent | repo | branch | depends_on | estimate | notes |
|
||||
| -------- | ----------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- | ---------- | ----------------- | --------------------------------- | ---------------------------------------------------------- | -------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| RI-0-001 | done | Bootstrap: issue #1275, PRD section, this DAG, scratchpad (docs only) | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-mission-bootstrap | — | 6K | PR #1276 (head 758659dd): docs-only, CI green (2475). Review requested from fargo. Merges first (no publish run). |
|
||||
| RI-1-001 | done | RI-N1: canonical terminal verification command + publish-pipeline exact-commit gate (every publish step depends on verify; commit identity check; fail closed) | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-publish-gate | RI-0-001 | 25K | PR #1277 (head 46784c8d): CI GREEN at head after serialized retry (pipeline 2476, 2026-08-18) - earlier red was CI-agent contention (web SPA timeouts under concurrent pipelines), not code. Review requested from fargo at pinned head (comms 20260818T021025Z). |
|
||||
| RI-1-002 | done | RI-N1 negative control: checked-in tests proving a broken mandatory check blocks every publish step and that DAG edges cannot be bypassed | #1275 | pi-glm-5.3 | mosaicstack/stack | test/ri-050-publish-gate-negative | RI-1-001 | 12K | |
|
||||
| RI-2-001 | done | RI-N2 (Forge): remove stub-executor false success; `--simulate` typed `simulated` results that satisfy nothing; literal-`true` gates and echo-review replaced with real gates or typed waiting-for-authority | #1275 | pi-glm-5.3 | mosaicstack/stack | fix/ri-050-forge-fail-closed | RI-0-001 | 20K | Independent review APPROVED 2026-08-17 (Gitea review 172 on PR #1278, head 99b8f6ea; reviewing seat fargo — recorded under shared host principal mos-dt-0, provenance correction posted by fred; wrapper gap filed by fred). Executed at head: forge tests 116/116, lint green, typecheck green after building macp dist (minimal-install artifact, not a defect), workspace typecheck 45/45, no external type consumers of the changed interfaces. CI red = known lane-wide fleet-test failure only, carries no information about this change (fred, log-content analysis, pipelines 2456-2458). Non-blocking finding: README L141-143 + skills/mosaic-forge/SKILL.md document bare forge run/resume, which now fails closed — fast-follow docs touch. Merge queued behind #1270. UPDATE 2026-08-18: #1270 merged; CI GREEN at head 4917df1f via serialized retry (pipeline 2477) - root cause of prior reds was CI-agent contention (web SPA timeouts under concurrent pipelines), superseding the fleet-test-failure theory. |
|
||||
| RI-2-002 | done | RI-N2 (MACP): gate runner fails closed on empty commands, stub executors, and unimplemented CI-provider gates unless explicit simulate; typed capability failures | #1275 | pi-glm-5.3 | mosaicstack/stack | fix/ri-050-macp-fail-closed | RI-0-001 | 15K | PR #1293 (head 2097379e): CI green (pipeline 2465), independent review APPROVED (Gitea review 173, jarvis seat, 2026-08-17) - macp 109/109 verified at head. Merge queued behind #1276/#1277/#1278. |
|
||||
| RI-3-001 | in-progress | RI-N4: complete probe inventory mapping every TS and shell quality-rail check to one canonical check with disposition (preserve/strengthen/retire, each named) | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-qr-probe-inventory | RI-0-001 | 12K | |
|
||||
| RI-3-002 | not-started | RI-N4: TS evaluator absorbs effective shell probes; typed results (passed/failed/blocked/error/not-applicable) with versioned digested check definitions; shell commands become thin adapters; contract/parity/negative-control tests | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-qr-evaluator | RI-3-001 | 30K | |
|
||||
| RI-4-001 | in-progress | RI-N3: one PRD application service — `mission --plan` persists mission↔PRD linkage (ids/versions/selected requirements); `mosaic prdy` routes through the service or becomes a named import/export adapter; Markdown is a labeled generated view; explicit conflict-aware import | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-prd-authority | RI-0-001 | 35K | PR #1294 (head 8d258e1d): CI green (pipeline 2466), independent review APPROVED (Gitea review 174, jarvis seat, 2026-08-17) - prdy 20/20 + command specs 9/9 at head. Merge queued behind #1276/#1277/#1278. |
|
||||
| RI-5-001 | done | RI-N5: typed freshness states (current/stale/partial/unknown/unavailable); no failed-fetch-renders-empty; stale derived verdicts → unknown; mutations disabled when stale; failure-matrix tests | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-web-stale-safety | RI-0-001 | 25K | |
|
||||
| RI-V-001 | not-started | Final verification + release evidence: all cards verified merged, negative controls demonstrated, real `next` publish run green on exact commit, evidence pack recorded | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-release-evidence | RI-1-002, RI-2-001, RI-2-002, RI-3-002, RI-4-001, RI-5-001 | 10K | |
|
||||
| id | status | description | issue | agent | repo | branch | depends_on | estimate | notes |
|
||||
| -------- | ----------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ----- | ---------- | ----------------- | --------------------------------- | ---------------------------------------------------------- | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
|
||||
| RI-0-001 | in-progress | Bootstrap: issue #1275, PRD section, this DAG, scratchpad (docs only) | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-mission-bootstrap | — | 6K | |
|
||||
| RI-1-001 | in-progress | RI-N1: canonical terminal verification command + publish-pipeline exact-commit gate (every publish step depends on verify; commit identity check; fail closed) | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-publish-gate | RI-0-001 | 25K | |
|
||||
| RI-1-002 | not-started | RI-N1 negative control: checked-in tests proving a broken mandatory check blocks every publish step and that DAG edges cannot be bypassed | #1275 | pi-glm-5.3 | mosaicstack/stack | test/ri-050-publish-gate-negative | RI-1-001 | 12K | |
|
||||
| RI-2-001 | in-progress | RI-N2 (Forge): remove stub-executor false success; `--simulate` typed `simulated` results that satisfy nothing; literal-`true` gates and echo-review replaced with real gates or typed waiting-for-authority | #1275 | pi-glm-5.3 | mosaicstack/stack | fix/ri-050-forge-fail-closed | RI-0-001 | 20K | Independent review APPROVED 2026-08-17 (Gitea review 172 on PR #1278, head 99b8f6ea; reviewing seat fargo — recorded under shared host principal mos-dt-0, provenance correction posted by fred; wrapper gap filed by fred). Executed at head: forge tests 116/116, lint green, typecheck green after building macp dist (minimal-install artifact, not a defect), workspace typecheck 45/45, no external type consumers of the changed interfaces. CI red = known lane-wide fleet-test failure only, carries no information about this change (fred, log-content analysis, pipelines 2456-2458). Non-blocking finding: README L141-143 + skills/mosaic-forge/SKILL.md document bare forge run/resume, which now fails closed — fast-follow docs touch. Merge queued behind #1270. |
|
||||
| RI-2-002 | in-progress | RI-N2 (MACP): gate runner fails closed on empty commands, stub executors, and unimplemented CI-provider gates unless explicit simulate; typed capability failures | #1275 | pi-glm-5.3 | mosaicstack/stack | fix/ri-050-macp-fail-closed | RI-0-001 | 15K | |
|
||||
| RI-3-001 | not-started | RI-N4: complete probe inventory mapping every TS and shell quality-rail check to one canonical check with disposition (preserve/strengthen/retire, each named) | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-qr-probe-inventory | RI-0-001 | 12K | |
|
||||
| RI-3-002 | not-started | RI-N4: TS evaluator absorbs effective shell probes; typed results (passed/failed/blocked/error/not-applicable) with versioned digested check definitions; shell commands become thin adapters; contract/parity/negative-control tests | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-qr-evaluator | RI-3-001 | 30K | |
|
||||
| RI-4-001 | in-progress | RI-N3: one PRD application service — `mission --plan` persists mission↔PRD linkage (ids/versions/selected requirements); `mosaic prdy` routes through the service or becomes a named import/export adapter; Markdown is a labeled generated view; explicit conflict-aware import | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-prd-authority | RI-0-001 | 35K | |
|
||||
| RI-5-001 | not-started | RI-N5: typed freshness states (current/stale/partial/unknown/unavailable); no failed-fetch-renders-empty; stale derived verdicts → unknown; mutations disabled when stale; failure-matrix tests | #1275 | pi-glm-5.3 | mosaicstack/stack | feat/ri-050-web-stale-safety | RI-0-001 | 25K | |
|
||||
| RI-V-001 | not-started | Final verification + release evidence: all cards verified merged, negative controls demonstrated, real `next` publish run green on exact commit, evidence pack recorded | #1275 | pi-glm-5.3 | mosaicstack/stack | docs/ri-050-release-evidence | RI-1-002, RI-2-001, RI-2-002, RI-3-002, RI-4-001, RI-5-001 | 10K | |
|
||||
|
||||
## Dispatch waves (max 2 parallel workers)
|
||||
|
||||
|
||||
@@ -34,7 +34,6 @@ export default tseslint.config(
|
||||
'packages/storage/vitest.config.ts',
|
||||
'packages/mosaic/vitest.config.ts',
|
||||
'packages/mosaic/__tests__/*.ts',
|
||||
'packages/forge/__tests__/*.ts',
|
||||
'tools/federation-harness/*.ts',
|
||||
],
|
||||
},
|
||||
|
||||
@@ -539,43 +539,3 @@ Not every brief needs full Board of Directors review. The classification system
|
||||
### Backward compatibility
|
||||
|
||||
Existing briefs without a `class` field are auto-classified. The default (no matching keywords) is `strategic`, so all existing runs get the full pipeline unless keywords trigger `technical`.
|
||||
|
||||
---
|
||||
|
||||
## Fail-Closed Execution & Explicit Simulation (SDLC-D-035)
|
||||
|
||||
**Added:** 2026-08-17
|
||||
|
||||
Forge fails closed when a required capability is missing. It never runs a
|
||||
pipeline with a stub executor and reports success.
|
||||
|
||||
### Normal mode (default)
|
||||
|
||||
- No task executor wired → the CLI exits nonzero with the typed capability
|
||||
error `FORGE_NO_EXECUTOR`. No run is created.
|
||||
- A stage whose gate is approval-based (board approval, planning approvals,
|
||||
remediation re-review, discovery/analysis attestations) records a typed
|
||||
`waiting-for-authority` stage result and raises `FORGE_AUTHORITY_REQUIRED`.
|
||||
It never passes vacuously.
|
||||
- A stage whose gate requires an unwired provider (AI reviewer, CI pipeline)
|
||||
records a typed `blocked` stage result and raises `FORGE_NO_REVIEWER` /
|
||||
`FORGE_NO_CI_PIPELINE`. The synthetic echo-review approval in `06-review`
|
||||
and all vacuous `true` gates were removed.
|
||||
|
||||
### Explicit simulation (`--simulate`)
|
||||
|
||||
Opts into stub/synthetic execution. Every stage result, every gate result, and
|
||||
the run manifest carry the distinct typed status `simulated` (manifest also
|
||||
records `mode: "simulated"`). `simulated` is a non-satisfying outcome:
|
||||
`isSatisfyingOutcome()` and all completion/gate consumers treat only `passed`
|
||||
as satisfying. The CLI exits 0 for a simulated run only because the caller
|
||||
explicitly passed `--simulate`, and prints a loud SIMULATED banner.
|
||||
|
||||
### Typed outcome model
|
||||
|
||||
Every gate/task outcome is one of the closed set
|
||||
`passed | failed | blocked | error | waiting-for-authority | simulated |
|
||||
not-applicable`, with the reason recorded on the stage status and each gate
|
||||
result in `manifest.json`. Missing implementations, missing gate evidence,
|
||||
unknown stages, process errors, and timeouts map to fail-closed members —
|
||||
never to `passed`.
|
||||
|
||||
@@ -1,319 +0,0 @@
|
||||
import fs from 'node:fs';
|
||||
import os from 'node:os';
|
||||
import path from 'node:path';
|
||||
import { describe, it, expect, beforeEach, afterEach } from 'vitest';
|
||||
|
||||
import { generateBoardTasks } from '../src/board-tasks.js';
|
||||
import { STAGE_SPECS } from '../src/constants.js';
|
||||
import { ForgeCapabilityError } from '../src/errors.js';
|
||||
import {
|
||||
evaluateStageGates,
|
||||
gateLabel,
|
||||
isCommandGate,
|
||||
isSatisfyingOutcome,
|
||||
} from '../src/outcomes.js';
|
||||
import { loadManifest, runPipeline } from '../src/pipeline-runner.js';
|
||||
import type { ForgeTask, ForgeTaskResult, TaskExecutor } from '../src/types.js';
|
||||
|
||||
/**
|
||||
* Mock real executor that returns typed results.
|
||||
*
|
||||
* Command gates are "verified" by the mock so normal-mode runs can pass
|
||||
* mechanically gated stages; authority/provider gates are never reported
|
||||
* because they have no mechanical implementation.
|
||||
*/
|
||||
function createTypedExecutor(options?: {
|
||||
failStage?: string;
|
||||
gateOutcomes?: Record<string, 'passed' | 'failed' | 'simulated' | 'error' | 'blocked'>;
|
||||
}): TaskExecutor & { submittedTasks: ForgeTask[] } {
|
||||
const submittedTasks: ForgeTask[] = [];
|
||||
return {
|
||||
submittedTasks,
|
||||
async submitTask(task: ForgeTask) {
|
||||
submittedTasks.push(task);
|
||||
},
|
||||
async waitForCompletion(taskId: string): Promise<ForgeTaskResult> {
|
||||
const task = submittedTasks.find((t) => t.id === taskId);
|
||||
const stageName = task?.metadata?.['stageName'] as string | undefined;
|
||||
|
||||
if (options?.failStage && stageName === options.failStage) {
|
||||
return {
|
||||
task_id: taskId,
|
||||
outcome: 'failed',
|
||||
reason: 'mock task failure',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 1,
|
||||
gate_results: [],
|
||||
};
|
||||
}
|
||||
|
||||
const gateResults = (task?.qualityGates ?? [])
|
||||
.filter((gate) => isCommandGate(gate))
|
||||
.map((gate) => {
|
||||
const label = gateLabel(gate);
|
||||
const outcome = options?.gateOutcomes?.[label] ?? 'passed';
|
||||
return {
|
||||
gate: label,
|
||||
outcome,
|
||||
reason: outcome === 'passed' ? 'mock verified' : `mock gate outcome: ${outcome}`,
|
||||
};
|
||||
});
|
||||
|
||||
return {
|
||||
task_id: taskId,
|
||||
outcome: 'passed',
|
||||
reason: 'mock verified',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: gateResults,
|
||||
};
|
||||
},
|
||||
async getTaskStatus() {
|
||||
return 'completed' as const;
|
||||
},
|
||||
};
|
||||
}
|
||||
|
||||
describe('fail-closed: no executor wired', () => {
|
||||
let tmpDir: string;
|
||||
let briefPath: string;
|
||||
|
||||
beforeEach(() => {
|
||||
tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'forge-failclosed-'));
|
||||
briefPath = path.join(tmpDir, 'brief.md');
|
||||
fs.writeFileSync(briefPath, '# Fix bug\n\nA bugfix for lint cleanup.');
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
fs.rmSync(tmpDir, { recursive: true, force: true });
|
||||
});
|
||||
|
||||
it('throws a typed FORGE_NO_EXECUTOR capability error without --simulate', async () => {
|
||||
await expect(
|
||||
runPipeline(briefPath, tmpDir, {
|
||||
// no executor, no simulate — must fail closed, never run with a stub
|
||||
stages: ['00-intake'],
|
||||
}),
|
||||
).rejects.toMatchObject({
|
||||
name: 'ForgeCapabilityError',
|
||||
code: 'FORGE_NO_EXECUTOR',
|
||||
capability: 'task-executor',
|
||||
});
|
||||
});
|
||||
|
||||
it('does not create a run directory when failing closed on a missing executor', async () => {
|
||||
try {
|
||||
await runPipeline(briefPath, tmpDir, { stages: ['00-intake'] });
|
||||
} catch {
|
||||
// expected
|
||||
}
|
||||
expect(fs.existsSync(path.join(tmpDir, '.forge', 'runs'))).toBe(false);
|
||||
});
|
||||
|
||||
it('completes with every result typed simulated when simulate is set', async () => {
|
||||
const result = await runPipeline(briefPath, tmpDir, {
|
||||
simulate: true,
|
||||
stages: ['00-intake', '00b-discovery', '02-planning-1', '06-review'],
|
||||
});
|
||||
|
||||
expect(result.manifest.mode).toBe('simulated');
|
||||
expect(result.manifest.status).toBe('simulated');
|
||||
|
||||
for (const stage of result.stages) {
|
||||
const stageStatus = result.manifest.stages[stage];
|
||||
expect(stageStatus?.status, `stage ${stage}`).toBe('simulated');
|
||||
expect(stageStatus?.status, `stage ${stage}`).not.toBe('passed');
|
||||
expect(stageStatus?.reason, `stage ${stage}`).toBeTruthy();
|
||||
for (const gateResult of stageStatus?.gateResults ?? []) {
|
||||
expect(gateResult.outcome, `gate ${gateResult.gate} of ${stage}`).toBe('simulated');
|
||||
expect(gateResult.outcome, `gate ${gateResult.gate} of ${stage}`).not.toBe('passed');
|
||||
}
|
||||
}
|
||||
|
||||
// The persisted manifest agrees.
|
||||
const persisted = loadManifest(result.runDir);
|
||||
expect(persisted.mode).toBe('simulated');
|
||||
expect(persisted.status).toBe('simulated');
|
||||
expect(persisted.stages['02-planning-1']?.status).toBe('simulated');
|
||||
});
|
||||
});
|
||||
|
||||
describe('fail-closed: typed outcome model', () => {
|
||||
it('only passed satisfies the gate/dependency predicate', () => {
|
||||
expect(isSatisfyingOutcome('passed')).toBe(true);
|
||||
expect(isSatisfyingOutcome('failed')).toBe(false);
|
||||
expect(isSatisfyingOutcome('blocked')).toBe(false);
|
||||
expect(isSatisfyingOutcome('error')).toBe(false);
|
||||
expect(isSatisfyingOutcome('waiting-for-authority')).toBe(false);
|
||||
expect(isSatisfyingOutcome('simulated')).toBe(false);
|
||||
expect(isSatisfyingOutcome('not-applicable')).toBe(false);
|
||||
});
|
||||
|
||||
it('a simulated gate result cannot satisfy the stage gate evaluation', () => {
|
||||
const evaluation = evaluateStageGates('05-coding', STAGE_SPECS['05-coding']!.qualityGates, {
|
||||
task_id: 'FORGE-x-05',
|
||||
outcome: 'passed',
|
||||
reason: 'executor claims success',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: [{ gate: 'pnpm lint', outcome: 'simulated', reason: 'simulated gate' }],
|
||||
});
|
||||
expect(isSatisfyingOutcome(evaluation.outcome)).toBe(false);
|
||||
expect(evaluation.outcome).toBe('error');
|
||||
});
|
||||
|
||||
it('a simulated task outcome cannot satisfy evaluation in normal mode', () => {
|
||||
const evaluation = evaluateStageGates('00-intake', [], {
|
||||
task_id: 'FORGE-x-00',
|
||||
outcome: 'simulated',
|
||||
reason: 'executor reported simulated',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: [],
|
||||
});
|
||||
expect(isSatisfyingOutcome(evaluation.outcome)).toBe(false);
|
||||
});
|
||||
|
||||
it('a missing gate result blocks the stage instead of passing vacuously', () => {
|
||||
const evaluation = evaluateStageGates('05-coding', STAGE_SPECS['05-coding']!.qualityGates, {
|
||||
task_id: 'FORGE-x-05',
|
||||
outcome: 'passed',
|
||||
reason: 'executor claims success',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: [],
|
||||
});
|
||||
expect(evaluation.outcome).toBe('blocked');
|
||||
});
|
||||
});
|
||||
|
||||
describe('fail-closed: authority and provider gates', () => {
|
||||
let tmpDir: string;
|
||||
let briefPath: string;
|
||||
|
||||
beforeEach(() => {
|
||||
tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'forge-authority-'));
|
||||
briefPath = path.join(tmpDir, 'brief.md');
|
||||
fs.writeFileSync(briefPath, '# Fix bug\n\nA bugfix for lint cleanup.');
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
fs.rmSync(tmpDir, { recursive: true, force: true });
|
||||
});
|
||||
|
||||
it.each(['02-planning-1', '03-planning-2', '04-planning-3', '07-remediate'])(
|
||||
'planning/remediation stage %s yields waiting-for-authority (not passed) in normal mode',
|
||||
async (stage) => {
|
||||
const executor = createTypedExecutor();
|
||||
let runDir: string | undefined;
|
||||
|
||||
try {
|
||||
await runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: [stage as string],
|
||||
});
|
||||
expect.unreachable('runPipeline should have failed closed');
|
||||
} catch (err) {
|
||||
expect(err).toBeInstanceOf(ForgeCapabilityError);
|
||||
expect((err as ForgeCapabilityError).code).toBe('FORGE_AUTHORITY_REQUIRED');
|
||||
runDir = path.join(tmpDir, '.forge', 'runs');
|
||||
}
|
||||
|
||||
const runIds = fs.readdirSync(runDir!);
|
||||
expect(runIds).toHaveLength(1);
|
||||
const manifest = loadManifest(path.join(runDir!, runIds[0]!));
|
||||
expect(manifest.stages[stage]?.status).toBe('waiting-for-authority');
|
||||
expect(manifest.stages[stage]?.status).not.toBe('passed');
|
||||
expect(manifest.status).toBe('waiting-for-authority');
|
||||
},
|
||||
);
|
||||
|
||||
it('review stage fails closed with a typed FORGE_NO_REVIEWER error in normal mode', async () => {
|
||||
const executor = createTypedExecutor();
|
||||
|
||||
try {
|
||||
await runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['06-review'],
|
||||
});
|
||||
expect.unreachable('runPipeline should have failed closed');
|
||||
} catch (err) {
|
||||
expect(err).toBeInstanceOf(ForgeCapabilityError);
|
||||
expect((err as ForgeCapabilityError).code).toBe('FORGE_NO_REVIEWER');
|
||||
expect((err as ForgeCapabilityError).capability).toBe('reviewer');
|
||||
}
|
||||
|
||||
const runsDir = path.join(tmpDir, '.forge', 'runs');
|
||||
const runIds = fs.readdirSync(runsDir);
|
||||
const manifest = loadManifest(path.join(runsDir, runIds[0]!));
|
||||
expect(manifest.stages['06-review']?.status).toBe('blocked');
|
||||
expect(manifest.stages['06-review']?.status).not.toBe('passed');
|
||||
expect(manifest.status).toBe('failed');
|
||||
});
|
||||
|
||||
it('review stage produces simulated results under --simulate', async () => {
|
||||
const result = await runPipeline(briefPath, tmpDir, {
|
||||
simulate: true,
|
||||
stages: ['06-review'],
|
||||
});
|
||||
|
||||
expect(result.manifest.mode).toBe('simulated');
|
||||
expect(result.manifest.stages['06-review']?.status).toBe('simulated');
|
||||
for (const gateResult of result.manifest.stages['06-review']?.gateResults ?? []) {
|
||||
expect(gateResult.outcome).toBe('simulated');
|
||||
}
|
||||
});
|
||||
|
||||
it('deploy stage fails closed without a wired ci-pipeline provider in normal mode', async () => {
|
||||
const executor = createTypedExecutor();
|
||||
|
||||
await expect(
|
||||
runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['09-deploy'],
|
||||
}),
|
||||
).rejects.toMatchObject({
|
||||
name: 'ForgeCapabilityError',
|
||||
code: 'FORGE_NO_CI_PIPELINE',
|
||||
});
|
||||
});
|
||||
});
|
||||
|
||||
describe('fail-closed: no vacuous gate commands remain', () => {
|
||||
it('stage constants contain no echo/synthetic-approval, vacuous true, or empty gate commands', () => {
|
||||
for (const [stageName, spec] of Object.entries(STAGE_SPECS)) {
|
||||
for (const gate of spec.qualityGates) {
|
||||
const serialized = JSON.stringify(gate);
|
||||
// The echo-review synthetic approval must be gone.
|
||||
expect(serialized, `stage ${stageName} gate ${serialized}`).not.toContain('echo');
|
||||
expect(serialized, `stage ${stageName} gate ${serialized}`).not.toMatch(/"verdict"\s*:/);
|
||||
expect(serialized, `stage ${stageName} gate ${serialized}`).not.toMatch(
|
||||
/"summary"\s*:\s*"review-pass"/,
|
||||
);
|
||||
// No vacuous literal `true` gate.
|
||||
expect(gate, `stage ${stageName}`).not.toBe('true');
|
||||
// Command gates must carry a real, non-empty command.
|
||||
if (isCommandGate(gate)) {
|
||||
const command = typeof gate === 'string' ? gate : gate.command;
|
||||
expect(command.trim().length, `stage ${stageName} gate ${serialized}`).toBeGreaterThan(0);
|
||||
}
|
||||
}
|
||||
}
|
||||
});
|
||||
|
||||
it('board tasks contain no vacuous true gates', () => {
|
||||
const tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'forge-board-gates-'));
|
||||
try {
|
||||
const tasks = generateBoardTasks('# Brief', [], tmpDir, 'BOARD-TEST');
|
||||
for (const task of tasks) {
|
||||
for (const gate of task.qualityGates) {
|
||||
expect(gate, `task ${task.id}`).not.toBe('true');
|
||||
const serialized = JSON.stringify(gate);
|
||||
expect(serialized, `task ${task.id} gate ${serialized}`).not.toContain('echo');
|
||||
}
|
||||
}
|
||||
} finally {
|
||||
fs.rmSync(tmpDir, { recursive: true, force: true });
|
||||
}
|
||||
});
|
||||
});
|
||||
@@ -12,10 +12,10 @@ import {
|
||||
resumePipeline,
|
||||
getPipelineStatus,
|
||||
} from '../src/pipeline-runner.js';
|
||||
import type { ForgeTask, ForgeTaskResult, RunManifest, TaskExecutor } from '../src/types.js';
|
||||
import { gateLabel, isCommandGate } from '../src/outcomes.js';
|
||||
import type { ForgeTask, RunManifest, TaskExecutor } from '../src/types.js';
|
||||
import type { TaskResult } from '@mosaicstack/macp';
|
||||
|
||||
/** Mock TaskExecutor that records submitted tasks and returns typed results. */
|
||||
/** Mock TaskExecutor that records submitted tasks and returns success. */
|
||||
function createMockExecutor(options?: {
|
||||
failStage?: string;
|
||||
}): TaskExecutor & { submittedTasks: ForgeTask[] } {
|
||||
@@ -25,7 +25,7 @@ function createMockExecutor(options?: {
|
||||
async submitTask(task: ForgeTask) {
|
||||
submittedTasks.push(task);
|
||||
},
|
||||
async waitForCompletion(taskId: string): Promise<ForgeTaskResult> {
|
||||
async waitForCompletion(taskId: string): Promise<TaskResult> {
|
||||
const failStage = options?.failStage;
|
||||
const task = submittedTasks.find((t) => t.id === taskId);
|
||||
const stageName = task?.metadata?.['stageName'] as string | undefined;
|
||||
@@ -33,8 +33,7 @@ function createMockExecutor(options?: {
|
||||
if (failStage && stageName === failStage) {
|
||||
return {
|
||||
task_id: taskId,
|
||||
outcome: 'failed',
|
||||
reason: 'mock task failure',
|
||||
status: 'failed',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 1,
|
||||
gate_results: [],
|
||||
@@ -42,17 +41,10 @@ function createMockExecutor(options?: {
|
||||
}
|
||||
return {
|
||||
task_id: taskId,
|
||||
outcome: 'passed',
|
||||
reason: 'mock verified',
|
||||
status: 'completed',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: (task?.qualityGates ?? [])
|
||||
.filter((gate) => isCommandGate(gate))
|
||||
.map((gate) => ({
|
||||
gate: gateLabel(gate),
|
||||
outcome: 'passed' as const,
|
||||
reason: 'mock verified',
|
||||
})),
|
||||
gate_results: [],
|
||||
};
|
||||
},
|
||||
async getTaskStatus() {
|
||||
@@ -164,13 +156,12 @@ describe('runPipeline', () => {
|
||||
const executor = createMockExecutor();
|
||||
const result = await runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['00-intake', '05-coding'],
|
||||
stages: ['00-intake', '00b-discovery'],
|
||||
});
|
||||
|
||||
expect(result.runId).toMatch(/^\d{8}-\d{6}$/);
|
||||
expect(result.stages).toEqual(['00-intake', '05-coding']);
|
||||
expect(result.stages).toEqual(['00-intake', '00b-discovery']);
|
||||
expect(result.manifest.status).toBe('completed');
|
||||
expect(result.manifest.mode).toBe('normal');
|
||||
expect(executor.submittedTasks).toHaveLength(2);
|
||||
});
|
||||
|
||||
@@ -189,17 +180,12 @@ describe('runPipeline', () => {
|
||||
const executor = createMockExecutor();
|
||||
const result = await runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['00-intake', '05-coding'],
|
||||
stages: ['00-intake', '00b-discovery'],
|
||||
});
|
||||
|
||||
const manifest = loadManifest(result.runDir);
|
||||
expect(manifest.stages['00-intake']?.status).toBe('passed');
|
||||
expect(manifest.stages['05-coding']?.status).toBe('passed');
|
||||
expect(manifest.stages['05-coding']?.gateResults?.map((g) => g.outcome)).toEqual([
|
||||
'passed',
|
||||
'passed',
|
||||
'passed',
|
||||
]);
|
||||
expect(manifest.stages['00b-discovery']?.status).toBe('passed');
|
||||
});
|
||||
|
||||
it('respects CLI class override', async () => {
|
||||
@@ -229,7 +215,7 @@ describe('runPipeline', () => {
|
||||
const executor = createMockExecutor();
|
||||
await runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['00-intake', '05-coding', '08-test'],
|
||||
stages: ['00-intake', '00b-discovery', '02-planning-1'],
|
||||
});
|
||||
|
||||
expect(executor.submittedTasks[0]!.dependsOn).toBeUndefined();
|
||||
@@ -238,14 +224,14 @@ describe('runPipeline', () => {
|
||||
});
|
||||
|
||||
it('handles stage failure', async () => {
|
||||
const executor = createMockExecutor({ failStage: '05-coding' });
|
||||
const executor = createMockExecutor({ failStage: '00b-discovery' });
|
||||
|
||||
await expect(
|
||||
runPipeline(briefPath, tmpDir, {
|
||||
executor,
|
||||
stages: ['00-intake', '05-coding'],
|
||||
stages: ['00-intake', '00b-discovery'],
|
||||
}),
|
||||
).rejects.toThrow('Stage 05-coding failed');
|
||||
).rejects.toThrow('Stage 00b-discovery failed');
|
||||
});
|
||||
|
||||
it('marks manifest as failed on stage failure', async () => {
|
||||
@@ -284,143 +270,30 @@ describe('resumePipeline', () => {
|
||||
fs.rmSync(tmpDir, { recursive: true, force: true });
|
||||
});
|
||||
|
||||
it('resumes from first incomplete stage and fails closed at the next provider gate', async () => {
|
||||
// Simulate a run whose authority stages were approved out-of-band
|
||||
// (recorded as passed) and whose coding stage failed mechanically.
|
||||
const runId = '20260101-000000';
|
||||
const runDir = path.join(tmpDir, '.forge', 'runs', runId);
|
||||
fs.mkdirSync(runDir, { recursive: true });
|
||||
const passed = { status: 'passed' as const, startedAt: '2026-01-01T00:00:00Z' };
|
||||
saveManifest(runDir, {
|
||||
runId,
|
||||
brief: briefPath,
|
||||
codebase: tmpDir,
|
||||
briefClass: 'hotfix',
|
||||
classSource: 'frontmatter',
|
||||
forceBoard: false,
|
||||
mode: 'normal',
|
||||
createdAt: '2026-01-01T00:00:00Z',
|
||||
updatedAt: '2026-01-01T00:00:00Z',
|
||||
currentStage: '05-coding',
|
||||
status: 'failed',
|
||||
stages: {
|
||||
'00-intake': passed,
|
||||
'00b-discovery': passed,
|
||||
'02-planning-1': passed,
|
||||
'03-planning-2': passed,
|
||||
'04-planning-3': passed,
|
||||
'05-coding': { status: 'failed', reason: 'gate failed' },
|
||||
},
|
||||
});
|
||||
it('resumes from first incomplete stage', async () => {
|
||||
// First run fails on discovery
|
||||
const executor1 = createMockExecutor({ failStage: '00b-discovery' });
|
||||
let runDir: string;
|
||||
|
||||
// Resume re-runs 05-coding (the first non-passed stage), then fails
|
||||
// closed at 06-review because no reviewer provider is wired.
|
||||
const executor = createMockExecutor();
|
||||
await expect(resumePipeline(runDir, executor)).rejects.toMatchObject({
|
||||
name: 'ForgeCapabilityError',
|
||||
code: 'FORGE_NO_REVIEWER',
|
||||
});
|
||||
|
||||
const manifest = loadManifest(runDir);
|
||||
expect(manifest.stages['05-coding']?.status).toBe('passed');
|
||||
expect(manifest.stages['06-review']?.status).toBe('blocked');
|
||||
expect(manifest.status).toBe('failed');
|
||||
});
|
||||
|
||||
it('resumes to completion as simulated under explicit simulate', async () => {
|
||||
const runId = '20260101-000003';
|
||||
const runDir = path.join(tmpDir, '.forge', 'runs', runId);
|
||||
fs.mkdirSync(runDir, { recursive: true });
|
||||
const passed = { status: 'passed' as const, startedAt: '2026-01-01T00:00:00Z' };
|
||||
saveManifest(runDir, {
|
||||
runId,
|
||||
brief: briefPath,
|
||||
codebase: tmpDir,
|
||||
briefClass: 'hotfix',
|
||||
classSource: 'frontmatter',
|
||||
forceBoard: false,
|
||||
mode: 'normal',
|
||||
createdAt: '2026-01-01T00:00:00Z',
|
||||
updatedAt: '2026-01-01T00:00:00Z',
|
||||
currentStage: '05-coding',
|
||||
status: 'failed',
|
||||
stages: {
|
||||
'00-intake': passed,
|
||||
'00b-discovery': passed,
|
||||
'02-planning-1': passed,
|
||||
'03-planning-2': passed,
|
||||
'04-planning-3': passed,
|
||||
'05-coding': { status: 'failed', reason: 'gate failed' },
|
||||
},
|
||||
});
|
||||
|
||||
const result = await resumePipeline(runDir, undefined, { simulate: true });
|
||||
|
||||
expect(result.manifest.status).toBe('simulated');
|
||||
expect(result.manifest.mode).toBe('simulated');
|
||||
expect(result.stages[0]).toBe('05-coding');
|
||||
for (const stage of result.stages) {
|
||||
expect(result.manifest.stages[stage]?.status).toBe('simulated');
|
||||
try {
|
||||
await runPipeline(briefPath, tmpDir, {
|
||||
executor: executor1,
|
||||
stages: ['00-intake', '00b-discovery', '02-planning-1'],
|
||||
});
|
||||
} catch {
|
||||
// expected
|
||||
}
|
||||
});
|
||||
|
||||
it('fails closed on resume when the next stage needs authority sign-off', async () => {
|
||||
const runId = '20260101-000001';
|
||||
const runDir = path.join(tmpDir, '.forge', 'runs', runId);
|
||||
fs.mkdirSync(runDir, { recursive: true });
|
||||
saveManifest(runDir, {
|
||||
runId,
|
||||
brief: briefPath,
|
||||
codebase: tmpDir,
|
||||
briefClass: 'hotfix',
|
||||
classSource: 'frontmatter',
|
||||
forceBoard: false,
|
||||
mode: 'normal',
|
||||
createdAt: '2026-01-01T00:00:00Z',
|
||||
updatedAt: '2026-01-01T00:00:00Z',
|
||||
currentStage: '00-intake',
|
||||
status: 'in_progress',
|
||||
stages: {
|
||||
'00-intake': { status: 'passed' },
|
||||
},
|
||||
});
|
||||
const runsDir = path.join(tmpDir, '.forge', 'runs');
|
||||
runDir = path.join(runsDir, fs.readdirSync(runsDir)[0]!);
|
||||
|
||||
const executor = createMockExecutor();
|
||||
await expect(resumePipeline(runDir, executor)).rejects.toMatchObject({
|
||||
name: 'ForgeCapabilityError',
|
||||
code: 'FORGE_AUTHORITY_REQUIRED',
|
||||
});
|
||||
// Resume should pick up from 00b-discovery
|
||||
const executor2 = createMockExecutor();
|
||||
const result = await resumePipeline(runDir, executor2);
|
||||
|
||||
const manifest = loadManifest(runDir);
|
||||
expect(manifest.stages['00b-discovery']?.status).toBe('waiting-for-authority');
|
||||
expect(manifest.status).toBe('waiting-for-authority');
|
||||
});
|
||||
|
||||
it('fails closed on resume without an executor or --simulate', async () => {
|
||||
const runId = '20260101-000002';
|
||||
const runDir = path.join(tmpDir, '.forge', 'runs', runId);
|
||||
fs.mkdirSync(runDir, { recursive: true });
|
||||
saveManifest(runDir, {
|
||||
runId,
|
||||
brief: briefPath,
|
||||
codebase: tmpDir,
|
||||
briefClass: 'hotfix',
|
||||
classSource: 'frontmatter',
|
||||
forceBoard: false,
|
||||
mode: 'normal',
|
||||
createdAt: '2026-01-01T00:00:00Z',
|
||||
updatedAt: '2026-01-01T00:00:00Z',
|
||||
currentStage: '00-intake',
|
||||
status: 'in_progress',
|
||||
stages: {
|
||||
'00-intake': { status: 'passed' },
|
||||
},
|
||||
});
|
||||
|
||||
await expect(resumePipeline(runDir)).rejects.toMatchObject({
|
||||
name: 'ForgeCapabilityError',
|
||||
code: 'FORGE_NO_EXECUTOR',
|
||||
});
|
||||
expect(result.manifest.status).toBe('completed');
|
||||
// Should have re-run from 00b-discovery onward
|
||||
expect(result.stages[0]).toBe('00b-discovery');
|
||||
});
|
||||
});
|
||||
|
||||
|
||||
@@ -95,14 +95,7 @@ export function generateBoardTasks(
|
||||
briefPath,
|
||||
resultPath: resultRelPath,
|
||||
timeoutSeconds: 120,
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'board-approval',
|
||||
reason:
|
||||
'persona evaluation is judged by board synthesis (authority review); no mechanical gate exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
metadata: {
|
||||
personaName: persona.name,
|
||||
personaSlug: persona.slug,
|
||||
@@ -128,13 +121,7 @@ export function generateBoardTasks(
|
||||
timeoutSeconds: 120,
|
||||
dependsOn: personaTaskIds,
|
||||
dependsOnPolicy: 'all_terminal',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'board-approval',
|
||||
reason: 'board synthesis is an authority decision; no mechanical gate exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
metadata: {
|
||||
resultOutputPath: synthesisResult,
|
||||
inputResultPaths: personaResultPaths,
|
||||
|
||||
@@ -1,11 +1,7 @@
|
||||
import fs from 'node:fs';
|
||||
import os from 'node:os';
|
||||
import path from 'node:path';
|
||||
import { Command } from 'commander';
|
||||
import { describe, expect, it, vi, beforeEach, afterEach } from 'vitest';
|
||||
import { describe, expect, it } from 'vitest';
|
||||
|
||||
import { registerForgeCommand } from './cli.js';
|
||||
import { loadManifest } from './pipeline-runner.js';
|
||||
|
||||
describe('registerForgeCommand', () => {
|
||||
it('registers a "forge" command on the parent program', () => {
|
||||
@@ -59,94 +55,3 @@ describe('registerForgeCommand', () => {
|
||||
}).not.toThrow();
|
||||
});
|
||||
});
|
||||
|
||||
describe('forge run fail-closed behavior (SDLC-D-035)', () => {
|
||||
let tmpDir: string;
|
||||
let briefPath: string;
|
||||
let errSpy: ReturnType<typeof vi.spyOn>;
|
||||
let logSpy: ReturnType<typeof vi.spyOn>;
|
||||
let prevExitCode: string | number | null | undefined;
|
||||
|
||||
const parse = (args: string[]) => {
|
||||
const program = new Command();
|
||||
registerForgeCommand(program);
|
||||
return program.parseAsync(['forge', ...args], { from: 'user' });
|
||||
};
|
||||
|
||||
beforeEach(() => {
|
||||
tmpDir = fs.mkdtempSync(path.join(os.tmpdir(), 'forge-cli-failclosed-'));
|
||||
briefPath = path.join(tmpDir, 'brief.md');
|
||||
fs.writeFileSync(briefPath, '# Fix bug\n\nA bugfix for lint cleanup.');
|
||||
errSpy = vi.spyOn(console, 'error').mockImplementation(() => {});
|
||||
logSpy = vi.spyOn(console, 'log').mockImplementation(() => {});
|
||||
prevExitCode = process.exitCode;
|
||||
});
|
||||
|
||||
afterEach(() => {
|
||||
errSpy.mockRestore();
|
||||
logSpy.mockRestore();
|
||||
process.exitCode = prevExitCode;
|
||||
fs.rmSync(tmpDir, { recursive: true, force: true });
|
||||
});
|
||||
|
||||
it('exits nonzero with a typed FORGE_NO_EXECUTOR error when no executor is wired and --simulate is absent', async () => {
|
||||
await parse(['run', '--brief', briefPath, '--codebase', tmpDir]);
|
||||
|
||||
expect(process.exitCode).toBe(1);
|
||||
const errText = errSpy.mock.calls.map((c) => c.join(' ')).join('\n');
|
||||
expect(errText).toContain('FORGE_NO_EXECUTOR');
|
||||
// It must never run the pipeline with a stub and report success.
|
||||
expect(fs.existsSync(path.join(tmpDir, '.forge', 'runs'))).toBe(false);
|
||||
});
|
||||
|
||||
it('completes with typed simulated results and exit 0 under explicit --simulate', async () => {
|
||||
await parse(['run', '--brief', briefPath, '--codebase', tmpDir, '--simulate']);
|
||||
|
||||
expect(process.exitCode).toBeUndefined();
|
||||
|
||||
// Loud simulated-mode summary.
|
||||
const logText = logSpy.mock.calls.map((c) => c.join(' ')).join('\n');
|
||||
expect(logText).toContain('SIMULATED');
|
||||
|
||||
// Manifest records the mode and simulated per-result statuses.
|
||||
const runsDir = path.join(tmpDir, '.forge', 'runs');
|
||||
const runIds = fs.readdirSync(runsDir);
|
||||
expect(runIds).toHaveLength(1);
|
||||
const manifest = loadManifest(path.join(runsDir, runIds[0]!));
|
||||
expect(manifest.mode).toBe('simulated');
|
||||
expect(manifest.status).toBe('simulated');
|
||||
for (const stageStatus of Object.values(manifest.stages)) {
|
||||
expect(stageStatus?.status).toBe('simulated');
|
||||
for (const gateResult of stageStatus?.gateResults ?? []) {
|
||||
expect(gateResult.outcome).toBe('simulated');
|
||||
}
|
||||
}
|
||||
});
|
||||
|
||||
it('resume exits nonzero with a typed FORGE_NO_EXECUTOR error without --simulate', async () => {
|
||||
const runDir = path.join(tmpDir, '.forge', 'runs', '20260101-000000');
|
||||
fs.mkdirSync(runDir, { recursive: true });
|
||||
fs.writeFileSync(
|
||||
path.join(runDir, 'manifest.json'),
|
||||
JSON.stringify({
|
||||
runId: '20260101-000000',
|
||||
brief: briefPath,
|
||||
codebase: tmpDir,
|
||||
briefClass: 'hotfix',
|
||||
classSource: 'frontmatter',
|
||||
forceBoard: false,
|
||||
createdAt: '2026-01-01T00:00:00Z',
|
||||
updatedAt: '2026-01-01T00:00:00Z',
|
||||
currentStage: '00-intake',
|
||||
status: 'in_progress',
|
||||
stages: { '00-intake': { status: 'passed' } },
|
||||
}),
|
||||
);
|
||||
|
||||
await parse(['resume', '20260101-000000', '--project', tmpDir]);
|
||||
|
||||
expect(process.exitCode).toBe(1);
|
||||
const errText = errSpy.mock.calls.map((c) => c.join(' ')).join('\n');
|
||||
expect(errText).toContain('FORGE_NO_EXECUTOR');
|
||||
});
|
||||
});
|
||||
|
||||
+48
-122
@@ -5,47 +5,37 @@ import type { Command } from 'commander';
|
||||
|
||||
import { classifyBrief } from './brief-classifier.js';
|
||||
import { STAGE_LABELS, STAGE_SEQUENCE } from './constants.js';
|
||||
import { ForgeCapabilityError } from './errors.js';
|
||||
import { getEffectivePersonas, loadBoardPersonas } from './persona-loader.js';
|
||||
import { generateRunId, getPipelineStatus, loadManifest, runPipeline } from './pipeline-runner.js';
|
||||
import { createSimulatedExecutor } from './simulated-executor.js';
|
||||
import type { PipelineOptions, RunManifest, RunMode } from './types.js';
|
||||
import type { PipelineOptions, RunManifest, TaskExecutor } from './types.js';
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// Stub executor — used when no real executor is wired at CLI invocation time.
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
const stubExecutor: TaskExecutor = {
|
||||
async submitTask(task) {
|
||||
console.log(` [forge] stage submitted: ${task.id} (${task.title})`);
|
||||
},
|
||||
async waitForCompletion(taskId, _timeoutMs) {
|
||||
console.log(` [forge] stage complete: ${taskId}`);
|
||||
return {
|
||||
task_id: taskId,
|
||||
status: 'completed' as const,
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: [],
|
||||
};
|
||||
},
|
||||
async getTaskStatus(_taskId) {
|
||||
return 'completed' as const;
|
||||
},
|
||||
};
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// Helpers
|
||||
// ---------------------------------------------------------------------------
|
||||
|
||||
/** Resolve a run's effective mode, defaulting legacy manifests to normal. */
|
||||
function runModeOf(manifest: RunManifest): RunMode {
|
||||
return manifest.mode ?? 'normal';
|
||||
}
|
||||
|
||||
/** Print a loud banner so a simulated run can never be misread as verified. */
|
||||
function printSimulatedBanner(): void {
|
||||
console.log('');
|
||||
console.log('[forge] ===============================================================');
|
||||
console.log('[forge] MODE: SIMULATED — no stage or gate was really executed.');
|
||||
console.log('[forge] All results are synthetic and MUST NOT be read as verified');
|
||||
console.log('[forge] success. Wire a real executor/providers and re-run to verify.');
|
||||
console.log('[forge] ===============================================================');
|
||||
}
|
||||
|
||||
/** Print a typed error line for fail-closed capability errors. */
|
||||
function printCapabilityError(err: ForgeCapabilityError): void {
|
||||
console.error(`[forge] error ${err.code}: ${err.message}`);
|
||||
console.error(`[forge] missing capability: ${err.capability}`);
|
||||
}
|
||||
|
||||
/** Handle a pipeline error uniformly: typed capability errors get their code. */
|
||||
function handlePipelineError(err: unknown): void {
|
||||
if (err instanceof ForgeCapabilityError) {
|
||||
printCapabilityError(err);
|
||||
} else {
|
||||
console.error(`[forge] pipeline failed: ${err instanceof Error ? err.message : String(err)}`);
|
||||
}
|
||||
process.exitCode = 1;
|
||||
}
|
||||
|
||||
function formatDuration(startedAt?: string, completedAt?: string): string {
|
||||
if (!startedAt || !completedAt) return '-';
|
||||
const ms = new Date(completedAt).getTime() - new Date(startedAt).getTime();
|
||||
@@ -54,24 +44,19 @@ function formatDuration(startedAt?: string, completedAt?: string): string {
|
||||
}
|
||||
|
||||
function printManifestTable(manifest: RunManifest): void {
|
||||
const mode = runModeOf(manifest);
|
||||
console.log(`\nRun ID : ${manifest.runId}`);
|
||||
console.log(`Status : ${manifest.status}`);
|
||||
console.log(`Mode : ${mode}`);
|
||||
if (mode === 'simulated') {
|
||||
console.log('WARNING: SIMULATED RUN — results are synthetic, not verified success.');
|
||||
}
|
||||
console.log(`Brief : ${manifest.brief}`);
|
||||
console.log(`Class : ${manifest.briefClass} (${manifest.classSource})`);
|
||||
console.log(`Updated: ${manifest.updatedAt}`);
|
||||
console.log('');
|
||||
console.log('Stage'.padEnd(22) + 'Status'.padEnd(24) + 'Duration');
|
||||
console.log('-'.repeat(60));
|
||||
console.log('Stage'.padEnd(22) + 'Status'.padEnd(14) + 'Duration');
|
||||
console.log('-'.repeat(50));
|
||||
for (const stage of STAGE_SEQUENCE) {
|
||||
const s = manifest.stages[stage];
|
||||
if (!s) continue;
|
||||
const label = (STAGE_LABELS[stage] ?? stage).padEnd(22);
|
||||
const status = s.status.padEnd(24);
|
||||
const status = s.status.padEnd(14);
|
||||
const dur = formatDuration(s.startedAt, s.completedAt);
|
||||
console.log(`${label}${status}${dur}`);
|
||||
}
|
||||
@@ -105,58 +90,23 @@ function listRecentRuns(projectRoot?: string): void {
|
||||
}
|
||||
|
||||
console.log('\nRecent runs:');
|
||||
console.log('Run ID'.padEnd(22) + 'Status'.padEnd(24) + 'Mode'.padEnd(12) + 'Brief');
|
||||
console.log('-'.repeat(80));
|
||||
console.log('Run ID'.padEnd(22) + 'Status'.padEnd(14) + 'Brief');
|
||||
console.log('-'.repeat(70));
|
||||
|
||||
for (const runId of entries) {
|
||||
const runDir = path.join(runsDir, runId);
|
||||
try {
|
||||
const manifest = loadManifest(runDir);
|
||||
const status = manifest.status.padEnd(24);
|
||||
const mode = runModeOf(manifest).padEnd(12);
|
||||
const status = manifest.status.padEnd(14);
|
||||
const brief = path.basename(manifest.brief);
|
||||
console.log(`${runId.padEnd(22)}${status}${mode}${brief}`);
|
||||
console.log(`${runId.padEnd(22)}${status}${brief}`);
|
||||
} catch {
|
||||
console.log(`${runId.padEnd(22)}${'(unreadable)'.padEnd(24)}`);
|
||||
console.log(`${runId.padEnd(22)}${'(unreadable)'.padEnd(14)}`);
|
||||
}
|
||||
}
|
||||
console.log('');
|
||||
}
|
||||
|
||||
/**
|
||||
* Apply the exit-code policy for a finished pipeline run (SDLC-D-035):
|
||||
*
|
||||
* - exit 0 only for a verified `completed` normal run, or for an overall
|
||||
* `simulated` run when the caller explicitly passed --simulate;
|
||||
* - anything else exits nonzero so it can never be read as success.
|
||||
*/
|
||||
function applyRunExitPolicy(result: { manifest: RunManifest; runDir: string }, simulate: boolean) {
|
||||
const { manifest } = result;
|
||||
|
||||
if (runModeOf(manifest) === 'simulated') {
|
||||
if (!simulate || manifest.status !== 'simulated') {
|
||||
console.error(
|
||||
'[forge] error FORGE_MODE_MISMATCH: run reports simulated results without an explicit, ' +
|
||||
'consistent --simulate request; refusing to report success.',
|
||||
);
|
||||
process.exitCode = 1;
|
||||
return;
|
||||
}
|
||||
printSimulatedBanner();
|
||||
console.log(`[forge] run directory: ${result.runDir}`);
|
||||
return; // exit 0 — the caller explicitly opted into simulation
|
||||
}
|
||||
|
||||
if (manifest.status !== 'completed') {
|
||||
console.error(`[forge] run did not complete: terminal status '${manifest.status}'`);
|
||||
process.exitCode = 1;
|
||||
return;
|
||||
}
|
||||
|
||||
console.log(`[forge] pipeline complete (mode: normal): ${manifest.runId}`);
|
||||
console.log(`[forge] run directory: ${result.runDir}`);
|
||||
}
|
||||
|
||||
// ---------------------------------------------------------------------------
|
||||
// Register function
|
||||
// ---------------------------------------------------------------------------
|
||||
@@ -179,11 +129,6 @@ export function registerForgeCommand(parent: Command): void {
|
||||
.option('--config <path>', 'Path to forge config file (.forge/config.yaml)')
|
||||
.option('--codebase <path>', 'Codebase root to pass to the pipeline', process.cwd())
|
||||
.option('--dry-run', 'Print planned stages without executing', false)
|
||||
.option(
|
||||
'--simulate',
|
||||
'Simulate execution without real providers (every result is typed simulated, never verified)',
|
||||
false,
|
||||
)
|
||||
.action(
|
||||
async (opts: {
|
||||
brief: string;
|
||||
@@ -192,7 +137,6 @@ export function registerForgeCommand(parent: Command): void {
|
||||
config?: string;
|
||||
codebase: string;
|
||||
dryRun: boolean;
|
||||
simulate: boolean;
|
||||
}) => {
|
||||
const briefPath = path.resolve(opts.brief);
|
||||
|
||||
@@ -205,22 +149,14 @@ export function registerForgeCommand(parent: Command): void {
|
||||
const briefContent = fs.readFileSync(briefPath, 'utf-8');
|
||||
const briefClass = classifyBrief(briefContent);
|
||||
const projectRoot = opts.codebase;
|
||||
// A real executor is never wired at CLI invocation time today, so the
|
||||
// only executor we may construct is the explicitly-requested simulated
|
||||
// one. Normal mode fails closed with FORGE_NO_EXECUTOR.
|
||||
const executor = opts.simulate ? createSimulatedExecutor() : undefined;
|
||||
|
||||
if (opts.resume) {
|
||||
const runId = opts.runId ?? generateRunId();
|
||||
const runDir = resolveRunDir(runId, projectRoot);
|
||||
console.log(`[forge] resuming run: ${runId}`);
|
||||
try {
|
||||
const { resumePipeline } = await import('./pipeline-runner.js');
|
||||
const result = await resumePipeline(runDir, executor, { simulate: opts.simulate });
|
||||
applyRunExitPolicy(result, opts.simulate);
|
||||
} catch (err) {
|
||||
handlePipelineError(err);
|
||||
}
|
||||
const { resumePipeline } = await import('./pipeline-runner.js');
|
||||
const result = await resumePipeline(runDir, stubExecutor);
|
||||
console.log(`[forge] pipeline complete: ${result.runId}`);
|
||||
return;
|
||||
}
|
||||
|
||||
@@ -228,8 +164,7 @@ export function registerForgeCommand(parent: Command): void {
|
||||
briefClass,
|
||||
codebase: projectRoot,
|
||||
dryRun: opts.dryRun,
|
||||
executor,
|
||||
simulate: opts.simulate,
|
||||
executor: stubExecutor,
|
||||
};
|
||||
|
||||
if (opts.dryRun) {
|
||||
@@ -245,15 +180,16 @@ export function registerForgeCommand(parent: Command): void {
|
||||
|
||||
console.log(`[forge] starting pipeline for brief: ${briefPath}`);
|
||||
console.log(`[forge] classified as: ${briefClass}`);
|
||||
if (opts.simulate) {
|
||||
console.log('[forge] mode: SIMULATED (explicit --simulate)');
|
||||
}
|
||||
|
||||
try {
|
||||
const result = await runPipeline(briefPath, projectRoot, pipelineOptions);
|
||||
applyRunExitPolicy(result, opts.simulate);
|
||||
console.log(`[forge] pipeline complete: ${result.runId}`);
|
||||
console.log(`[forge] run directory: ${result.runDir}`);
|
||||
} catch (err) {
|
||||
handlePipelineError(err);
|
||||
console.error(
|
||||
`[forge] pipeline failed: ${err instanceof Error ? err.message : String(err)}`,
|
||||
);
|
||||
process.exitCode = 1;
|
||||
}
|
||||
},
|
||||
);
|
||||
@@ -288,12 +224,7 @@ export function registerForgeCommand(parent: Command): void {
|
||||
.command('resume <runId>')
|
||||
.description('Resume a stopped or failed pipeline run')
|
||||
.option('--project <path>', 'Project root (defaults to cwd)', process.cwd())
|
||||
.option(
|
||||
'--simulate',
|
||||
'Simulate execution without real providers (every result is typed simulated, never verified)',
|
||||
false,
|
||||
)
|
||||
.action(async (runId: string, opts: { project: string; simulate: boolean }) => {
|
||||
.action(async (runId: string, opts: { project: string }) => {
|
||||
const runDir = resolveRunDir(runId, opts.project);
|
||||
|
||||
if (!fs.existsSync(runDir)) {
|
||||
@@ -303,20 +234,15 @@ export function registerForgeCommand(parent: Command): void {
|
||||
}
|
||||
|
||||
console.log(`[forge] resuming run: ${runId}`);
|
||||
if (opts.simulate) {
|
||||
console.log('[forge] mode: SIMULATED (explicit --simulate)');
|
||||
}
|
||||
|
||||
// No real executor is wired at CLI invocation time; only the explicitly
|
||||
// requested simulated executor may be constructed (fail closed otherwise).
|
||||
const executor = opts.simulate ? createSimulatedExecutor() : undefined;
|
||||
|
||||
try {
|
||||
const { resumePipeline } = await import('./pipeline-runner.js');
|
||||
const result = await resumePipeline(runDir, executor, { simulate: opts.simulate });
|
||||
applyRunExitPolicy(result, opts.simulate);
|
||||
const result = await resumePipeline(runDir, stubExecutor);
|
||||
console.log(`[forge] pipeline complete: ${result.runId}`);
|
||||
console.log(`[forge] run directory: ${result.runDir}`);
|
||||
} catch (err) {
|
||||
handlePipelineError(err);
|
||||
console.error(`[forge] resume failed: ${err instanceof Error ? err.message : String(err)}`);
|
||||
process.exitCode = 1;
|
||||
}
|
||||
});
|
||||
|
||||
|
||||
@@ -9,16 +9,7 @@ export const PACKAGE_ROOT = path.resolve(path.dirname(fileURLToPath(import.meta.
|
||||
/** Pipeline asset directory (stages, agents, rails, gates, templates). */
|
||||
export const PIPELINE_DIR = path.join(PACKAGE_ROOT, 'pipeline');
|
||||
|
||||
/** Stage specifications — defines every pipeline stage.
|
||||
*\n * Gate semantics (SDLC-D-035): every gate is one of
|
||||
* - a real command string / GateEntry a mechanical runner can execute,
|
||||
* - an `authority` gate (human/board sign-off; produces waiting-for-authority),
|
||||
* - a `provider` gate (requires a wired provider such as a reviewer or CI pipeline).
|
||||
*
|
||||
* Vacuous gates (`true`, echo'd synthetic approvals, placeholder ci-pipeline
|
||||
* commands) are forbidden: a stage whose gate has no real implementation
|
||||
* fails closed instead of passing.
|
||||
*/
|
||||
/** Stage specifications — defines every pipeline stage. */
|
||||
export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
'00-intake': {
|
||||
number: '00',
|
||||
@@ -36,13 +27,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'research',
|
||||
gate: 'discovery-complete',
|
||||
promptFile: '00b-discovery.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'discovery-complete',
|
||||
reason: 'discovery completion is attested by an authority; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'01-board': {
|
||||
number: '01',
|
||||
@@ -51,13 +36,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'review',
|
||||
gate: 'board-approval',
|
||||
promptFile: '01-board.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'board-approval',
|
||||
reason: 'board approval is a board/human decision; no mechanical gate exists',
|
||||
},
|
||||
],
|
||||
qualityGates: [{ type: 'ci-pipeline', command: 'board-approval (via board-tasks)' }],
|
||||
},
|
||||
'01b-brief-analyzer': {
|
||||
number: '01b',
|
||||
@@ -66,13 +45,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'research',
|
||||
gate: 'brief-analysis-complete',
|
||||
promptFile: '01-board.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'brief-analysis-complete',
|
||||
reason: 'brief analysis completion is attested by an authority; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'02-planning-1': {
|
||||
number: '02',
|
||||
@@ -81,13 +54,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'research',
|
||||
gate: 'architecture-approval',
|
||||
promptFile: '02-planning-1-architecture.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'architecture-approval',
|
||||
reason: 'ADR approval requires authority sign-off; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'03-planning-2': {
|
||||
number: '03',
|
||||
@@ -96,14 +63,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'research',
|
||||
gate: 'implementation-approval',
|
||||
promptFile: '03-planning-2-implementation.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'implementation-approval',
|
||||
reason:
|
||||
'implementation spec approval requires authority sign-off; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'04-planning-3': {
|
||||
number: '04',
|
||||
@@ -112,14 +72,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'research',
|
||||
gate: 'decomposition-approval',
|
||||
promptFile: '04-planning-3-decomposition.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 'decomposition-approval',
|
||||
reason:
|
||||
'task decomposition approval requires authority sign-off; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'05-coding': {
|
||||
number: '05',
|
||||
@@ -139,10 +92,9 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
promptFile: '06-review.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'provider',
|
||||
capability: 'reviewer',
|
||||
reason:
|
||||
'review verdicts require a wired reviewer provider; synthetic approvals are not permitted',
|
||||
type: 'ai-review',
|
||||
command:
|
||||
'echo \'{"summary":"review-pass","verdict":"approve","findings":[],"stats":{"blockers":0,"should_fix":0,"suggestions":0}}\'',
|
||||
},
|
||||
],
|
||||
},
|
||||
@@ -153,13 +105,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'coding',
|
||||
gate: 're-review',
|
||||
promptFile: '07-remediate.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'authority',
|
||||
capability: 're-review',
|
||||
reason: 'remediation re-review is an approval-based gate; no mechanical check exists',
|
||||
},
|
||||
],
|
||||
qualityGates: ['true'],
|
||||
},
|
||||
'08-test': {
|
||||
number: '08',
|
||||
@@ -177,13 +123,7 @@ export const STAGE_SPECS: Record<string, StageSpec> = {
|
||||
type: 'deploy',
|
||||
gate: 'deploy-verification',
|
||||
promptFile: '09-deploy.md',
|
||||
qualityGates: [
|
||||
{
|
||||
kind: 'provider',
|
||||
capability: 'ci-pipeline',
|
||||
reason: 'deploy verification requires a wired CI pipeline provider',
|
||||
},
|
||||
],
|
||||
qualityGates: [{ type: 'ci-pipeline', command: 'deploy-verification' }],
|
||||
},
|
||||
};
|
||||
|
||||
|
||||
@@ -1,46 +0,0 @@
|
||||
/**
|
||||
* Typed fail-closed capability errors (SDLC-D-035).
|
||||
*
|
||||
* A Forge run must fail closed when a required capability (executor, reviewer
|
||||
* provider, CI pipeline, authority sign-off) is missing. These typed errors
|
||||
* name the missing capability so callers can distinguish "not wired" from
|
||||
* ordinary execution failures.
|
||||
*/
|
||||
|
||||
/** Closed set of typed Forge capability error codes. */
|
||||
export const FORGE_ERROR_CODES = [
|
||||
'FORGE_NO_EXECUTOR',
|
||||
'FORGE_NO_REVIEWER',
|
||||
'FORGE_NO_CI_PIPELINE',
|
||||
'FORGE_NO_PROVIDER',
|
||||
'FORGE_AUTHORITY_REQUIRED',
|
||||
] as const;
|
||||
|
||||
export type ForgeErrorCode = (typeof FORGE_ERROR_CODES)[number];
|
||||
|
||||
/** Raised when a required capability is missing and the pipeline must fail closed. */
|
||||
export class ForgeCapabilityError extends Error {
|
||||
/** Typed error code from the closed FORGE_ERROR_CODES set. */
|
||||
readonly code: ForgeErrorCode;
|
||||
/** The missing capability, e.g. `task-executor`, `reviewer`, `board-approval`. */
|
||||
readonly capability: string;
|
||||
|
||||
constructor(code: ForgeErrorCode, capability: string, message: string) {
|
||||
super(message);
|
||||
this.name = 'ForgeCapabilityError';
|
||||
this.code = code;
|
||||
this.capability = capability;
|
||||
}
|
||||
}
|
||||
|
||||
/** Map a provider gate capability to its typed error code. */
|
||||
export function providerErrorCode(capability: string): ForgeErrorCode {
|
||||
switch (capability) {
|
||||
case 'reviewer':
|
||||
return 'FORGE_NO_REVIEWER';
|
||||
case 'ci-pipeline':
|
||||
return 'FORGE_NO_CI_PIPELINE';
|
||||
default:
|
||||
return 'FORGE_NO_PROVIDER';
|
||||
}
|
||||
}
|
||||
@@ -5,13 +5,6 @@ export type {
|
||||
StageSpec,
|
||||
BriefClass,
|
||||
ClassSource,
|
||||
ForgeOutcome,
|
||||
AuthorityGate,
|
||||
ProviderGate,
|
||||
ForgeGate,
|
||||
ForgeGateResult,
|
||||
ForgeTaskResult,
|
||||
RunMode,
|
||||
StageStatus,
|
||||
RunManifest,
|
||||
ForgeTaskStatus,
|
||||
@@ -88,24 +81,5 @@ export {
|
||||
getPipelineStatus,
|
||||
} from './pipeline-runner.js';
|
||||
|
||||
// Fail-closed errors and typed outcome model (SDLC-D-035)
|
||||
export { FORGE_ERROR_CODES, ForgeCapabilityError, providerErrorCode } from './errors.js';
|
||||
export type { ForgeErrorCode } from './errors.js';
|
||||
export {
|
||||
isSatisfyingOutcome,
|
||||
isCapabilityGate,
|
||||
isCommandGate,
|
||||
gateLabel,
|
||||
uniformGateResults,
|
||||
simulatedGateResults,
|
||||
waitingGateResults,
|
||||
blockedGateResults,
|
||||
evaluateStageGates,
|
||||
} from './outcomes.js';
|
||||
export type { StageEvaluation } from './outcomes.js';
|
||||
|
||||
// Simulated executor (explicit --simulate only)
|
||||
export { createSimulatedExecutor } from './simulated-executor.js';
|
||||
|
||||
// CLI
|
||||
export { registerForgeCommand } from './cli.js';
|
||||
|
||||
@@ -1,147 +0,0 @@
|
||||
import type { GateEntry } from '@mosaicstack/macp';
|
||||
|
||||
import type {
|
||||
AuthorityGate,
|
||||
ForgeGate,
|
||||
ForgeGateResult,
|
||||
ForgeOutcome,
|
||||
ForgeTaskResult,
|
||||
ProviderGate,
|
||||
} from './types.js';
|
||||
|
||||
/**
|
||||
* Gate and dependency satisfaction predicate (SDLC-D-035).
|
||||
*
|
||||
* ONLY a verified `passed` outcome satisfies. Every other member of the closed
|
||||
* outcome set — including `simulated` — is non-satisfying, so a simulated or
|
||||
* authority-blocked result can never be read as success-by-verification.
|
||||
*/
|
||||
export function isSatisfyingOutcome(outcome: ForgeOutcome): boolean {
|
||||
return outcome === 'passed';
|
||||
}
|
||||
|
||||
/** Whether a gate is an authority or provider gate (capability-based, command-less). */
|
||||
export function isCapabilityGate(gate: ForgeGate): gate is AuthorityGate | ProviderGate {
|
||||
if (typeof gate !== 'object' || gate === null) return false;
|
||||
const kind = (gate as Record<string, unknown>)['kind'];
|
||||
return kind === 'authority' || kind === 'provider';
|
||||
}
|
||||
|
||||
/** Whether a gate definition carries a real command a mechanical runner can execute. */
|
||||
export function isCommandGate(gate: ForgeGate): gate is string | GateEntry {
|
||||
if (typeof gate === 'string') {
|
||||
return gate.trim().length > 0;
|
||||
}
|
||||
if (isCapabilityGate(gate)) {
|
||||
// Authority and provider gates are satisfied by a capability, not a command.
|
||||
return false;
|
||||
}
|
||||
return typeof gate.command === 'string' && gate.command.trim().length > 0;
|
||||
}
|
||||
|
||||
/** Typed label identifying a gate in results and logs. */
|
||||
export function gateLabel(gate: ForgeGate): string {
|
||||
if (typeof gate === 'string') return gate;
|
||||
if (isCapabilityGate(gate)) return `${gate.kind}:${gate.capability}`;
|
||||
return gate.command || gate.type || 'unnamed-gate';
|
||||
}
|
||||
|
||||
/** Reason string stamped on every simulated gate result. */
|
||||
export const SIMULATED_GATE_REASON =
|
||||
'simulated execution (--simulate): gate was not evaluated by a real implementation';
|
||||
|
||||
/** Build typed gate results with a uniform outcome for a stage's declared gates. */
|
||||
export function uniformGateResults(
|
||||
gates: ForgeGate[],
|
||||
outcome: ForgeOutcome,
|
||||
reason: string,
|
||||
): ForgeGateResult[] {
|
||||
return gates.map((gate) => ({ gate: gateLabel(gate), outcome, reason }));
|
||||
}
|
||||
|
||||
/** Typed simulated gate results — used exclusively in `--simulate` runs. */
|
||||
export function simulatedGateResults(gates: ForgeGate[]): ForgeGateResult[] {
|
||||
return uniformGateResults(gates, 'simulated', SIMULATED_GATE_REASON);
|
||||
}
|
||||
|
||||
/** Typed waiting-for-authority gate results for approval-based stages. */
|
||||
export function waitingGateResults(gates: ForgeGate[], reason: string): ForgeGateResult[] {
|
||||
return uniformGateResults(gates, 'waiting-for-authority', reason);
|
||||
}
|
||||
|
||||
/** Typed blocked gate results for stages whose provider capability is not wired. */
|
||||
export function blockedGateResults(gates: ForgeGate[], reason: string): ForgeGateResult[] {
|
||||
return uniformGateResults(gates, 'blocked', reason);
|
||||
}
|
||||
|
||||
/** Outcome of evaluating a completed stage in normal mode. */
|
||||
export interface StageEvaluation {
|
||||
outcome: ForgeOutcome;
|
||||
reason: string;
|
||||
gateResults: ForgeGateResult[];
|
||||
}
|
||||
|
||||
/**
|
||||
* Evaluate a stage's declared gates against the executor's typed result.
|
||||
*
|
||||
* Fail-closed mapping:
|
||||
* - a `simulated` task or gate outcome in normal mode maps to `error`
|
||||
* - a missing gate result for a required command gate maps to `blocked`
|
||||
* - a non-passing task outcome propagates as the stage outcome
|
||||
* - only verified `passed` task and gate outcomes yield a `passed` stage
|
||||
*/
|
||||
export function evaluateStageGates(
|
||||
stageName: string,
|
||||
gates: ForgeGate[],
|
||||
result: ForgeTaskResult,
|
||||
): StageEvaluation {
|
||||
const gateResults = result.gate_results ?? [];
|
||||
|
||||
if (result.outcome === 'simulated') {
|
||||
return {
|
||||
outcome: 'error',
|
||||
reason: `executor reported a simulated outcome for stage '${stageName}' in normal mode — refusing to treat simulated results as verified`,
|
||||
gateResults,
|
||||
};
|
||||
}
|
||||
|
||||
if (!isSatisfyingOutcome(result.outcome)) {
|
||||
return {
|
||||
outcome: result.outcome,
|
||||
reason: `task outcome is '${result.outcome}': ${result.reason}`,
|
||||
gateResults,
|
||||
};
|
||||
}
|
||||
|
||||
for (const gate of gates) {
|
||||
// Authority and provider gates are pre-flighted before execution; they have
|
||||
// no mechanical result to verify here.
|
||||
if (!isCommandGate(gate)) continue;
|
||||
|
||||
const label = gateLabel(gate);
|
||||
const gateResult = gateResults.find((r) => r.gate === label);
|
||||
if (!gateResult) {
|
||||
return {
|
||||
outcome: 'blocked',
|
||||
reason: `no gate result was reported for required gate '${label}' (stage '${stageName}')`,
|
||||
gateResults,
|
||||
};
|
||||
}
|
||||
if (!isSatisfyingOutcome(gateResult.outcome)) {
|
||||
return {
|
||||
outcome: gateResult.outcome === 'simulated' ? 'error' : gateResult.outcome,
|
||||
reason: `gate '${label}' outcome is '${gateResult.outcome}': ${gateResult.reason}`,
|
||||
gateResults,
|
||||
};
|
||||
}
|
||||
}
|
||||
|
||||
return {
|
||||
outcome: 'passed',
|
||||
reason:
|
||||
gates.length === 0
|
||||
? "stage declares no gates; task outcome 'passed' accepted"
|
||||
: 'all declared gates verified passed',
|
||||
gateResults,
|
||||
};
|
||||
}
|
||||
@@ -1,33 +1,18 @@
|
||||
import fs from 'node:fs';
|
||||
import path from 'node:path';
|
||||
|
||||
import { STAGE_SEQUENCE, STAGE_SPECS } from './constants.js';
|
||||
import { STAGE_SEQUENCE } from './constants.js';
|
||||
import { determineBriefClass, stagesForClass } from './brief-classifier.js';
|
||||
import { ForgeCapabilityError, providerErrorCode } from './errors.js';
|
||||
import {
|
||||
blockedGateResults,
|
||||
evaluateStageGates,
|
||||
isCapabilityGate,
|
||||
simulatedGateResults,
|
||||
waitingGateResults,
|
||||
} from './outcomes.js';
|
||||
import { mapStageToTask } from './stage-adapter.js';
|
||||
import { createSimulatedExecutor } from './simulated-executor.js';
|
||||
import type {
|
||||
ForgeTask,
|
||||
ForgeTaskResult,
|
||||
PipelineOptions,
|
||||
PipelineResult,
|
||||
RunManifest,
|
||||
RunMode,
|
||||
StageStatus,
|
||||
TaskExecutor,
|
||||
} from './types.js';
|
||||
|
||||
/** Reason stamped on stages that complete under explicit simulation. */
|
||||
const SIMULATED_STAGE_REASON =
|
||||
'simulated execution (--simulate): stage was not executed by a real executor';
|
||||
|
||||
/**
|
||||
* Generate a timestamp-based run ID.
|
||||
*/
|
||||
@@ -62,7 +47,6 @@ function createManifest(opts: {
|
||||
briefClass: RunManifest['briefClass'];
|
||||
classSource: RunManifest['classSource'];
|
||||
forceBoard: boolean;
|
||||
mode: RunMode;
|
||||
runDir: string;
|
||||
}): RunManifest {
|
||||
const ts = nowISO();
|
||||
@@ -73,7 +57,6 @@ function createManifest(opts: {
|
||||
briefClass: opts.briefClass,
|
||||
classSource: opts.classSource,
|
||||
forceBoard: opts.forceBoard,
|
||||
mode: opts.mode,
|
||||
createdAt: ts,
|
||||
updatedAt: ts,
|
||||
currentStage: '',
|
||||
@@ -125,199 +108,20 @@ export function selectStages(stages?: string[], skipTo?: string): string[] {
|
||||
return selected.slice(skipIndex);
|
||||
}
|
||||
|
||||
/**
|
||||
* Fail closed when the required executor capability is missing (SDLC-D-035).
|
||||
*/
|
||||
function requireExecutor(executor: TaskExecutor | undefined, simulate: boolean): TaskExecutor {
|
||||
if (executor) return executor;
|
||||
if (simulate) return createSimulatedExecutor({ log: false });
|
||||
throw new ForgeCapabilityError(
|
||||
'FORGE_NO_EXECUTOR',
|
||||
'task-executor',
|
||||
'no task executor is wired; refusing to run the pipeline with a stub executor (fail closed). ' +
|
||||
'Pass --simulate to opt into explicitly simulated execution.',
|
||||
);
|
||||
}
|
||||
|
||||
/**
|
||||
* Pre-flight a stage's gates in normal mode (fail closed, SDLC-D-035).
|
||||
*
|
||||
* - authority gates: record a typed `waiting-for-authority` stage result and
|
||||
* raise FORGE_AUTHORITY_REQUIRED — approval-based gates never pass vacuously.
|
||||
* - provider gates: record a typed `blocked` stage result and raise the typed
|
||||
* capability error for the missing provider.
|
||||
*
|
||||
* Returns the stage status to record when the pre-flight blocks, or undefined
|
||||
* when the stage may proceed.
|
||||
*/
|
||||
function preflightStageGates(
|
||||
stageName: string,
|
||||
manifest: RunManifest,
|
||||
): { status: StageStatus; error: ForgeCapabilityError } | undefined {
|
||||
const spec = STAGE_SPECS[stageName];
|
||||
if (!spec) throw new Error(`Unknown Forge stage: ${stageName}`);
|
||||
|
||||
for (const gate of spec.qualityGates) {
|
||||
if (!isCapabilityGate(gate)) continue;
|
||||
|
||||
const startedAt = manifest.stages[stageName]?.startedAt;
|
||||
const completedAt = nowISO();
|
||||
|
||||
if (gate.kind === 'authority') {
|
||||
const reason = `gate '${gate.capability}' requires authority sign-off; no mechanical implementation exists (${gate.reason})`;
|
||||
return {
|
||||
status: {
|
||||
status: 'waiting-for-authority',
|
||||
reason,
|
||||
startedAt,
|
||||
completedAt,
|
||||
gateResults: waitingGateResults(spec.qualityGates, reason),
|
||||
},
|
||||
error: new ForgeCapabilityError(
|
||||
'FORGE_AUTHORITY_REQUIRED',
|
||||
gate.capability,
|
||||
`stage '${stageName}' is blocked on authority gate '${gate.capability}': ${gate.reason}. ` +
|
||||
'The pipeline fails closed instead of passing vacuously. Record the approval out-of-band ' +
|
||||
'or run with --simulate for explicitly simulated execution.',
|
||||
),
|
||||
};
|
||||
}
|
||||
|
||||
const reason = `gate '${gate.capability}' requires provider '${gate.capability}' and none is wired (${gate.reason})`;
|
||||
return {
|
||||
status: {
|
||||
status: 'blocked',
|
||||
reason,
|
||||
startedAt,
|
||||
completedAt,
|
||||
gateResults: blockedGateResults(spec.qualityGates, reason),
|
||||
},
|
||||
error: new ForgeCapabilityError(
|
||||
providerErrorCode(gate.capability),
|
||||
gate.capability,
|
||||
`stage '${stageName}' requires provider '${gate.capability}' which is not wired: ${gate.reason}. ` +
|
||||
'The pipeline fails closed instead of passing vacuously.',
|
||||
),
|
||||
};
|
||||
}
|
||||
|
||||
return undefined;
|
||||
}
|
||||
|
||||
/**
|
||||
* Execute the given stage tasks sequentially, updating the manifest.
|
||||
*
|
||||
* Normal mode requires a real executor and evaluates every declared command
|
||||
* gate through the typed outcome model; any non-verified result fails closed.
|
||||
* Simulate mode types every stage and gate result as `simulated`.
|
||||
*/
|
||||
async function executeStages(opts: {
|
||||
manifest: RunManifest;
|
||||
runDir: string;
|
||||
tasks: ForgeTask[];
|
||||
stageNames: string[];
|
||||
executor: TaskExecutor;
|
||||
simulate: boolean;
|
||||
}): Promise<void> {
|
||||
const { manifest, runDir, tasks, stageNames, executor, simulate } = opts;
|
||||
|
||||
for (let i = 0; i < tasks.length; i++) {
|
||||
const task = tasks[i]!;
|
||||
const stageName = stageNames[i]!;
|
||||
const spec = STAGE_SPECS[stageName];
|
||||
if (!spec) throw new Error(`Unknown Forge stage: ${stageName}`);
|
||||
|
||||
// Update manifest: stage in progress
|
||||
manifest.currentStage = stageName;
|
||||
manifest.stages[stageName] = {
|
||||
status: 'in_progress',
|
||||
startedAt: nowISO(),
|
||||
};
|
||||
saveManifest(runDir, manifest);
|
||||
|
||||
// Fail-closed pre-flight (normal mode only): authority/provider gates have
|
||||
// no mechanical implementation and must never pass vacuously.
|
||||
if (!simulate) {
|
||||
const blocked = preflightStageGates(stageName, manifest);
|
||||
if (blocked) {
|
||||
manifest.stages[stageName] = blocked.status;
|
||||
manifest.status =
|
||||
blocked.status.status === 'waiting-for-authority' ? 'waiting-for-authority' : 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw blocked.error;
|
||||
}
|
||||
}
|
||||
|
||||
let result: ForgeTaskResult;
|
||||
try {
|
||||
await executor.submitTask(task);
|
||||
result = await executor.waitForCompletion(task.id, task.timeoutSeconds * 1000);
|
||||
} catch (error) {
|
||||
// Process errors (including timeouts) map to the fail-closed `error` outcome.
|
||||
const reason = error instanceof Error ? error.message : String(error);
|
||||
manifest.stages[stageName] = {
|
||||
status: 'error',
|
||||
reason: `executor error: ${reason}`,
|
||||
startedAt: manifest.stages[stageName]?.startedAt,
|
||||
completedAt: nowISO(),
|
||||
gateResults: [],
|
||||
};
|
||||
manifest.status = 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw error instanceof Error ? error : new Error(reason);
|
||||
}
|
||||
|
||||
if (simulate) {
|
||||
manifest.stages[stageName] = {
|
||||
status: 'simulated',
|
||||
reason: SIMULATED_STAGE_REASON,
|
||||
startedAt: manifest.stages[stageName]?.startedAt,
|
||||
completedAt: nowISO(),
|
||||
gateResults: simulatedGateResults(spec.qualityGates),
|
||||
};
|
||||
saveManifest(runDir, manifest);
|
||||
continue;
|
||||
}
|
||||
|
||||
const evaluation = evaluateStageGates(stageName, spec.qualityGates, result);
|
||||
manifest.stages[stageName] = {
|
||||
status: evaluation.outcome,
|
||||
reason: evaluation.reason,
|
||||
startedAt: manifest.stages[stageName]?.startedAt,
|
||||
completedAt: nowISO(),
|
||||
gateResults: evaluation.gateResults,
|
||||
};
|
||||
|
||||
if (evaluation.outcome !== 'passed') {
|
||||
manifest.status =
|
||||
evaluation.outcome === 'waiting-for-authority' ? 'waiting-for-authority' : 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw new Error(`Stage ${stageName} ${evaluation.outcome}: ${evaluation.reason}`);
|
||||
}
|
||||
|
||||
saveManifest(runDir, manifest);
|
||||
}
|
||||
}
|
||||
|
||||
/**
|
||||
* Run the Forge pipeline.
|
||||
*
|
||||
* 1. Fail closed unless a real executor is wired or simulation is explicit
|
||||
* 2. Classify the brief
|
||||
* 3. Generate a run ID and create run directory
|
||||
* 4. Map stages to tasks and submit to TaskExecutor
|
||||
* 5. Track manifest with typed stage outcomes
|
||||
* 6. Return pipeline result
|
||||
* 1. Classify the brief
|
||||
* 2. Generate a run ID and create run directory
|
||||
* 3. Map stages to tasks and submit to TaskExecutor
|
||||
* 4. Track manifest with stage statuses
|
||||
* 5. Return pipeline result
|
||||
*/
|
||||
export async function runPipeline(
|
||||
briefPath: string,
|
||||
projectRoot: string,
|
||||
options: PipelineOptions,
|
||||
): Promise<PipelineResult> {
|
||||
const simulate = options.simulate ?? false;
|
||||
const executor = requireExecutor(options.executor, simulate);
|
||||
const mode: RunMode = simulate ? 'simulated' : 'normal';
|
||||
|
||||
const resolvedRoot = path.resolve(projectRoot);
|
||||
const resolvedBrief = path.resolve(briefPath);
|
||||
const briefContent = fs.readFileSync(resolvedBrief, 'utf-8');
|
||||
@@ -342,7 +146,6 @@ export async function runPipeline(
|
||||
briefClass,
|
||||
classSource,
|
||||
forceBoard: options.forceBoard ?? false,
|
||||
mode,
|
||||
runDir,
|
||||
});
|
||||
|
||||
@@ -369,10 +172,54 @@ export async function runPipeline(
|
||||
}
|
||||
|
||||
// Execute stages
|
||||
await executeStages({ manifest, runDir, tasks, stageNames: selectedStages, executor, simulate });
|
||||
const { executor } = options;
|
||||
for (let i = 0; i < tasks.length; i++) {
|
||||
const task = tasks[i]!;
|
||||
const stageName = selectedStages[i]!;
|
||||
|
||||
// All stages reached a terminal state for this mode
|
||||
manifest.status = simulate ? 'simulated' : 'completed';
|
||||
// Update manifest: stage in progress
|
||||
manifest.currentStage = stageName;
|
||||
manifest.stages[stageName] = {
|
||||
status: 'in_progress',
|
||||
startedAt: nowISO(),
|
||||
};
|
||||
saveManifest(runDir, manifest);
|
||||
|
||||
try {
|
||||
await executor.submitTask(task);
|
||||
const result = await executor.waitForCompletion(task.id, task.timeoutSeconds * 1000);
|
||||
|
||||
// Update manifest: stage completed or failed
|
||||
const stageStatus: StageStatus = {
|
||||
status: result.status === 'completed' ? 'passed' : 'failed',
|
||||
startedAt: manifest.stages[stageName]!.startedAt,
|
||||
completedAt: nowISO(),
|
||||
};
|
||||
manifest.stages[stageName] = stageStatus;
|
||||
|
||||
if (result.status !== 'completed') {
|
||||
manifest.status = 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw new Error(`Stage ${stageName} failed with status: ${result.status}`);
|
||||
}
|
||||
|
||||
saveManifest(runDir, manifest);
|
||||
} catch (error) {
|
||||
if (!manifest.stages[stageName]?.completedAt) {
|
||||
manifest.stages[stageName] = {
|
||||
status: 'failed',
|
||||
startedAt: manifest.stages[stageName]?.startedAt,
|
||||
completedAt: nowISO(),
|
||||
};
|
||||
}
|
||||
manifest.status = 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
|
||||
// All stages passed
|
||||
manifest.status = 'completed';
|
||||
saveManifest(runDir, manifest);
|
||||
|
||||
return {
|
||||
@@ -387,30 +234,22 @@ export async function runPipeline(
|
||||
}
|
||||
|
||||
/**
|
||||
* Resume a pipeline from the last non-passed stage.
|
||||
* Resume a pipeline from the last incomplete stage.
|
||||
*/
|
||||
export async function resumePipeline(
|
||||
runDir: string,
|
||||
executor?: TaskExecutor,
|
||||
options?: { simulate?: boolean },
|
||||
executor: TaskExecutor,
|
||||
): Promise<PipelineResult> {
|
||||
const simulate = options?.simulate ?? false;
|
||||
const wiredExecutor = requireExecutor(executor, simulate);
|
||||
const mode: RunMode = simulate ? 'simulated' : 'normal';
|
||||
|
||||
const manifest = loadManifest(runDir);
|
||||
const resolvedRoot = path.dirname(path.dirname(path.dirname(runDir))); // .forge/runs/{id} → project root
|
||||
|
||||
const briefContent = fs.readFileSync(manifest.brief, 'utf-8');
|
||||
const allStages = stagesForClass(manifest.briefClass, manifest.forceBoard);
|
||||
|
||||
manifest.mode = mode;
|
||||
|
||||
// Find first non-satisfying stage (only a verified `passed` counts as done;
|
||||
// simulated and waiting-for-authority stages are re-run).
|
||||
// Find first non-passed stage
|
||||
const resumeFrom = allStages.find((s) => manifest.stages[s]?.status !== 'passed');
|
||||
if (!resumeFrom) {
|
||||
manifest.status = mode === 'simulated' ? 'simulated' : 'completed';
|
||||
manifest.status = 'completed';
|
||||
saveManifest(runDir, manifest);
|
||||
return {
|
||||
runId: manifest.runId,
|
||||
@@ -445,16 +284,49 @@ export async function resumePipeline(
|
||||
tasks.push(task);
|
||||
}
|
||||
|
||||
await executeStages({
|
||||
manifest,
|
||||
runDir,
|
||||
tasks,
|
||||
stageNames: remainingStages,
|
||||
executor: wiredExecutor,
|
||||
simulate,
|
||||
});
|
||||
for (let i = 0; i < tasks.length; i++) {
|
||||
const task = tasks[i]!;
|
||||
const stageName = remainingStages[i]!;
|
||||
|
||||
manifest.status = simulate ? 'simulated' : 'completed';
|
||||
manifest.currentStage = stageName;
|
||||
manifest.stages[stageName] = {
|
||||
status: 'in_progress',
|
||||
startedAt: nowISO(),
|
||||
};
|
||||
saveManifest(runDir, manifest);
|
||||
|
||||
try {
|
||||
await executor.submitTask(task);
|
||||
const result = await executor.waitForCompletion(task.id, task.timeoutSeconds * 1000);
|
||||
|
||||
manifest.stages[stageName] = {
|
||||
status: result.status === 'completed' ? 'passed' : 'failed',
|
||||
startedAt: manifest.stages[stageName]!.startedAt,
|
||||
completedAt: nowISO(),
|
||||
};
|
||||
|
||||
if (result.status !== 'completed') {
|
||||
manifest.status = 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw new Error(`Stage ${stageName} failed with status: ${result.status}`);
|
||||
}
|
||||
|
||||
saveManifest(runDir, manifest);
|
||||
} catch (error) {
|
||||
if (!manifest.stages[stageName]?.completedAt) {
|
||||
manifest.stages[stageName] = {
|
||||
status: 'failed',
|
||||
startedAt: manifest.stages[stageName]?.startedAt,
|
||||
completedAt: nowISO(),
|
||||
};
|
||||
}
|
||||
manifest.status = 'failed';
|
||||
saveManifest(runDir, manifest);
|
||||
throw error;
|
||||
}
|
||||
}
|
||||
|
||||
manifest.status = 'completed';
|
||||
saveManifest(runDir, manifest);
|
||||
|
||||
return {
|
||||
|
||||
@@ -1,32 +0,0 @@
|
||||
import type { ForgeTask, ForgeTaskResult, TaskExecutor } from './types.js';
|
||||
|
||||
/**
|
||||
* Simulated executor — used ONLY when the caller explicitly passes --simulate.
|
||||
*
|
||||
* It submits no real work and returns typed `simulated` results so a simulated
|
||||
* run can never be confused with a verified one. In normal mode (no --simulate)
|
||||
* the CLI refuses to run at all with FORGE_NO_EXECUTOR instead of wiring this
|
||||
* stub (fail closed, SDLC-D-035).
|
||||
*/
|
||||
export function createSimulatedExecutor(options?: { log?: boolean }): TaskExecutor {
|
||||
const log = options?.log ?? true;
|
||||
return {
|
||||
async submitTask(task: ForgeTask) {
|
||||
if (log) console.log(` [forge:simulated] stage submitted: ${task.id} (${task.title})`);
|
||||
},
|
||||
async waitForCompletion(taskId: string): Promise<ForgeTaskResult> {
|
||||
if (log) console.log(` [forge:simulated] stage complete: ${taskId}`);
|
||||
return {
|
||||
task_id: taskId,
|
||||
outcome: 'simulated',
|
||||
reason: 'no executor wired; simulated execution requested via --simulate',
|
||||
completed_at: new Date().toISOString(),
|
||||
exit_code: 0,
|
||||
gate_results: [],
|
||||
};
|
||||
},
|
||||
async getTaskStatus() {
|
||||
return 'completed' as const;
|
||||
},
|
||||
};
|
||||
}
|
||||
@@ -1,4 +1,4 @@
|
||||
import type { GateEntry } from '@mosaicstack/macp';
|
||||
import type { GateEntry, TaskResult } from '@mosaicstack/macp';
|
||||
|
||||
/** Stage dispatch mode. */
|
||||
export type StageDispatch = 'exec' | 'yolo' | 'pi';
|
||||
@@ -6,58 +6,6 @@ export type StageDispatch = 'exec' | 'yolo' | 'pi';
|
||||
/** Stage type — determines agent selection and gate requirements. */
|
||||
export type StageType = 'research' | 'review' | 'coding' | 'deploy';
|
||||
|
||||
/**
|
||||
* Typed outcome for every gate and stage evaluation — closed set (SDLC-D-035).
|
||||
*
|
||||
* Only `passed` means "verified by a real implementation". `simulated` is
|
||||
* produced exclusively in explicit `--simulate` runs and is never satisfying.
|
||||
*/
|
||||
export type ForgeOutcome =
|
||||
| 'passed'
|
||||
| 'failed'
|
||||
| 'blocked'
|
||||
| 'error'
|
||||
| 'waiting-for-authority'
|
||||
| 'simulated'
|
||||
| 'not-applicable';
|
||||
|
||||
/** A gate that requires authority (human/board) sign-off; no mechanical command can satisfy it. */
|
||||
export interface AuthorityGate {
|
||||
kind: 'authority';
|
||||
capability: string;
|
||||
reason: string;
|
||||
}
|
||||
|
||||
/** A gate that requires a wired provider (e.g. an AI reviewer, CI pipeline) to evaluate. */
|
||||
export interface ProviderGate {
|
||||
kind: 'provider';
|
||||
capability: string;
|
||||
reason: string;
|
||||
}
|
||||
|
||||
/** Forge quality gate: a real command, an authority sign-off, or a provider-backed check. */
|
||||
export type ForgeGate = string | GateEntry | AuthorityGate | ProviderGate;
|
||||
|
||||
/** Typed result of evaluating a single quality gate. */
|
||||
export interface ForgeGateResult {
|
||||
gate: string;
|
||||
outcome: ForgeOutcome;
|
||||
reason: string;
|
||||
exitCode?: number;
|
||||
output?: string;
|
||||
timedOut?: boolean;
|
||||
}
|
||||
|
||||
/** Typed result of a task/stage execution returned by a TaskExecutor. */
|
||||
export interface ForgeTaskResult {
|
||||
task_id: string;
|
||||
outcome: ForgeOutcome;
|
||||
reason: string;
|
||||
completed_at: string;
|
||||
exit_code: number;
|
||||
gate_results: ForgeGateResult[];
|
||||
}
|
||||
|
||||
/** Stage specification — defines a single pipeline stage. */
|
||||
export interface StageSpec {
|
||||
number: string;
|
||||
@@ -66,7 +14,7 @@ export interface StageSpec {
|
||||
type: StageType;
|
||||
gate: string;
|
||||
promptFile: string;
|
||||
qualityGates: ForgeGate[];
|
||||
qualityGates: (string | GateEntry)[];
|
||||
}
|
||||
|
||||
/** Brief classification. */
|
||||
@@ -77,18 +25,11 @@ export type ClassSource = 'cli' | 'frontmatter' | 'auto';
|
||||
|
||||
/** Per-stage status within a run manifest. */
|
||||
export interface StageStatus {
|
||||
status: 'pending' | 'in_progress' | ForgeOutcome;
|
||||
/** Why the stage reached its current (terminal) outcome, when applicable. */
|
||||
reason?: string;
|
||||
status: 'pending' | 'in_progress' | 'passed' | 'failed';
|
||||
startedAt?: string;
|
||||
completedAt?: string;
|
||||
/** Typed per-gate results recorded alongside the stage outcome. */
|
||||
gateResults?: ForgeGateResult[];
|
||||
}
|
||||
|
||||
/** Execution mode of a run. */
|
||||
export type RunMode = 'normal' | 'simulated';
|
||||
|
||||
/** Run manifest — persisted to disk as manifest.json. */
|
||||
export interface RunManifest {
|
||||
runId: string;
|
||||
@@ -97,23 +38,10 @@ export interface RunManifest {
|
||||
briefClass: BriefClass;
|
||||
classSource: ClassSource;
|
||||
forceBoard: boolean;
|
||||
/**
|
||||
* Execution mode. `simulated` runs stub execution; their results are typed
|
||||
* `simulated` and must never be read as verified success. Optional because
|
||||
* manifests written before this field existed default to `normal`.
|
||||
*/
|
||||
mode?: RunMode;
|
||||
createdAt: string;
|
||||
updatedAt: string;
|
||||
currentStage: string;
|
||||
status:
|
||||
| 'in_progress'
|
||||
| 'completed'
|
||||
| 'failed'
|
||||
| 'interrupted'
|
||||
| 'rejected'
|
||||
| 'simulated'
|
||||
| 'waiting-for-authority';
|
||||
status: 'in_progress' | 'completed' | 'failed' | 'interrupted' | 'rejected';
|
||||
stages: Record<string, StageStatus>;
|
||||
}
|
||||
|
||||
@@ -137,7 +65,7 @@ export interface ForgeTask {
|
||||
briefPath: string;
|
||||
resultPath: string;
|
||||
timeoutSeconds: number;
|
||||
qualityGates: ForgeGate[];
|
||||
qualityGates: (string | GateEntry)[];
|
||||
worktree?: string;
|
||||
command?: string;
|
||||
dependsOn?: string[];
|
||||
@@ -148,7 +76,7 @@ export interface ForgeTask {
|
||||
/** Abstract task executor — decouples from packages/coord. */
|
||||
export interface TaskExecutor {
|
||||
submitTask(task: ForgeTask): Promise<void>;
|
||||
waitForCompletion(taskId: string, timeoutMs: number): Promise<ForgeTaskResult>;
|
||||
waitForCompletion(taskId: string, timeoutMs: number): Promise<TaskResult>;
|
||||
getTaskStatus(taskId: string): Promise<ForgeTaskStatus>;
|
||||
}
|
||||
|
||||
@@ -194,16 +122,7 @@ export interface PipelineOptions {
|
||||
stages?: string[];
|
||||
skipTo?: string;
|
||||
dryRun?: boolean;
|
||||
/**
|
||||
* Real task executor. Required in normal mode: the pipeline fails closed
|
||||
* with FORGE_NO_EXECUTOR when it is absent.
|
||||
*/
|
||||
executor?: TaskExecutor;
|
||||
/**
|
||||
* Explicit opt-in to simulated execution. Every stage and gate result is
|
||||
* typed `simulated` and is never satisfying.
|
||||
*/
|
||||
simulate?: boolean;
|
||||
executor: TaskExecutor;
|
||||
}
|
||||
|
||||
/** Pipeline run result. */
|
||||
|
||||
@@ -255,68 +255,6 @@ fleet_declared_transport() {
|
||||
printf '%s\n' "${declared:-tmux}"
|
||||
}
|
||||
|
||||
# Brain-home fleet-state resolution (#1298; canon STRUCTURE-CANON §2).
|
||||
#
|
||||
# Seat launch envs, roles.local overrides, and profile working copies resolve
|
||||
# from the brain home when one is active; roster, baseline roles, run/, and
|
||||
# services stay under MOSAIC_HOME. This check surfaces which tree fleet state
|
||||
# resolves from and the drift a launch would otherwise hit at runtime:
|
||||
#
|
||||
# - a stale MOSAIC_BRAIN_HOME pointing at a directory with no fleet/agents is a
|
||||
# misconfiguration the resolver honors (explicit wins) — warn, don't pass;
|
||||
# - a symlinked brain or agents dir defeats the managed-directory boundary;
|
||||
# - a group/world-readable agents dir violates the 0700 projection boundary;
|
||||
# - env files left in the config-home tree while a brain is active are split
|
||||
# state — the write path rejects NEW split writes, but nothing would ever
|
||||
# tell the operator the old files are stranded.
|
||||
resolve_brain_home() {
|
||||
local explicit="${MOSAIC_BRAIN_HOME:-}"
|
||||
if [[ -n "$(printf '%s' "$explicit" | tr -d '[:space:]')" ]]; then
|
||||
printf '%s' "$explicit"
|
||||
return
|
||||
fi
|
||||
if [[ "$(cd "$MOSAIC_HOME" 2>/dev/null && pwd -P)" == "$HOME/.config/mosaic" \
|
||||
&& -d "$HOME/.mosaic/fleet/agents" ]]; then
|
||||
printf '%s' "$HOME/.mosaic"
|
||||
return
|
||||
fi
|
||||
printf '%s' "$MOSAIC_HOME"
|
||||
}
|
||||
|
||||
check_brain_home() {
|
||||
local brain agents mode
|
||||
brain="$(resolve_brain_home)"
|
||||
|
||||
if [[ "$brain" == "$MOSAIC_HOME" ]]; then
|
||||
pass "Fleet state home: $MOSAIC_HOME (legacy single-tree; no brain adopted)"
|
||||
return
|
||||
fi
|
||||
|
||||
agents="$brain/fleet/agents"
|
||||
if [[ ! -d "$agents" ]]; then
|
||||
warn "Brain home '$brain' has no fleet/agents — seat envs will not resolve from it. Point MOSAIC_BRAIN_HOME at a brain carrying fleet/agents, or unset it."
|
||||
return
|
||||
fi
|
||||
if [[ -L "$brain" || -L "$agents" ]]; then
|
||||
warn "Brain fleet-state path resolves through a symlink ($brain) — the managed-directory boundary requires regular directories."
|
||||
return
|
||||
fi
|
||||
|
||||
mode="$(stat -c '%a' -- "$agents" 2>/dev/null)" || mode=""
|
||||
if [[ -n "$mode" ]] && (( (8#$mode & 8#077) != 0 )); then
|
||||
warn "Brain agents dir '$agents' is group/world-accessible (mode $mode) — the projection boundary requires 0700."
|
||||
return
|
||||
fi
|
||||
|
||||
if [[ -d "$MOSAIC_HOME/fleet/agents" ]] \
|
||||
&& ls "$MOSAIC_HOME/fleet/agents/"*.env* >/dev/null 2>&1; then
|
||||
warn "Fleet env files exist in BOTH trees — brain '$brain' is active but '$MOSAIC_HOME/fleet/agents' still carries env files (split state). Migrate them (mosaic fleet regen) and remove the config-home copies."
|
||||
return
|
||||
fi
|
||||
|
||||
pass "Fleet state home: $brain (brain active); roster + templates: $MOSAIC_HOME"
|
||||
}
|
||||
|
||||
check_fleet_transport() {
|
||||
local transport
|
||||
transport="$(fleet_declared_transport)"
|
||||
@@ -335,8 +273,6 @@ check_fleet_transport() {
|
||||
|
||||
check_fleet_transport
|
||||
|
||||
check_brain_home
|
||||
|
||||
# Legacy migration surfaces should no longer contain symlink trees.
|
||||
legacy_paths=(
|
||||
"$HOME/.claude/agent-guides"
|
||||
|
||||
@@ -1,108 +0,0 @@
|
||||
#!/usr/bin/env bash
|
||||
# Covers the brain-home fleet-state check in `mosaic-doctor` (#1298 follow-up).
|
||||
#
|
||||
# The functions are extracted from the shipped script rather than copied here
|
||||
# (same discipline as test-fleet-transport-check.sh): a test that carries its
|
||||
# own copy of the logic keeps passing after the shipped copy changes.
|
||||
# Extraction is by exact function header and a closing brace in column one.
|
||||
set -euo pipefail
|
||||
|
||||
SCRIPT_DIR=$(cd -- "$(dirname "$0")" && pwd)
|
||||
DOCTOR="$SCRIPT_DIR/mosaic-doctor"
|
||||
|
||||
fail() {
|
||||
echo "FAIL: $*" >&2
|
||||
exit 1
|
||||
}
|
||||
|
||||
[ -f "$DOCTOR" ] || fail "missing mosaic-doctor at $DOCTOR"
|
||||
|
||||
extract_function() {
|
||||
local name="$1"
|
||||
local extracted
|
||||
extracted=$(sed -n "/^${name}() {/,/^}/p" "$DOCTOR")
|
||||
[ -n "$extracted" ] || fail "could not extract ${name}() from mosaic-doctor — script reshaped?"
|
||||
printf '%s\n' "$extracted"
|
||||
}
|
||||
|
||||
for fn in resolve_brain_home check_brain_home; do
|
||||
extract_function "$fn" >/dev/null
|
||||
done
|
||||
|
||||
warn_count=0
|
||||
warn() { warn_count=$((warn_count + 1)); echo "[WARN] $*"; }
|
||||
pass() { echo "[OK] $*"; return 0; }
|
||||
|
||||
eval "$(extract_function resolve_brain_home)"
|
||||
eval "$(extract_function check_brain_home)"
|
||||
|
||||
ROOT=$(mktemp -d)
|
||||
trap 'rm -rf "$ROOT"' EXIT
|
||||
|
||||
run_case() {
|
||||
# label, expect (ok|warn), then env assignments as arguments.
|
||||
# The check runs under `env` in a subshell, so its warn() also prints a
|
||||
# sentinel the parent counts — a subshell counter would never be visible.
|
||||
local label="$1" expect="$2"
|
||||
shift 2
|
||||
local out warns
|
||||
out=$(env "$@" bash -c "warn() { echo \"[WARN] \$*\"; }; pass() { echo \"[OK] \$*\"; return 0; }; $(extract_function resolve_brain_home); $(extract_function check_brain_home); check_brain_home" 2>&1)
|
||||
warns=$(printf '%s\n' "$out" | grep -c '^\[WARN\]' || true)
|
||||
if [[ "$expect" == ok && "$warns" -eq 0 ]]; then
|
||||
echo "ok - $label"
|
||||
elif [[ "$expect" == warn && "$warns" -gt 0 ]]; then
|
||||
echo "ok - $label (warned)"
|
||||
else
|
||||
echo "output: $out" >&2
|
||||
fail "$label: expected $expect (warns=$warns)"
|
||||
fi
|
||||
}
|
||||
|
||||
# ── legacy: no brain, custom home never adopts ─────────────────────────────
|
||||
mkdir -p "$ROOT/legacy-mosaic/fleet/agents"
|
||||
run_case "custom home without brain stays legacy" ok \
|
||||
MOSAIC_HOME="$ROOT/legacy-mosaic" HOME="$ROOT"
|
||||
|
||||
# ── healthy brain at the default config home ───────────────────────────────
|
||||
mkdir -p "$ROOT/home/.config/mosaic" "$ROOT/home/.mosaic/fleet/agents"
|
||||
chmod 700 "$ROOT/home/.mosaic/fleet/agents"
|
||||
run_case "default home adopts healthy brain" ok \
|
||||
MOSAIC_HOME="$ROOT/home/.config/mosaic" HOME="$ROOT/home"
|
||||
|
||||
# ── explicit MOSAIC_BRAIN_HOME to a brain without fleet/agents → warn ──────
|
||||
mkdir -p "$ROOT/brain-noagents/fleet" "$ROOT/config"
|
||||
run_case "explicit brain without agents warns" warn \
|
||||
MOSAIC_HOME="$ROOT/config" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-noagents"
|
||||
|
||||
# ── explicit MOSAIC_BRAIN_HOME to a healthy brain → ok ─────────────────────
|
||||
mkdir -p "$ROOT/brain-ok/fleet/agents" "$ROOT/config2"
|
||||
chmod 700 "$ROOT/brain-ok/fleet/agents"
|
||||
run_case "explicit healthy brain passes" ok \
|
||||
MOSAIC_HOME="$ROOT/config2" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-ok"
|
||||
|
||||
# ── group-readable agents dir → warn (0700 boundary) ───────────────────────
|
||||
mkdir -p "$ROOT/brain-loose/fleet/agents" "$ROOT/config3"
|
||||
chmod 750 "$ROOT/brain-loose/fleet/agents"
|
||||
run_case "group-readable brain agents warns" warn \
|
||||
MOSAIC_HOME="$ROOT/config3" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-loose"
|
||||
|
||||
# ── symlinked agents dir → warn (managed-directory boundary) ───────────────
|
||||
mkdir -p "$ROOT/brain-link/real-agents" "$ROOT/brain-link/fleet" "$ROOT/config4"
|
||||
ln -s "$ROOT/brain-link/real-agents" "$ROOT/brain-link/fleet/agents"
|
||||
run_case "symlinked brain agents warns" warn \
|
||||
MOSAIC_HOME="$ROOT/config4" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-link"
|
||||
|
||||
# ── split state: envs in BOTH trees → warn ─────────────────────────────────
|
||||
mkdir -p "$ROOT/brain-split/fleet/agents" "$ROOT/config5/fleet/agents"
|
||||
chmod 700 "$ROOT/brain-split/fleet/agents" "$ROOT/config5/fleet/agents"
|
||||
touch "$ROOT/config5/fleet/agents/coder0.env.generated"
|
||||
run_case "env files in both trees warns (split state)" warn \
|
||||
MOSAIC_HOME="$ROOT/config5" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-split"
|
||||
|
||||
# ── config-home agents dir WITHOUT env files alongside a brain → ok ────────
|
||||
mkdir -p "$ROOT/brain-clean/fleet/agents" "$ROOT/config6/fleet/agents"
|
||||
chmod 700 "$ROOT/brain-clean/fleet/agents" "$ROOT/config6/fleet/agents"
|
||||
run_case "empty config-home agents dir alongside brain passes" ok \
|
||||
MOSAIC_HOME="$ROOT/config6" HOME="$ROOT" MOSAIC_BRAIN_HOME="$ROOT/brain-clean"
|
||||
|
||||
echo "ok - mosaic-doctor brain-home check"
|
||||
@@ -25,7 +25,7 @@
|
||||
"lint": "eslint src",
|
||||
"typecheck": "tsc --noEmit",
|
||||
"test": "vitest run --passWithNoTests && pnpm run test:framework-shell",
|
||||
"test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh && bash framework/tools/glpi/test-list-http-status.sh && bash framework/tools/orchestrator/test-board-roll.sh && bash framework/tools/woodpecker/test-ci-wait-exit-matrix.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh && bash framework/tools/_scripts/test-brain-home-check.sh"
|
||||
"test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh && bash framework/tools/glpi/test-list-http-status.sh && bash framework/tools/orchestrator/test-board-roll.sh && bash framework/tools/woodpecker/test-ci-wait-exit-matrix.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh"
|
||||
},
|
||||
"dependencies": {
|
||||
"@mosaicstack/brain": "workspace:*",
|
||||
|
||||
@@ -0,0 +1,45 @@
|
||||
# RI-1-002 — Publish-gate negative controls (SDLC-D-034 second half)
|
||||
|
||||
- Task: RI-1-002 (docs/release-integrity workstream, PRD item RI-N1), issue ref #1275
|
||||
- Branch: `test/ri-050-publish-gate-negative` (base `origin/next` @ d8e0aec9 = PR #1277, RI-1-001)
|
||||
- Budget: worker estimate ~45K tokens; keep scoped to the two test files + scratchpad.
|
||||
|
||||
## Objective
|
||||
|
||||
Checked-in negative-control tests that PROVE the publish gate fails when it must:
|
||||
|
||||
1. Broken mandatory check blocks every publish step (structural DAG proof from `.woodpecker/publish.yml`).
|
||||
2. Bypass shapes fail the checker: missing edge, hidden effect (non-`publish` name), detached verify, always-pass verify (`failure: ignore` / `success` override), conditional verify (`when`).
|
||||
3. Exact-commit identity: no HEAD-moving step between verify and publish effects; legitimate re-checkout requires verify to re-run after it.
|
||||
4. `verify-release.mjs` composition control: a SUBSET stage list fails the composition check.
|
||||
|
||||
## Plan
|
||||
|
||||
- NEW `scripts/publish-gate-structure.test.mjs` — self-contained structural checker (`assertPublishGateBlocksOnVerify`) + positive control on the real pipeline + one negative-control test per bypass shape (S1–S6, documented in file header) + positive control for the legitimate re-checkout shape.
|
||||
- EXTEND `scripts/verify-release.test.mjs` — refactor the stage-mirror test body into `assertStagesMirrorCi(stages, ci)`; add negative control dropping each stage one at a time (subset must throw).
|
||||
|
||||
## Conventions confirmed
|
||||
|
||||
- Root `test:checkout` = `node --test scripts/*.test.mjs` → new file auto-joins `pnpm test`.
|
||||
- Test-enumeration guard population is `*test*.sh` under `packages/mosaic/framework/tools/` only → unaffected.
|
||||
- Root eslint covers only `**/*.{ts,tsx}` → .mjs files need Prettier style only (printWidth 100, singleQuote, semi, trailingComma all).
|
||||
- Do NOT touch docs/TASKS.md, docs/release-integrity/TASKS.md, docs/scratchpads/.
|
||||
|
||||
## Progress log
|
||||
|
||||
- [x] Base verified: publish.yml `verify` step + verify-release.mjs present; HEAD contains origin/next.
|
||||
- [x] Wrote scripts/publish-gate-structure.test.mjs
|
||||
- [x] Extended scripts/verify-release.test.mjs (mirror fn + subset negative control)
|
||||
- [x] Gates: node --test scripts (31 tests pass), prettier clean on touched files, pnpm typecheck PASS, pnpm lint PASS, pnpm format:check PASS
|
||||
- [x] Committed ff585b88 + pushed, PR #1305 → next (no conflicts). Stopped before merge per task instruction.
|
||||
|
||||
## Evidence
|
||||
|
||||
- `node --test scripts/verify-release.test.mjs scripts/publish-gate-structure.test.mjs` → 31 tests, 0 fail.
|
||||
- Mutation sanity: temporarily removing the `verify` edge from build-gateway in publish.yml → structure test goes red (verified manually during dev, then reverted).
|
||||
- Gates run from repo root on this worktree; results in Progress log.
|
||||
|
||||
## Risks / notes
|
||||
|
||||
- Effect detection (`isPublishCommand`) is deliberately over-broad (any npm/pnpm/yarn command mentioning `publish`, any kaniko/docker-push/`--destination`) — fail-closed: a false positive forces justification, a false negative is the actual hazard.
|
||||
- `git fetch` flagged as HEAD-moving even though fetch alone doesn't move HEAD — fail-closed on the classic `fetch && reset` pair.
|
||||
@@ -0,0 +1,310 @@
|
||||
import assert from 'node:assert/strict';
|
||||
import { readFile } from 'node:fs/promises';
|
||||
import { createRequire } from 'node:module';
|
||||
import path from 'node:path';
|
||||
import test from 'node:test';
|
||||
|
||||
// RI-1-002 / RI-N1 publish-gate NEGATIVE CONTROLS (SDLC-D-034).
|
||||
//
|
||||
// scripts/verify-release.test.mjs pins the POSITIVE structure of the publish
|
||||
// gate: every publish effect declares a direct `depends_on: verify` edge and
|
||||
// the verify step asserts commit identity + runs the canonical command. This
|
||||
// suite is the negative-control set: each test feeds a structural gate
|
||||
// checker a pipeline in which the gate is bypassed by ONE specific shape and
|
||||
// asserts the checker goes RED. The controls prove from the pipeline FILE —
|
||||
// never by executing Woodpecker — that a verify step that FAILS (nonzero
|
||||
// exit) blocks every publish effect.
|
||||
//
|
||||
// Woodpecker semantics these controls rely on:
|
||||
// - A step that exits nonzero FAILS, and every step that transitively
|
||||
// depends on a failed step is SKIPPED — never run. That skip is the only
|
||||
// thing standing between a failed mandatory check and a publish effect.
|
||||
// - `detach: true` removes the step from the wait graph: the pipeline does
|
||||
// not wait for detached steps, so their failure can never block anything.
|
||||
// - `failure: ignore` reports a failed step as success to the DAG.
|
||||
// - `success: [codes...]` overrides which exit codes count as success;
|
||||
// admitting any nonzero code launders a failed verification into green.
|
||||
// - `when` on the verify step would skip verification entirely on some
|
||||
// event/path classes while publish effects still run.
|
||||
//
|
||||
// Bypass shapes covered (one negative-control test each):
|
||||
// S1 Missing edge — a publish effect whose dependency closure does not
|
||||
// contain `verify` (a refactor drops the depends_on entry).
|
||||
// S2 Hidden effect — a step whose NAME does not start with `publish` but
|
||||
// whose COMMANDS publish npm packages or push images. Effects are
|
||||
// classified by commands, so renaming a step cannot un-gate it.
|
||||
// S3 Detached verify — `verify: { detach: true }`: publish steps no longer
|
||||
// wait for verify, so the depends_on edge is decorative.
|
||||
// S4 Always-pass verify — `failure: ignore`, or a `success` override
|
||||
// admitting nonzero exit codes: verify fails, the DAG sees success.
|
||||
// S5 Conditional verify — a `when`/path filter on verify itself.
|
||||
// S6 Exact-commit drift — a HEAD-moving step (git checkout/switch/reset/
|
||||
// clean/pull/clone/fetch) ordered between `verify` and a publish
|
||||
// effect: the verified commit would not be the published commit. A
|
||||
// LEGITIMATE re-checkout is allowed only when `verify` itself runs
|
||||
// after it — positive control included.
|
||||
// S7 Gate removal — the verify step deleted or renamed away entirely.
|
||||
|
||||
// Reuse the monorepo's existing YAML parser (@mosaicstack/mosaic's direct
|
||||
// dependency) instead of adding a root dependency or vendoring a parser.
|
||||
const mosaicRequire = createRequire(
|
||||
path.resolve(process.cwd(), 'packages', 'mosaic', 'package.json'),
|
||||
);
|
||||
const { parse: parseYaml } = mosaicRequire('yaml');
|
||||
|
||||
const publishYmlPath = path.join(process.cwd(), '.woodpecker', 'publish.yml');
|
||||
|
||||
async function readPublishPipeline() {
|
||||
return parseYaml(await readFile(publishYmlPath, 'utf8'));
|
||||
}
|
||||
|
||||
// A command has a publish EFFECT when it publishes npm packages (`publish`
|
||||
// anywhere after a package-manager token — `pnpm --filter "@x/*" publish`
|
||||
// puts flags and quoted filters between the binary and the subcommand) or
|
||||
// pushes an image (kaniko, docker push, or a registry --destination).
|
||||
// Deliberately over-broad: a false positive forces justification, a false
|
||||
// negative is the actual hazard.
|
||||
function isPublishCommand(command) {
|
||||
return (
|
||||
/(^|\s)\/kaniko\/executor\b/.test(command) ||
|
||||
/(^|\s)docker\s+push\b/.test(command) ||
|
||||
/(^|\s)--destination(\s|=)/.test(command) ||
|
||||
(/\bpublish\b/.test(command) && /(^|\s)(npm|pnpm|yarn)(\s|$)/.test(command))
|
||||
);
|
||||
}
|
||||
|
||||
function hasPublishEffect(step) {
|
||||
return (step.commands ?? []).some(isPublishCommand);
|
||||
}
|
||||
|
||||
// A step is a publish effect when its name says so OR (S2) when any of its
|
||||
// commands does — classification must not depend on the name alone.
|
||||
function publishEffectSteps(pipeline) {
|
||||
return Object.entries(pipeline.steps ?? {})
|
||||
.filter(([name, step]) => name.startsWith('publish') || hasPublishEffect(step))
|
||||
.map(([name]) => name);
|
||||
}
|
||||
|
||||
// Transitive closure of a step's depends_on graph.
|
||||
function dependencyClosure(pipeline, stepName, seen = new Set()) {
|
||||
const dependencies = pipeline.steps?.[stepName]?.depends_on ?? [];
|
||||
for (const dependency of dependencies) {
|
||||
if (seen.has(dependency)) continue;
|
||||
seen.add(dependency);
|
||||
dependencyClosure(pipeline, dependency, seen);
|
||||
}
|
||||
return seen;
|
||||
}
|
||||
|
||||
// Deliberately over-broad: `git fetch` alone does not move HEAD, but the
|
||||
// classic re-checkout pair is `git fetch && git reset --hard <remote>`; a
|
||||
// fetch step sitting between verify and a publish effect deserves scrutiny,
|
||||
// so the gate fails closed on it.
|
||||
function movesHead(step) {
|
||||
return (step.commands ?? []).some((command) =>
|
||||
/(^|\s)git\s+(checkout|switch|reset|clean|pull|clone|fetch)\b/.test(command),
|
||||
);
|
||||
}
|
||||
|
||||
// The structural gate checker: green only when a failed (nonzero-exit)
|
||||
// verify provably blocks every publish effect on the same commit.
|
||||
function assertPublishGateBlocksOnVerify(pipeline) {
|
||||
assert.ok(pipeline.steps, 'publish pipeline must define steps');
|
||||
const verify = pipeline.steps.verify;
|
||||
assert.ok(verify, 'publish pipeline must define a `verify` step (S7)');
|
||||
|
||||
// S5: a skipped verification authorizes publishes exactly as much as a
|
||||
// failed one — verify must be unconditional.
|
||||
assert.equal(verify.when, undefined, '`verify` must not carry a when/path filter (S5)');
|
||||
|
||||
// S3/S4: the depends_on edges are only meaningful if verify's own failure
|
||||
// is both awaited and terminal for the DAG.
|
||||
assert.equal(verify.detach, undefined, '`verify` must not be detached (S3)');
|
||||
assert.equal(
|
||||
verify.failure,
|
||||
undefined,
|
||||
'`verify` must not tolerate its own failure (S4: failure: ignore launders a failed gate into success)',
|
||||
);
|
||||
assert.equal(
|
||||
verify.success,
|
||||
undefined,
|
||||
'`verify` must not override success exit codes (S4: nonzero codes would make failed verification pass)',
|
||||
);
|
||||
|
||||
const effects = publishEffectSteps(pipeline);
|
||||
assert.ok(effects.length > 0, 'publish pipeline must contain publish effect steps to guard');
|
||||
|
||||
const verifyClosure = dependencyClosure(pipeline, 'verify');
|
||||
for (const stepName of effects) {
|
||||
// S1: only the failure-skip semantics of the DAG stand between a failed
|
||||
// verify and this effect — the verify edge in its closure is the proof.
|
||||
const closure = dependencyClosure(pipeline, stepName);
|
||||
assert.ok(
|
||||
closure.has('verify'),
|
||||
`publish effect '${stepName}' must transitively depend on verify (S1) — a failed verify must skip it`,
|
||||
);
|
||||
|
||||
// S6: any step ordered after verify (outside its closure) but inside the
|
||||
// effect's chain must not be able to move HEAD. If the pipeline
|
||||
// legitimately re-checks-out, verify must run after the re-checkout.
|
||||
for (const chainStep of closure) {
|
||||
if (chainStep === 'verify' || verifyClosure.has(chainStep)) continue;
|
||||
assert.ok(
|
||||
!movesHead(pipeline.steps[chainStep]),
|
||||
`step '${chainStep}' sits between verify and publish effect '${stepName}' and can move HEAD (S6)` +
|
||||
' — verify must re-run after any re-checkout',
|
||||
);
|
||||
}
|
||||
}
|
||||
return effects;
|
||||
}
|
||||
|
||||
// A minimal but healthy gate used as the base for every negative-control
|
||||
// mutation: verify (identity + canonical command) → build → publish-npm,
|
||||
// with the publish effect blocked by verify both directly and through build.
|
||||
const HEALTHY_GATE_YAML = `
|
||||
steps:
|
||||
verify:
|
||||
image: node:24-alpine
|
||||
commands:
|
||||
- |
|
||||
if [ -z "$CI_COMMIT_SHA" ] || [ "$CI_COMMIT_SHA" != "$(git rev-parse HEAD)" ]; then
|
||||
echo "identity mismatch" >&2
|
||||
exit 1
|
||||
fi
|
||||
- pnpm verify:release
|
||||
build:
|
||||
image: node:24-alpine
|
||||
commands:
|
||||
- pnpm build
|
||||
depends_on:
|
||||
- verify
|
||||
publish-npm:
|
||||
image: node:24-alpine
|
||||
commands:
|
||||
- npm publish
|
||||
depends_on:
|
||||
- build
|
||||
- verify
|
||||
`;
|
||||
|
||||
// Fresh parse per call so every negative control mutates its own object.
|
||||
function healthyPipeline() {
|
||||
return parseYaml(HEALTHY_GATE_YAML);
|
||||
}
|
||||
|
||||
test('the real publish pipeline: a failed verify provably blocks every publish effect', async () => {
|
||||
const pipeline = await readPublishPipeline();
|
||||
const effects = assertPublishGateBlocksOnVerify(pipeline);
|
||||
assert.deepEqual(effects.sort(), [
|
||||
'build-appservice',
|
||||
'build-gateway',
|
||||
'build-web',
|
||||
'publish-next-npm',
|
||||
'publish-npm',
|
||||
]);
|
||||
});
|
||||
|
||||
test('fixture sanity: the healthy gate base passes the checker unmutated', () => {
|
||||
assertPublishGateBlocksOnVerify(healthyPipeline());
|
||||
});
|
||||
|
||||
test('S1 negative control: a publish effect with no verify edge fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps['publish-npm'].depends_on = ['build'];
|
||||
pipeline.steps.build.depends_on = [];
|
||||
assert.throws(
|
||||
() => assertPublishGateBlocksOnVerify(pipeline),
|
||||
/publish-npm.*must transitively depend on verify/s,
|
||||
);
|
||||
});
|
||||
|
||||
test('S2 negative control: an npm publish hidden behind a non-publish step name fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
delete pipeline.steps['publish-npm'];
|
||||
pipeline.steps.build.depends_on = [];
|
||||
pipeline.steps.deploy = {
|
||||
image: 'node:24-alpine',
|
||||
commands: ['npm publish'],
|
||||
depends_on: ['build'],
|
||||
};
|
||||
// Detection must be by COMMAND: the name says "deploy", the commands say
|
||||
// publish — an un-gated effect under either reading.
|
||||
assert.throws(
|
||||
() => assertPublishGateBlocksOnVerify(pipeline),
|
||||
/deploy.*must transitively depend on verify/s,
|
||||
);
|
||||
});
|
||||
|
||||
test('S2 negative control: a kaniko image push under a build-* name fails the checker when ungated', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
delete pipeline.steps['publish-npm'];
|
||||
pipeline.steps.build.depends_on = [];
|
||||
pipeline.steps['push-platform-image'] = {
|
||||
image: 'gcr.io/kaniko-project/executor:debug',
|
||||
commands: ['/kaniko/executor --context . --destination reg.example/img:latest'],
|
||||
depends_on: ['build'],
|
||||
};
|
||||
assert.throws(
|
||||
() => assertPublishGateBlocksOnVerify(pipeline),
|
||||
/push-platform-image.*must transitively depend on verify/s,
|
||||
);
|
||||
});
|
||||
|
||||
test('S3 negative control: a detached verify fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.verify.detach = true;
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /detached \(S3\)/);
|
||||
});
|
||||
|
||||
test('S4 negative control: failure: ignore on verify fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.verify.failure = 'ignore';
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /tolerate its own failure/);
|
||||
});
|
||||
|
||||
test('S4 negative control: a success override admitting nonzero exit codes fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.verify.success = [0, 1];
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /success exit codes/);
|
||||
});
|
||||
|
||||
test('S5 negative control: a when filter on verify fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.verify.when = [{ event: 'push' }];
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /when\/path filter \(S5\)/);
|
||||
});
|
||||
|
||||
test('S6 negative control: a HEAD-moving step between verify and publish fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.resync = {
|
||||
image: 'node:24-alpine',
|
||||
commands: ['git fetch origin', 'git reset --hard origin/main'],
|
||||
depends_on: [],
|
||||
};
|
||||
pipeline.steps.build.depends_on = ['verify', 'resync'];
|
||||
// resync sits AFTER verify in the publish chain (verify does not depend on
|
||||
// it), so the verified commit could be replaced before publishing.
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /resync.*can move HEAD/s);
|
||||
});
|
||||
|
||||
test('S6 positive control: a legitimate re-checkout passes when verify re-runs after it', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
pipeline.steps.resync = {
|
||||
image: 'node:24-alpine',
|
||||
commands: ['git fetch origin', 'git reset --hard origin/main'],
|
||||
depends_on: [],
|
||||
};
|
||||
pipeline.steps.verify.depends_on = ['resync'];
|
||||
pipeline.steps.build.depends_on = ['verify'];
|
||||
// resync precedes verify in the chain, so verification covers the
|
||||
// re-checked-out HEAD — the exact-commit contract holds.
|
||||
assertPublishGateBlocksOnVerify(pipeline);
|
||||
});
|
||||
|
||||
test('S7 negative control: deleting the verify step entirely fails the checker', () => {
|
||||
const pipeline = healthyPipeline();
|
||||
delete pipeline.steps.verify;
|
||||
pipeline.steps['publish-npm'].depends_on = ['build'];
|
||||
assert.throws(() => assertPublishGateBlocksOnVerify(pipeline), /`verify` step/);
|
||||
});
|
||||
@@ -9,7 +9,10 @@ import { STAGES } from './verify-release.mjs';
|
||||
// SDLC-D-034 checkout invariant: publication in .woodpecker/publish.yml is
|
||||
// bound to exact-commit terminal verification. This suite parses the real
|
||||
// pipeline files and fails red when the gate is bypassed, weakened, or drifts
|
||||
// out of sync with the canonical `pnpm verify:release` command.
|
||||
// out of sync with the canonical `pnpm verify:release` command. The negative
|
||||
// controls for pipeline DAG/bypass shapes live in
|
||||
// scripts/publish-gate-structure.test.mjs (RI-1-002); this file owns the
|
||||
// canonical-command composition controls.
|
||||
|
||||
// Reuse the monorepo's existing YAML parser (@mosaicstack/mosaic's direct
|
||||
// dependency) instead of adding a root dependency or vendoring a parser.
|
||||
@@ -219,13 +222,16 @@ steps:
|
||||
assert.throws(() => assertPublishGate(parseYaml(noIdentityPipeline)), /CI_COMMIT_SHA/);
|
||||
});
|
||||
|
||||
test('the canonical verify:release stages mirror the PR CI pipeline one-for-one', async () => {
|
||||
const ci = parseYaml(await readFile(ciYmlPath, 'utf8'));
|
||||
const canonical = Object.fromEntries(STAGES.map((stage) => [stage.name, stage.commands]));
|
||||
// The composition check: the canonical stage table must mirror the PR CI
|
||||
// pipeline's complete mandatory set. Parameterized by the stage list so the
|
||||
// subset negative control below can prove a dropped stage goes red (RI-1-002:
|
||||
// the canonical command cannot silently lose a check).
|
||||
function assertStagesMirrorCi(stages, ci) {
|
||||
const canonical = Object.fromEntries(stages.map((stage) => [stage.name, stage.commands]));
|
||||
|
||||
// The complete mandatory set, in gate order.
|
||||
assert.deepEqual(
|
||||
STAGES.map((stage) => stage.name),
|
||||
stages.map((stage) => stage.name),
|
||||
['sanitization', 'upgrade-guard', 'typecheck', 'lint', 'format', 'test', 'build'],
|
||||
);
|
||||
|
||||
@@ -269,6 +275,26 @@ test('the canonical verify:release stages mirror the PR CI pipeline one-for-one'
|
||||
`ci.yml test step must keep its pipeline-level prerequisite '${fragment}'`,
|
||||
);
|
||||
}
|
||||
}
|
||||
|
||||
test('the canonical verify:release stages mirror the PR CI pipeline one-for-one', async () => {
|
||||
const ci = parseYaml(await readFile(ciYmlPath, 'utf8'));
|
||||
assertStagesMirrorCi(STAGES, ci);
|
||||
});
|
||||
|
||||
test('a subset stage list fails the composition check — a dropped stage cannot pass silently', async () => {
|
||||
const ci = parseYaml(await readFile(ciYmlPath, 'utf8'));
|
||||
// Drop each stage one at a time: every stage is load-bearing, so every drop
|
||||
// must go red. If any drop went green, a refactor could silently delete a
|
||||
// mandatory check from the canonical command.
|
||||
for (const stage of STAGES) {
|
||||
const subset = STAGES.filter((entry) => entry.name !== stage.name);
|
||||
assert.throws(
|
||||
() => assertStagesMirrorCi(subset, ci),
|
||||
Error,
|
||||
`composition check must fail when the '${stage.name}' stage is dropped from the table`,
|
||||
);
|
||||
}
|
||||
});
|
||||
|
||||
test('the root package.json exposes verify:release as the canonical command', async () => {
|
||||
|
||||
@@ -133,13 +133,10 @@ When the full `@mosaicstack/forge` package is available, Forge uses MACP task ex
|
||||
|
||||
```bash
|
||||
# Run from CLI
|
||||
# Fails closed with a typed FORGE_NO_EXECUTOR capability error when no real
|
||||
# executor is wired — pass --simulate to opt into explicit typed simulation
|
||||
# (every result carries status `simulated`, which satisfies nothing).
|
||||
mosaic forge run path/to/brief.md [--simulate]
|
||||
mosaic forge run path/to/brief.md
|
||||
|
||||
# Resume interrupted run (same fail-closed rule as forge run)
|
||||
mosaic forge resume .forge/runs/20260401-143022/ [--simulate]
|
||||
# Resume interrupted run
|
||||
mosaic forge resume .forge/runs/20260401-143022/
|
||||
|
||||
# Check status
|
||||
mosaic forge status .forge/runs/20260401-143022/
|
||||
|
||||
Reference in New Issue
Block a user