@tea-agent/loop-agent 0.13.0-beta.0 → 0.13.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (282) hide show
  1. package/AGENTS.md +157 -155
  2. package/CHANGELOG.md +301 -322
  3. package/README.md +335 -345
  4. package/bin/agent-worker.js +22 -22
  5. package/bin/loop-agent.js +21 -21
  6. package/dist/commands/cursor-prompt.js +6 -6
  7. package/dist/commands/init.js +597 -528
  8. package/dist/commands/loop-benchmark.js +11 -11
  9. package/dist/commands/pi-reuse-benchmark.js +16 -16
  10. package/dist/executors/shell-executor.js +200 -21
  11. package/dist/infrastructure/evaluation/candidate-store.js +5 -1
  12. package/dist/sidecars/cursor-prompt/executor.js +1 -1
  13. package/dist/task/runtime.js +27 -27
  14. package/dist/worker/observe/static/api.js +46 -46
  15. package/dist/worker/observe/static/app.js +150 -150
  16. package/dist/worker/observe/static/constants.js +148 -148
  17. package/dist/worker/observe/static/copy.js +67 -67
  18. package/dist/worker/observe/static/dag-helpers.js +172 -172
  19. package/dist/worker/observe/static/dag-layout.d.ts +31 -31
  20. package/dist/worker/observe/static/dag-layout.js +83 -83
  21. package/dist/worker/observe/static/dag-model.js +72 -72
  22. package/dist/worker/observe/static/dom.js +212 -53
  23. package/dist/worker/observe/static/format-pool.js +67 -67
  24. package/dist/worker/observe/static/format.js +292 -292
  25. package/dist/worker/observe/static/index.html +308 -308
  26. package/dist/worker/observe/static/kpi.js +94 -94
  27. package/dist/worker/observe/static/relations.js +133 -133
  28. package/dist/worker/observe/static/router.js +93 -93
  29. package/dist/worker/observe/static/run-processing.js +148 -148
  30. package/dist/worker/observe/static/shell-chrome.js +68 -68
  31. package/dist/worker/observe/static/state.js +267 -253
  32. package/dist/worker/observe/static/styles.css +1902 -1902
  33. package/dist/worker/observe/static/views/batch.js +227 -227
  34. package/dist/worker/observe/static/views/dag-graph.js +172 -172
  35. package/dist/worker/observe/static/views/dag-inspector.js +627 -607
  36. package/dist/worker/observe/static/views/dag.js +371 -362
  37. package/dist/worker/observe/static/views/dashboard.js +509 -252
  38. package/dist/worker/observe/static/views/failures.js +143 -143
  39. package/dist/worker/observe/static/views/feature.js +492 -492
  40. package/dist/worker/observe/static/views/pool.js +350 -350
  41. package/dist/worker/observe/static/views/run.js +453 -453
  42. package/dist/worker/observe/static/views/session-timeline.js +219 -205
  43. package/dist/worker/observe/static/views/shell.js +7 -7
  44. package/dist/worker/observe/static/views/task.js +314 -314
  45. package/dist/worker/observe/static/views/timeline.js +163 -163
  46. package/dist/workflows/dag/backend-test-case-manifest.js +503 -0
  47. package/dist/workflows/dag/backend-test-execution-contract.js +353 -0
  48. package/dist/workflows/dag/backend-test-result-contract.js +568 -0
  49. package/dist/workflows/dag/canvas-observer.js +275 -275
  50. package/dist/workflows/dag/decision-envelope.js +57 -2
  51. package/dist/workflows/dag/frontend-implementation-contract.js +240 -0
  52. package/dist/workflows/dag/frontend-project-capability.js +309 -0
  53. package/dist/workflows/dag/frontend-repair.js +341 -0
  54. package/dist/workflows/dag/frontend-risk.js +161 -0
  55. package/dist/workflows/dag/frontend-verification-trace.js +190 -0
  56. package/dist/workflows/dag/init-hybrid.js +1020 -125
  57. package/dist/workflows/dag/repair-artifact.js +43 -3
  58. package/dist/workflows/dag/skill-instructions.js +4 -2
  59. package/dist/workflows/dag/types.js +29 -8
  60. package/docs/README.md +105 -104
  61. package/docs/agent-dag-recovery-playbook.md +195 -195
  62. package/docs/agent-dag-runner.md +67 -67
  63. package/docs/architecture/README.md +26 -26
  64. package/docs/architecture/dag-execution.md +140 -140
  65. package/docs/architecture/evolution.md +54 -54
  66. package/docs/architecture/facts-and-state.md +71 -71
  67. package/docs/architecture/runtime-boundaries.md +191 -191
  68. package/docs/architecture/system-overview.md +93 -93
  69. package/docs/architecture/worker-and-feature.md +85 -85
  70. package/docs/cursor-prompt-sidecar.md +36 -36
  71. package/docs/decisions/README.md +18 -18
  72. package/docs/design/README.md +167 -167
  73. package/docs/development-principles.md +73 -73
  74. package/docs/exec-plans/README.md +6 -6
  75. package/docs/exec-plans/active/README.md +1 -4
  76. package/docs/exec-plans/completed/README.md +106 -84
  77. package/docs/feature-workflow.md +414 -389
  78. package/docs/harness-methodology-debugging.md +153 -153
  79. package/docs/harness-methodology-tdd.md +130 -130
  80. package/docs/harness-methodology-verification.md +27 -27
  81. package/docs/init-surface.manifest.json +307 -289
  82. package/docs/loop-agent-harness.md +142 -142
  83. package/docs/production-readiness.md +96 -96
  84. package/docs/progress/README.md +76 -60
  85. package/docs/reports/README.md +150 -108
  86. package/docs/skills/README.md +7 -7
  87. package/docs/skills/vetted-skill-registry.md +29 -29
  88. package/docs/templates/adr.md +60 -60
  89. package/docs/templates/agent-dag-authority-surface-audit.prompt.md +94 -94
  90. package/docs/templates/agent-dag-decision-envelope.schema.json +213 -213
  91. package/docs/templates/agent-dag-decision-gate-dogfood-report.md +117 -117
  92. package/docs/templates/agent-dag-decision-gate.prompt.md +246 -246
  93. package/docs/templates/agent-dag-process-supervisor.prompt.md +98 -98
  94. package/docs/templates/agent-dag-report.schema.json +473 -473
  95. package/docs/templates/agent-dag-review-verdict.prompt.md +68 -68
  96. package/docs/templates/agent-dag.base.json +190 -190
  97. package/docs/templates/agent-dag.final-verification.json +185 -185
  98. package/docs/templates/agent-dag.schema.json +411 -411
  99. package/docs/templates/agent-dag.supervised-implementation.json +620 -501
  100. package/docs/templates/backend-test-analysis.schema.json +44 -44
  101. package/docs/templates/backend-test-case-manifest.schema.json +190 -0
  102. package/docs/templates/backend-test-dag.classify.prompt.md +75 -0
  103. package/docs/templates/backend-test-dag.generate-pytest.prompt.md +204 -202
  104. package/docs/templates/backend-test-dag.json +559 -311
  105. package/docs/templates/backend-test-dag.retrospect.prompt.md +139 -125
  106. package/docs/templates/backend-test-dag.review-cases.prompt.md +83 -81
  107. package/docs/templates/backend-test-execution.schema.json +133 -0
  108. package/docs/templates/backend-test-result.schema.json +99 -0
  109. package/docs/templates/branch-merge-report.md +93 -0
  110. package/docs/templates/exec-plan.md +64 -64
  111. package/docs/templates/feature-spec.md +53 -53
  112. package/docs/templates/frontend-design-contract.md +42 -42
  113. package/docs/templates/frontend-eval/fixtures/failures/01-type-build-error.md +17 -0
  114. package/docs/templates/frontend-eval/fixtures/failures/02-unit-component-test-fail.md +16 -0
  115. package/docs/templates/frontend-eval/fixtures/failures/03-fixture-schema-drift.md +16 -0
  116. package/docs/templates/frontend-eval/fixtures/failures/04-missing-loading-empty-error-state.md +16 -0
  117. package/docs/templates/frontend-eval/fixtures/failures/05-forbidden-write-writeset-expansion.md +16 -0
  118. package/docs/templates/frontend-eval/fixtures/failures/06-unapproved-dependency-add.md +16 -0
  119. package/docs/templates/frontend-eval/fixtures/failures/07-mock-production-on.md +21 -0
  120. package/docs/templates/frontend-eval/fixtures/functional/01-simple-component-style.md +29 -0
  121. package/docs/templates/frontend-eval/fixtures/functional/02-form-validation.md +28 -0
  122. package/docs/templates/frontend-eval/fixtures/functional/03-list-detail-page.md +28 -0
  123. package/docs/templates/frontend-eval/fixtures/functional/04-api-mock.md +29 -0
  124. package/docs/templates/frontend-eval/fixtures/functional/05-permission-auth-gated-ui.md +27 -0
  125. package/docs/templates/frontend-eval/fixtures/functional/06-ssr-server-client-boundary.md +28 -0
  126. package/docs/templates/frontend-eval/fixtures/functional/07-shared-public-component-api.md +28 -0
  127. package/docs/templates/frontend-eval/fixtures/functional/08-pure-local-no-remote.md +27 -0
  128. package/docs/templates/frontend-eval/metrics.md +138 -0
  129. package/docs/templates/frontend-eval/smoke-targets.md +53 -0
  130. package/docs/templates/frontend-implementation-contract.schema.json +27 -0
  131. package/docs/templates/frontend-task-constraints.md +35 -35
  132. package/docs/templates/frontend-task-requirement.md +70 -70
  133. package/docs/templates/frontend-test-dag.generate-cases.prompt.md +5 -5
  134. package/docs/templates/frontend-test-dag.json +23 -23
  135. package/docs/templates/frontend-test-dag.retrieve-context.prompt.md +3 -3
  136. package/docs/templates/frontend-test-dag.retrospect.prompt.md +3 -3
  137. package/docs/templates/frontend-test-dag.review-cases.prompt.md +3 -3
  138. package/docs/templates/frontend-test-dag.review-execution.prompt.md +3 -3
  139. package/docs/templates/harness.schema.json +221 -221
  140. package/docs/templates/hybrid-dag.json +188 -188
  141. package/docs/templates/init-evolution-review.md +35 -35
  142. package/docs/templates/interactive-ui-round2-experiment.md +66 -66
  143. package/docs/templates/knowledge-graph-bootstrap-dag.json +118 -118
  144. package/docs/templates/knowledge-sync-dag.json +178 -178
  145. package/docs/templates/knowledge-sync-draft.schema.json +71 -71
  146. package/docs/templates/product-line/AGENTS.md +8 -8
  147. package/docs/templates/product-line/README.md +9 -9
  148. package/docs/templates/product-line/acceptance.yaml +14 -14
  149. package/docs/templates/product-line/closeout.yaml +9 -9
  150. package/docs/templates/product-line/design.md +13 -13
  151. package/docs/templates/product-line/links.md +10 -10
  152. package/docs/templates/product-line/requirement.md +17 -17
  153. package/docs/templates/product-line/task-graph.yaml +15 -15
  154. package/docs/templates/product-line/task.yaml +64 -64
  155. package/docs/templates/product-line/test-plan.md +7 -7
  156. package/docs/templates/production-readiness-checklist.md +57 -57
  157. package/docs/templates/progress-log.md +17 -17
  158. package/docs/templates/project-start-checklist.md +9 -9
  159. package/docs/templates/qa-report.md +48 -48
  160. package/docs/templates/sprint-contract.md +29 -29
  161. package/docs/templates/worker-dogfood-evidence.md +80 -80
  162. package/docs/templates/worker-dogfood-setup.md +68 -68
  163. package/docs/verification-matrix.md +70 -70
  164. package/examples/decision-gate-agent-dag.json +177 -177
  165. package/examples/example-dag.json +46 -46
  166. package/examples/hybrid-loop-agent-dag.json +189 -189
  167. package/harness.json +66 -66
  168. package/package.json +52 -88
  169. package/scripts/check-product-line-docs.sh +29 -29
  170. package/scripts/check-task-pool-root.sh +32 -32
  171. package/scripts/kb-bootstrap-init-skeleton.sh +240 -240
  172. package/scripts/kb-graph-incremental-prepare.mjs +386 -386
  173. package/scripts/kb-graph-incremental-prepare.sh +5 -5
  174. package/scripts/kb-graph-materialize.mjs +105 -105
  175. package/scripts/kb-graph-materialize.sh +4 -4
  176. package/scripts/kb-graph-promote.mjs +164 -164
  177. package/scripts/kb-graph-promote.sh +4 -4
  178. package/scripts/kb-query.mjs +554 -554
  179. package/scripts/kb-query.sh +5 -5
  180. package/skills/agent-worker/SKILL.md +39 -39
  181. package/skills/agent-worker/references/agent-worker-operator.md +60 -60
  182. package/skills/ai-engineering-context/SKILL.md +48 -48
  183. package/skills/analyze-product-dependencies/SKILL.md +67 -67
  184. package/skills/analyze-product-dependencies/agents/openai.yaml +4 -4
  185. package/skills/analyze-product-dependencies/references/api-documentation-schema.md +30 -30
  186. package/skills/analyze-product-dependencies/references/dependency-analysis-schema.md +28 -28
  187. package/skills/analyze-product-dependencies/references/example.md +76 -76
  188. package/skills/analyze-product-dependencies/references/forward-test-cases.md +35 -35
  189. package/skills/analyze-product-dependencies/references/input-contract.md +11 -11
  190. package/skills/analyze-product-dependencies/references/scouting-rules.md +61 -61
  191. package/skills/analyze-product-dependencies/scripts/test-validators.mjs +267 -267
  192. package/skills/analyze-product-dependencies/scripts/validate-api-documentation.mjs +101 -101
  193. package/skills/analyze-product-dependencies/scripts/validate-dependency-analysis.mjs +142 -142
  194. package/skills/analyze-product-dependencies/scripts/validate-product-requirement-input.mjs +76 -76
  195. package/skills/analyze-product-dependencies/scripts/validation-helpers.mjs +146 -146
  196. package/skills/analyze-product-requirements/SKILL.md +90 -90
  197. package/skills/analyze-product-requirements/agents/openai.yaml +4 -4
  198. package/skills/analyze-product-requirements/references/acceptance-criteria.md +91 -91
  199. package/skills/analyze-product-requirements/references/clarification-and-knowledge.md +56 -56
  200. package/skills/analyze-product-requirements/references/example.md +86 -86
  201. package/skills/analyze-product-requirements/references/forward-test-cases.md +66 -66
  202. package/skills/analyze-product-requirements/references/product-analysis-schema.md +32 -32
  203. package/skills/analyze-product-requirements/references/product-requirement-schema.md +33 -33
  204. package/skills/analyze-product-requirements/references/requirement-clarification-schema.md +35 -35
  205. package/skills/analyze-product-requirements/scripts/test-validators.mjs +193 -193
  206. package/skills/analyze-product-requirements/scripts/validate-product-analysis.mjs +69 -69
  207. package/skills/analyze-product-requirements/scripts/validate-product-requirement.mjs +97 -97
  208. package/skills/analyze-product-requirements/scripts/validate-requirement-clarification.mjs +98 -98
  209. package/skills/analyze-product-requirements/scripts/validation-helpers.mjs +156 -156
  210. package/skills/browser-tools/SKILL.md +196 -0
  211. package/skills/browser-tools/browser-content.js +103 -0
  212. package/skills/browser-tools/browser-cookies.js +35 -0
  213. package/skills/browser-tools/browser-eval.js +53 -0
  214. package/skills/browser-tools/browser-hn-scraper.js +108 -0
  215. package/skills/browser-tools/browser-nav.js +44 -0
  216. package/skills/browser-tools/browser-pick.js +162 -0
  217. package/skills/browser-tools/browser-screenshot.js +34 -0
  218. package/skills/browser-tools/browser-start.js +86 -0
  219. package/skills/browser-tools/package-lock.json +2556 -0
  220. package/skills/browser-tools/package.json +19 -0
  221. package/skills/code-review-core/SKILL.md +20 -20
  222. package/skills/codebase-scout/SKILL.md +19 -19
  223. package/skills/frontend-design-review/SKILL.md +66 -66
  224. package/skills/frontend-design-review/references/review-checklist.md +58 -58
  225. package/skills/frontend-implementation/SKILL.md +49 -47
  226. package/skills/frontend-implementation/references/code-standards.md +32 -32
  227. package/skills/frontend-implementation/references/design-spec.md +46 -46
  228. package/skills/frontend-implementation/references/node-contracts.md +27 -76
  229. package/skills/frontend-review/SKILL.md +59 -59
  230. package/skills/frontend-review/references/review-findings.md +47 -47
  231. package/skills/frontend-verification/SKILL.md +53 -53
  232. package/skills/frontend-verification/references/verification-checklist.md +68 -68
  233. package/skills/grill-me/SKILL.md +10 -10
  234. package/skills/grill-with-docs/SKILL.md +88 -88
  235. package/skills/grill-with-docs/adr-format.md +47 -47
  236. package/skills/grill-with-docs/context-format.md +60 -60
  237. package/skills/init-capability-evolution/SKILL.md +70 -70
  238. package/skills/loop-agent/SKILL.md +151 -151
  239. package/skills/loop-agent/references/README.md +67 -67
  240. package/skills/loop-agent/references/command-reference.md +527 -505
  241. package/skills/loop-agent/references/docs-converge.md +126 -126
  242. package/skills/loop-agent/references/harness-policy.md +263 -263
  243. package/skills/loop-agent/references/hybrid-dag.md +243 -238
  244. package/skills/loop-agent/references/learned/README.md +21 -21
  245. package/skills/loop-agent/references/long-running-loop.md +57 -57
  246. package/skills/loop-agent/references/model-routing.md +36 -36
  247. package/skills/loop-agent/references/multi-worktree.md +54 -54
  248. package/skills/loop-agent/references/one-shot-runs.md +85 -85
  249. package/skills/loop-agent/references/orchestrator-and-interventions.md +169 -169
  250. package/skills/loop-agent/references/pi-prompt.md +23 -23
  251. package/skills/loop-agent/references/pi-subagent-assisted-mode.md +84 -84
  252. package/skills/loop-agent/references/post-implementation-and-patterns.md +44 -44
  253. package/skills/loop-agent/references/task-workflow.md +89 -89
  254. package/skills/loop-agent/references/verification-and-failure-handling.md +141 -139
  255. package/skills/playwright-cli/SKILL.md +420 -420
  256. package/skills/playwright-cli/references/element-attributes.md +23 -23
  257. package/skills/playwright-cli/references/playwright-tests.md +39 -39
  258. package/skills/playwright-cli/references/request-mocking.md +87 -87
  259. package/skills/playwright-cli/references/running-code.md +241 -241
  260. package/skills/playwright-cli/references/session-management.md +225 -225
  261. package/skills/playwright-cli/references/storage-state.md +275 -275
  262. package/skills/playwright-cli/references/test-generation.md +433 -433
  263. package/skills/playwright-cli/references/tracing.md +139 -139
  264. package/skills/playwright-cli/references/video-recording.md +143 -143
  265. package/skills/playwright-cli-case-generator/SKILL.md +74 -74
  266. package/skills/requesting-code-review/SKILL.md +101 -101
  267. package/skills/requesting-code-review/code-reviewer.md +168 -168
  268. package/skills/systematic-debugging/CREATION-LOG.md +119 -119
  269. package/skills/systematic-debugging/SKILL.md +296 -296
  270. package/skills/systematic-debugging/condition-based-waiting-example.ts +158 -158
  271. package/skills/systematic-debugging/condition-based-waiting.md +115 -115
  272. package/skills/systematic-debugging/defense-in-depth.md +122 -122
  273. package/skills/systematic-debugging/find-polluter.sh +63 -63
  274. package/skills/systematic-debugging/root-cause-tracing.md +169 -169
  275. package/skills/systematic-debugging/test-academic.md +14 -14
  276. package/skills/systematic-debugging/test-pressure-1.md +58 -58
  277. package/skills/systematic-debugging/test-pressure-2.md +68 -68
  278. package/skills/systematic-debugging/test-pressure-3.md +69 -69
  279. package/skills/test-driven-development/SKILL.md +20 -20
  280. package/skills/using-git-worktrees/SKILL.md +215 -215
  281. package/skills/verification-before-completion/SKILL.md +154 -154
  282. package/skills/webapp-testing/SKILL.md +19 -19
@@ -1,12 +1,13 @@
1
1
  import { createHash } from "node:crypto";
2
- import { access, readdir, readFile, realpath, writeFile } from "node:fs/promises";
2
+ import { access, readdir, readFile, realpath, writeFile, } from "node:fs/promises";
3
3
  import os from "node:os";
4
4
  import path from "node:path";
5
5
  import { assertValidDagSpec } from "./validate.js";
6
6
  import { DAG_AGENT_RUNTIME_PI_ONLY, DAG_REPAIR_WRITER_PROTOCOL_EXPLICIT_NODE_V1, DAG_RUNTIME_CONTRACT_SCHEMA_VERSION, DEFAULT_DAG_OUTPUT_LANGUAGE, DEFAULT_DAG_EXECUTOR_MODELS, parseDagSpec, } from "./types.js";
7
7
  import { pathMatchesPattern } from "../../shared/git-progress.js";
8
8
  import { BASELINE_FORBIDDEN_PATHS } from "./governance-constants.js";
9
- import { DEFAULT_READ_ONLY_PI_RETRY_POLICY, isSafeReadOnlyPiRetryCandidate } from "./retry-policy.js";
9
+ import { buildDecisionEnvelopePromptContract } from "./decision-envelope.js";
10
+ import { DEFAULT_READ_ONLY_PI_RETRY_POLICY, isSafeReadOnlyPiRetryCandidate, } from "./retry-policy.js";
10
11
  import { resolveAdapter } from "../../adapters/index.js";
11
12
  import { loadHarnessManifest } from "../../governance/harness.js";
12
13
  import { buildAuthoritySurfaceAuditNode, buildAuthoritySurfaceGateNode, resolveAuthoritySurfaceAudit, } from "./authority-surface.js";
@@ -16,6 +17,10 @@ import { materializeTaskReferenceDocs } from "../../task/source-references.js";
16
17
  import { resolveVerifyPreset } from "../../executors/shell-verification.js";
17
18
  import { resolveExecutorModelMatrices } from "../../executors/model-routing.js";
18
19
  import { normalizeTaskRequirementText, resolveTaskDagTemplateSelection, } from "./task-demand-routing.js";
20
+ import { BACKEND_TEST_EXECUTION_DEFAULT_TEST_ROOT, buildBackendTestExecutionPreflightShellSnippet, } from "./backend-test-execution-contract.js";
21
+ import { buildBackendTestOutcomeGateShellSnippet } from "./backend-test-result-contract.js";
22
+ import { classifyFrontendRisk, } from "./frontend-risk.js";
23
+ import { discoverFrontendProjectCapability, } from "./frontend-project-capability.js";
19
24
  const REQUIREMENT_FILE = "需求.md";
20
25
  const CONSTRAINT_FILE = "执行约束.md";
21
26
  const REFERENCE_DIRECTORY = "references";
@@ -185,7 +190,10 @@ async function hasDirectDependency(repoRoot, depName) {
185
190
  try {
186
191
  const raw = await readFile(path.join(repoRoot, "package.json"), "utf-8");
187
192
  const pkg = JSON.parse(raw);
188
- const deps = { ...(pkg.dependencies ?? {}), ...(pkg.devDependencies ?? {}) };
193
+ const deps = {
194
+ ...(pkg.dependencies ?? {}),
195
+ ...(pkg.devDependencies ?? {}),
196
+ };
189
197
  return depName in deps;
190
198
  }
191
199
  catch {
@@ -207,7 +215,10 @@ async function discoverMockHandlerFiles(repoRoot, serviceRoot) {
207
215
  "mocks/browser.ts",
208
216
  "mocks/browser.js",
209
217
  ].map((candidatePath) => ({ framework: "msw", path: candidatePath })),
210
- ...["db.json", "mock/db.json", "src/mock/db.json"].map((candidatePath) => ({ framework: "json-server", path: candidatePath })),
218
+ ...["db.json", "mock/db.json", "src/mock/db.json"].map((candidatePath) => ({
219
+ framework: "json-server",
220
+ path: candidatePath,
221
+ })),
211
222
  ];
212
223
  const directoryCandidates = [
213
224
  ...(serviceRoot ? [{ path: serviceRoot }] : []),
@@ -331,7 +342,13 @@ export async function discoverFrontendMockCapability(repoRoot, taskConfig) {
331
342
  // A script alone is not strong evidence unless we also find config/entry
332
343
  }
333
344
  // 2. Check for direct mock framework dependencies
334
- const mockDeps = ["msw", "mockjs", "miragejs", "json-server", "vite-plugin-mock"];
345
+ const mockDeps = [
346
+ "msw",
347
+ "mockjs",
348
+ "miragejs",
349
+ "json-server",
350
+ "vite-plugin-mock",
351
+ ];
335
352
  const foundDeps = [];
336
353
  for (const dep of mockDeps) {
337
354
  if (await hasDirectDependency(repoRoot, dep)) {
@@ -385,7 +402,9 @@ export async function discoverFrontendMockCapability(repoRoot, taskConfig) {
385
402
  if (foundDeps.length > 0 && handlerResult.paths.length === 0) {
386
403
  reasons.push(`Mock dependency found (${foundDeps.join(", ")}) but no handler/bootstrap files detected`);
387
404
  }
388
- if (scriptResult.hasScript && handlerResult.paths.length === 0 && foundDeps.length === 0) {
405
+ if (scriptResult.hasScript &&
406
+ handlerResult.paths.length === 0 &&
407
+ foundDeps.length === 0) {
389
408
  reasons.push("Mock scripts exist but no handler files or direct mock dependencies found");
390
409
  }
391
410
  }
@@ -396,8 +415,11 @@ export async function discoverFrontendMockCapability(repoRoot, taskConfig) {
396
415
  cwd: repoRoot,
397
416
  timeoutMs: cmd.timeoutMs,
398
417
  }));
399
- const verifyCommands = [...configuredVerifyCommands, ...scriptResult.verifyCommands]
400
- .filter((command, index, commands) => commands.findIndex((candidate) => candidate.label === command.label) === index);
418
+ const verifyCommands = [
419
+ ...configuredVerifyCommands,
420
+ ...scriptResult.verifyCommands,
421
+ ].filter((command, index, commands) => commands.findIndex((candidate) => candidate.label === command.label) ===
422
+ index);
401
423
  if (strongEvidenceCount > 0) {
402
424
  return {
403
425
  status: "present",
@@ -475,7 +497,10 @@ async function frontendMockServiceRootSafetyViolation(repoRoot, taskConfig) {
475
497
  if (!frontendMockServiceRootAllowed({
476
498
  ...taskConfig,
477
499
  frontendMock: {
478
- ...(taskConfig.frontendMock ?? { policy: "auto", verifyCommands: [] }),
500
+ ...(taskConfig.frontendMock ?? {
501
+ policy: "auto",
502
+ verifyCommands: [],
503
+ }),
479
504
  serviceRoot: normalizedRealPath,
480
505
  },
481
506
  })) {
@@ -504,8 +529,7 @@ export function hasApiDependency(sources) {
504
529
  if (sources.taskConfig.frontendMock?.policy === "required") {
505
530
  return true;
506
531
  }
507
- const requirement = normalizeTaskRequirementText(sources.requirementMarkdown)
508
- .replace(/`[^`\n]*`/g, " ");
532
+ const requirement = normalizeTaskRequirementText(sources.requirementMarkdown).replace(/`[^`\n]*`/g, " ");
509
533
  const dependencyPatterns = [
510
534
  /(?:接口文档|接口定义|接口协议|后端接口|服务端接口|接口联调|请求|响应|远程数据|异步数据|数据获取|模拟接口|模拟数据)/,
511
535
  /(?<![A-Za-z0-9_])API(?![A-Za-z0-9_])/i,
@@ -539,7 +563,8 @@ export function hasApiDependency(sources) {
539
563
  export function resolveFrontendMockMode(capability, taskConfig, hasApiDep) {
540
564
  const policy = taskConfig.frontendMock?.policy ?? "auto";
541
565
  const hasDeterministicMockVerification = capability.verifyCommands.length > 0;
542
- if (capability.safetyViolation || !frontendMockServiceRootAllowed(taskConfig)) {
566
+ if (capability.safetyViolation ||
567
+ !frontendMockServiceRootAllowed(taskConfig)) {
543
568
  return "blocked";
544
569
  }
545
570
  // disabled policy: must respect project mock rules (can't override spec)
@@ -615,7 +640,8 @@ function isMetadataLine(line) {
615
640
  }
616
641
  if (/^(权威来源|SHA-256|冲突时以)/.test(trimmed))
617
642
  return true;
618
- if (/^>/.test(trimmed) && /(权威来源|SHA-256|原始 PRD|reference)/i.test(trimmed)) {
643
+ if (/^>/.test(trimmed) &&
644
+ /(权威来源|SHA-256|原始 PRD|reference)/i.test(trimmed)) {
619
645
  return true;
620
646
  }
621
647
  return false;
@@ -702,7 +728,9 @@ function shellQuote(value) {
702
728
  }
703
729
  function verifyCommandToShell(repoRoot, command) {
704
730
  const relativeCwd = path.relative(repoRoot, command.cwd);
705
- const cwdPrefix = relativeCwd && !relativeCwd.startsWith("..") && !path.isAbsolute(relativeCwd)
731
+ const cwdPrefix = relativeCwd &&
732
+ !relativeCwd.startsWith("..") &&
733
+ !path.isAbsolute(relativeCwd)
706
734
  ? `cd ${shellQuote(relativeCwd)} && `
707
735
  : command.cwd !== repoRoot
708
736
  ? `cd ${shellQuote(command.cwd)} && `
@@ -851,9 +879,7 @@ function deriveFrontendBehaviorPaths(taskConfig) {
851
879
  return taskConfig.allowedPaths;
852
880
  }
853
881
  function toTaskRelativeSourcePath(sources, absolutePath) {
854
- return path
855
- .relative(sources.taskDir, absolutePath)
856
- .replaceAll(path.sep, "/");
882
+ return path.relative(sources.taskDir, absolutePath).replaceAll(path.sep, "/");
857
883
  }
858
884
  function extractExplicitRequirementIds(...markdownInputs) {
859
885
  const ids = [];
@@ -873,9 +899,25 @@ function extractExplicitRequirementIds(...markdownInputs) {
873
899
  }
874
900
  function buildDagSourceBinding(sources) {
875
901
  const sourceEntries = [
876
- { kind: "requirement", path: sources.requirementPath, markdown: sources.requirementMarkdown },
877
- ...(sources.constraintMarkdown ? [{ kind: "constraint", path: sources.constraintPath, markdown: sources.constraintMarkdown }] : []),
878
- ...(sources.referenceDocuments ?? []).map((reference) => ({ kind: "reference", path: reference.path, markdown: reference.markdown })),
902
+ {
903
+ kind: "requirement",
904
+ path: sources.requirementPath,
905
+ markdown: sources.requirementMarkdown,
906
+ },
907
+ ...(sources.constraintMarkdown
908
+ ? [
909
+ {
910
+ kind: "constraint",
911
+ path: sources.constraintPath,
912
+ markdown: sources.constraintMarkdown,
913
+ },
914
+ ]
915
+ : []),
916
+ ...(sources.referenceDocuments ?? []).map((reference) => ({
917
+ kind: "reference",
918
+ path: reference.path,
919
+ markdown: reference.markdown,
920
+ })),
879
921
  ];
880
922
  return {
881
923
  schemaVersion: 1,
@@ -883,7 +925,9 @@ function buildDagSourceBinding(sources) {
883
925
  sources: sourceEntries.map((source) => ({
884
926
  kind: source.kind,
885
927
  path: toTaskRelativeSourcePath(sources, source.path),
886
- sha256: createHash("sha256").update(source.markdown, "utf8").digest("hex"),
928
+ sha256: createHash("sha256")
929
+ .update(source.markdown, "utf8")
930
+ .digest("hex"),
887
931
  })),
888
932
  requirementIds: extractExplicitRequirementIds(sources.requirementMarkdown, sources.constraintMarkdown, ...(sources.referenceDocuments ?? []).map((reference) => reference.markdown)),
889
933
  };
@@ -893,10 +937,7 @@ function buildSourceContextBlock(sources) {
893
937
  const requirementExcerpt = excerptMarkdown(sources.requirementMarkdown, {
894
938
  sourceRef: requirementRef,
895
939
  });
896
- const parts = [
897
- "## Task source: 需求.md",
898
- requirementExcerpt.text,
899
- ];
940
+ const parts = ["## Task source: 需求.md", requirementExcerpt.text];
900
941
  if (sources.constraintMarkdown) {
901
942
  const constraintRef = toTaskRelativeSourcePath(sources, sources.constraintPath);
902
943
  const constraintExcerpt = excerptMarkdown(sources.constraintMarkdown, {
@@ -940,7 +981,8 @@ async function loadMaterializedSourceReferences(sourceDir) {
940
981
  }
941
982
  else if (entry.isFile()) {
942
983
  // Skip index/manifest sidecars; keep only user/source reference content.
943
- if (entry.name === "index.json" || entry.name === "source-manifest.json") {
984
+ if (entry.name === "index.json" ||
985
+ entry.name === "source-manifest.json") {
944
986
  continue;
945
987
  }
946
988
  referencePaths.push(entryPath);
@@ -949,7 +991,9 @@ async function loadMaterializedSourceReferences(sourceDir) {
949
991
  }
950
992
  await collect(referenceDir);
951
993
  referencePaths.sort((left, right) => left.localeCompare(right));
952
- return Promise.all(referencePaths.slice(0, MAX_SOURCE_REFERENCE_DOCUMENTS).map(async (filePath) => ({
994
+ return Promise.all(referencePaths
995
+ .slice(0, MAX_SOURCE_REFERENCE_DOCUMENTS)
996
+ .map(async (filePath) => ({
953
997
  path: filePath,
954
998
  markdown: await readFile(filePath, "utf-8"),
955
999
  })));
@@ -1043,10 +1087,26 @@ async function prepareFrontendMockSources(sources) {
1043
1087
  capability.verifyCommands.push(command);
1044
1088
  }
1045
1089
  }
1090
+ const projectCapability = await discoverFrontendProjectCapability(repoRoot);
1091
+ const frontendRisk = classifyFrontendRisk({
1092
+ title: sources.taskConfig.title,
1093
+ requirementMarkdown: sources.requirementMarkdown,
1094
+ constraintMarkdown: sources.constraintMarkdown ?? undefined,
1095
+ allowedPaths: sources.taskConfig.allowedPaths,
1096
+ forbiddenPaths: sources.taskConfig.forbiddenPaths,
1097
+ complexity: sources.taskConfig.complexity,
1098
+ manifestEvidence: [
1099
+ projectCapability.framework,
1100
+ projectCapability.frameworkVersion ?? "",
1101
+ projectCapability.adapterGuidance,
1102
+ ].join("\n"),
1103
+ });
1046
1104
  return {
1047
1105
  ...sources,
1048
1106
  frontendMockCapability: capability,
1049
1107
  frontendMockMode: resolveFrontendMockMode(capability, sources.taskConfig, hasApiDependency(sources)),
1108
+ frontendProjectCapability: projectCapability,
1109
+ frontendRisk,
1050
1110
  };
1051
1111
  }
1052
1112
  function mergeFinalVerifyCommands(repoRoot, taskConfig, adapterCommands) {
@@ -1080,14 +1140,17 @@ export function buildStandardHybridDagFromTask(sources) {
1080
1140
  fallbackCommands: [],
1081
1141
  });
1082
1142
  const verifyShellTask = verifyShellCommands.length > 0
1083
- ? [{
1143
+ ? [
1144
+ {
1084
1145
  id: "verify-shell",
1085
1146
  depends_on: [implementId],
1086
1147
  role: "verifier",
1087
1148
  executor: "shell",
1088
1149
  complexity: "LOW",
1089
1150
  writePolicy: "read-only",
1090
- allowedPaths: taskConfig.allowedPaths.length > 0 ? taskConfig.allowedPaths : ["**"],
1151
+ allowedPaths: taskConfig.allowedPaths.length > 0
1152
+ ? taskConfig.allowedPaths
1153
+ : ["**"],
1091
1154
  forbiddenPaths,
1092
1155
  outputContract: "Archived final shell verification stdout/stderr with exit codes; no worktree writes.",
1093
1156
  subtask_prompt: "Run the adapter-resolved final verification commands before read-only verification review.",
@@ -1104,7 +1167,8 @@ export function buildStandardHybridDagFromTask(sources) {
1104
1167
  cwd: ".",
1105
1168
  timeoutMs: 300000,
1106
1169
  },
1107
- }]
1170
+ },
1171
+ ]
1108
1172
  : [];
1109
1173
  const sourceContext = buildSourceContextBlock(sources);
1110
1174
  const globalConstraints = [
@@ -1209,7 +1273,9 @@ export function buildStandardHybridDagFromTask(sources) {
1209
1273
  "Stay within writeSet. Do not write root artifacts/** unless artifacts paths are explicitly declared in writeSet.",
1210
1274
  writerDeliveryContract(taskConfig),
1211
1275
  sourceContext,
1212
- ].filter((value) => Boolean(value)).join("\n\n"),
1276
+ ]
1277
+ .filter((value) => Boolean(value))
1278
+ .join("\n\n"),
1213
1279
  },
1214
1280
  ...verifyShellTask,
1215
1281
  {
@@ -1505,13 +1571,131 @@ function resolveFrontendMockContextBlock(sources) {
1505
1571
  }
1506
1572
  return parts.join("\n");
1507
1573
  }
1574
+ function resolveFrontendCapabilityContextBlock(sources) {
1575
+ const risk = sources.frontendRisk;
1576
+ const capability = sources.frontendProjectCapability;
1577
+ const parts = ["## Frontend risk & project capability", ""];
1578
+ if (risk) {
1579
+ parts.push(`Risk level: ${risk.selectedRisk}`, `Risk reason: ${risk.reason}`, `Risk signals: ${risk.signals.join(", ") || "(none)"}`, `Rejected signals: ${risk.rejectedSignals.join(", ") || "(none)"}`, `Force full gates: ${risk.forceFullGates}`);
1580
+ }
1581
+ else {
1582
+ parts.push("Risk level: standard (not precomputed)");
1583
+ }
1584
+ if (capability) {
1585
+ parts.push("", capability.adapterGuidance);
1586
+ if (capability.designEvidence.normativePaths.length > 0) {
1587
+ parts.push(`openSpec normative candidates: ${capability.designEvidence.normativePaths.slice(0, 12).join(", ")}`);
1588
+ }
1589
+ parts.push(`A11y capability: ${capability.a11y.status}` +
1590
+ (capability.a11y.tools.length
1591
+ ? ` (${capability.a11y.tools.join(", ")})`
1592
+ : ""));
1593
+ parts.push("Browser accessibility verification: not-run (out of scope for this workflow).");
1594
+ }
1595
+ return parts.join("\n");
1596
+ }
1597
+ function pruneFrontendTasksForRisk(tasks, risk) {
1598
+ if (risk.forceFullGates || risk.selectedRisk !== "small") {
1599
+ return tasks;
1600
+ }
1601
+ // small topology: drop first design gate + plan revision + second design cycle;
1602
+ // keep mock assess/gate, single design review (final), contract, implement, verify, repair chain, review.
1603
+ const drop = new Set([
1604
+ "frontend-design-gate-pi",
1605
+ "frontend-first-design-gate-shell",
1606
+ "frontend-plan-revision-pi",
1607
+ ]);
1608
+ // If we drop plan-revision, contract shell must depend on plan-pi instead; final design review depends on plan.
1609
+ const filtered = tasks.filter((task) => !drop.has(task.id));
1610
+ const byId = new Map(filtered.map((task) => [task.id, task]));
1611
+ const remap = (deps) => {
1612
+ if (!deps)
1613
+ return [];
1614
+ const next = [];
1615
+ for (const dep of deps) {
1616
+ if (dep === "frontend-plan-revision-pi") {
1617
+ if (byId.has("frontend-plan-pi"))
1618
+ next.push("frontend-plan-pi");
1619
+ continue;
1620
+ }
1621
+ if (dep === "frontend-first-design-gate-shell" ||
1622
+ dep === "frontend-design-gate-pi") {
1623
+ // skip removed gates
1624
+ continue;
1625
+ }
1626
+ if (byId.has(dep) || dep === "frontend-implement-pi")
1627
+ next.push(dep);
1628
+ }
1629
+ return [...new Set(next)];
1630
+ };
1631
+ return filtered.map((task) => {
1632
+ const depends_on = remap(task.depends_on);
1633
+ // Ensure final design review still has plan + mock + contract path
1634
+ if (task.id === "frontend-final-design-review-pi") {
1635
+ for (const need of [
1636
+ "frontend-plan-pi",
1637
+ "frontend-mock-assess-pi",
1638
+ "frontend-implementation-contract-shell",
1639
+ ]) {
1640
+ if (byId.has(need) && !depends_on.includes(need))
1641
+ depends_on.push(need);
1642
+ }
1643
+ }
1644
+ if (task.id === "frontend-requirement-coverage-shell") {
1645
+ const shell = task.shell?.requirementCoverageGate
1646
+ ? {
1647
+ ...task.shell,
1648
+ requirementCoverageGate: {
1649
+ ...task.shell.requirementCoverageGate,
1650
+ fromNodeIds: task.shell.requirementCoverageGate.fromNodeIds.map((nodeId) => nodeId === "frontend-plan-revision-pi"
1651
+ ? "frontend-plan-pi"
1652
+ : nodeId),
1653
+ },
1654
+ }
1655
+ : task.shell;
1656
+ return { ...task, depends_on, shell };
1657
+ }
1658
+ if (task.id === "frontend-implementation-contract-shell") {
1659
+ const nextDeps = depends_on.filter((dep) => dep !== "frontend-plan-revision-pi");
1660
+ if (!nextDeps.includes("frontend-plan-pi") &&
1661
+ byId.has("frontend-plan-pi")) {
1662
+ nextDeps.push("frontend-plan-pi");
1663
+ }
1664
+ const shell = task.shell
1665
+ ? {
1666
+ ...task.shell,
1667
+ jsonArtifactGate: task.shell.jsonArtifactGate
1668
+ ? {
1669
+ ...task.shell.jsonArtifactGate,
1670
+ fromNodeId: "frontend-plan-pi",
1671
+ }
1672
+ : task.shell.jsonArtifactGate,
1673
+ }
1674
+ : task.shell;
1675
+ return { ...task, depends_on: nextDeps, shell };
1676
+ }
1677
+ if (task.id === "frontend-implement-pi") {
1678
+ // still requires final design gate
1679
+ for (const need of [
1680
+ "frontend-final-design-gate-shell",
1681
+ "frontend-implementation-contract-shell",
1682
+ ]) {
1683
+ if (byId.has(need) && !depends_on.includes(need))
1684
+ depends_on.push(need);
1685
+ }
1686
+ }
1687
+ return { ...task, depends_on };
1688
+ });
1689
+ }
1508
1690
  function buildFrontendHybridDagFromTask(sources) {
1509
1691
  const { taskConfig } = sources;
1510
1692
  const mockCapability = sources.frontendMockCapability ?? {
1511
1693
  status: "absent",
1512
1694
  evidencePaths: [],
1513
1695
  verifyCommands: [],
1514
- reasons: ["Frontend Mock capability was not precomputed; assessment must verify repository evidence."],
1696
+ reasons: [
1697
+ "Frontend Mock capability was not precomputed; assessment must verify repository evidence.",
1698
+ ],
1515
1699
  };
1516
1700
  const mockMode = sources.frontendMockMode ??
1517
1701
  resolveFrontendMockMode(mockCapability, taskConfig, hasApiDependency(sources));
@@ -1523,8 +1707,22 @@ function buildFrontendHybridDagFromTask(sources) {
1523
1707
  const forbiddenPaths = mergeForbiddenPaths(taskConfig);
1524
1708
  const implementPaths = resolveImplementPaths(taskConfig);
1525
1709
  const implementId = frontendImplementationNodeId();
1526
- const sourceContext = buildSourceContextBlock(sources);
1527
1710
  const mockContextBlock = resolveFrontendMockContextBlock(frontendSources);
1711
+ const capabilityContextBlock = resolveFrontendCapabilityContextBlock(frontendSources);
1712
+ const frontendRisk = frontendSources.frontendRisk ??
1713
+ classifyFrontendRisk({
1714
+ title: taskConfig.title,
1715
+ requirementMarkdown: sources.requirementMarkdown,
1716
+ constraintMarkdown: sources.constraintMarkdown ?? undefined,
1717
+ allowedPaths: taskConfig.allowedPaths,
1718
+ complexity: taskConfig.complexity,
1719
+ });
1720
+ const sourceContext = [
1721
+ buildSourceContextBlock(sources),
1722
+ capabilityContextBlock,
1723
+ ]
1724
+ .filter(Boolean)
1725
+ .join("\n\n");
1528
1726
  const hasMockVerifyCommands = (taskConfig.frontendMock?.verifyCommands.length ?? 0) > 0 ||
1529
1727
  mockCapability.verifyCommands.length > 0;
1530
1728
  const requirementIds = buildDagSourceBinding(sources).requirementIds;
@@ -1549,6 +1747,10 @@ function buildFrontendHybridDagFromTask(sources) {
1549
1747
  "Mock-backed behavior evidence proves only the documented frontend contract, never real API integration.",
1550
1748
  "frontend-implementation DAGs must complete deterministic static verification and behavior verification before final review.",
1551
1749
  "frontend review must block closeout unless review verdict is exactly VERDICT: pass.",
1750
+ `Frontend risk classification: ${frontendRisk.selectedRisk} — ${frontendRisk.reason}`,
1751
+ frontendRisk.forceFullGates
1752
+ ? "High-risk or supervised: keep full design gates; do not weaken write boundaries."
1753
+ : "Risk-aware topology may omit redundant design revision nodes for small tasks only.",
1552
1754
  ];
1553
1755
  // Guard: blocked mode — generate assessment-only DAG with no writer reachable
1554
1756
  if (mockMode === "blocked") {
@@ -1671,11 +1873,12 @@ function buildFrontendHybridDagFromTask(sources) {
1671
1873
  allowedPaths: readOnlyPaths,
1672
1874
  forbiddenPaths,
1673
1875
  skills: FRONTEND_IMPLEMENTATION_SKILLS,
1674
- outputContract: "Markdown implementation plan with Requirement Coverage, Implementation Steps, Target Files, UI State Handling, Styling / Component Strategy, Interaction Notes, Mock / API Strategy, Dependency Policy, Verification Plan, Real Integration Gap, and Residual Risks. No file writes.",
1876
+ outputContract: "Markdown implementation plan with Requirement Coverage, Implementation Steps, Target Files, UI State Handling, Styling / Component Strategy, Interaction Notes, Mock / API Strategy, Dependency Policy, Verification Plan, Real Integration Gap, and Residual Risks, followed by exactly one fenced json object conforming to frontend-implementation-contract-v1 when this node is the effective plan source. No file writes.",
1675
1877
  subtask_prompt: [
1676
1878
  "Based on frontend-contract-pi, frontend-scout-pi, and the gated frontend-mock-assess-pi strategy, return a minimal frontend implementation plan.",
1677
1879
  "Carry the selected Mock / API strategy, endpoint/fixture mapping, explicit activation, production-default-off rule, verification commands, and Real Integration Gap into the plan.",
1678
1880
  "Include ordered steps, target files, UI state handling, styling/component strategy, interaction notes, Mock/API strategy, dependency policy, deterministic verification entrypoints, and residual risks. Use only the fixed entrypoints below; implementation may add tests behind them but cannot replace them.",
1881
+ "End with exactly one fenced json object conforming to frontend-implementation-contract-v1 so small topology can materialize the contract without plan-revision.",
1679
1882
  requirementCoverageInstruction,
1680
1883
  "Read-only: do not modify code, docs, artifacts, or repository files.",
1681
1884
  fixedVerificationContext,
@@ -1742,7 +1945,7 @@ function buildFrontendHybridDagFromTask(sources) {
1742
1945
  allowedPaths: readOnlyPaths,
1743
1946
  forbiddenPaths,
1744
1947
  skills: FRONTEND_IMPLEMENTATION_SKILLS,
1745
- outputContract: "Markdown revision plan (pass case: first line PASS_NO_REVISION_NEEDED with Requirement Coverage confirmation; request-revision case: complete revised implementation plan with corrections from design findings and Requirement Coverage). No file writes.",
1948
+ outputContract: "Markdown revision plan followed by exactly one fenced json object conforming to frontend-implementation-contract-v1. The JSON is the authoritative materialization input. No file writes.",
1746
1949
  subtask_prompt: [
1747
1950
  "Consume frontend-plan-pi (original plan) and frontend-design-gate-pi (first design review findings).",
1748
1951
  "If the first design gate passed (VERDICT: pass from frontend-design-gate-pi), output exactly:",
@@ -1755,27 +1958,59 @@ function buildFrontendHybridDagFromTask(sources) {
1755
1958
  requirementCoverageInstruction,
1756
1959
  "Do not turn MOCK_STRATEGY: blocked into an implementable strategy without new repository or contract evidence that resolves every blocker.",
1757
1960
  "Read-only: do not modify code, docs, artifacts, or repository files. This node revises the plan only.",
1961
+ "End the response with exactly one fenced json object conforming to frontend-implementation-contract-v1. Bind it to the supplied task sources; map every requirement and applicable UI state to concrete implementation and verification targets or an explicit blocking evidence gap. Do not include secrets or unsafe paths.",
1758
1962
  sourceContext,
1759
1963
  ].join("\n\n"),
1760
1964
  },
1761
- ...(requirementIds.length > 0 ? [{
1762
- id: "frontend-requirement-coverage-shell",
1763
- depends_on: ["frontend-plan-revision-pi"],
1764
- role: "verifier",
1765
- executor: "shell",
1766
- complexity: "LOW",
1767
- writePolicy: "read-only",
1768
- allowedPaths: readOnlyPaths,
1769
- forbiddenPaths,
1770
- outputContract: "Deterministic current-run evidence that the original or revised frontend plan retains every explicit REQ-/BR-/AC- identifier from the bound task sources.",
1771
- subtask_prompt: "Block final design review when the current run's plan facts omit any explicit requirement identifier from the authoritative task sources.",
1772
- shell: {
1773
- commands: [],
1774
- requirementCoverageGate: { fromNodeIds: ["frontend-plan-revision-pi"], requiredIds: requirementIds, label: "frontend requirement coverage" },
1775
- cwd: ".",
1776
- timeoutMs: 60000,
1965
+ ...(requirementIds.length > 0
1966
+ ? [
1967
+ {
1968
+ id: "frontend-requirement-coverage-shell",
1969
+ depends_on: ["frontend-plan-revision-pi"],
1970
+ role: "verifier",
1971
+ executor: "shell",
1972
+ complexity: "LOW",
1973
+ writePolicy: "read-only",
1974
+ allowedPaths: readOnlyPaths,
1975
+ forbiddenPaths,
1976
+ outputContract: "Deterministic current-run evidence that the original or revised frontend plan retains every explicit REQ-/BR-/AC- identifier from the bound task sources.",
1977
+ subtask_prompt: "Block final design review when the current run's plan facts omit any explicit requirement identifier from the authoritative task sources.",
1978
+ shell: {
1979
+ commands: [],
1980
+ requirementCoverageGate: {
1981
+ fromNodeIds: ["frontend-plan-revision-pi"],
1982
+ requiredIds: requirementIds,
1983
+ label: "frontend requirement coverage",
1984
+ },
1985
+ cwd: ".",
1986
+ timeoutMs: 60000,
1987
+ },
1777
1988
  },
1778
- }] : []),
1989
+ ]
1990
+ : []),
1991
+ {
1992
+ id: "frontend-implementation-contract-shell",
1993
+ depends_on: ["frontend-plan-revision-pi"],
1994
+ role: "verifier",
1995
+ executor: "shell",
1996
+ complexity: "LOW",
1997
+ writePolicy: "read-only",
1998
+ allowedPaths: readOnlyPaths,
1999
+ forbiddenPaths,
2000
+ outputContract: "Run-owned validated frontend-implementation-contract-v1 artifact path, schema id, and SHA-256.",
2001
+ subtask_prompt: "Materialize the effective frontend plan as a source-bound structured contract; fail closed on missing or invalid output.",
2002
+ shell: {
2003
+ commands: [],
2004
+ jsonArtifactGate: {
2005
+ fromNodeId: "frontend-plan-revision-pi",
2006
+ schemaId: "frontend-implementation-contract-v1",
2007
+ artifactName: "frontend-implementation-contract.json",
2008
+ outputDir: "contracts",
2009
+ },
2010
+ cwd: ".",
2011
+ timeoutMs: 60000,
2012
+ },
2013
+ },
1779
2014
  {
1780
2015
  id: "frontend-final-design-review-pi",
1781
2016
  depends_on: [
@@ -1783,7 +2018,10 @@ function buildFrontendHybridDagFromTask(sources) {
1783
2018
  "frontend-plan-pi",
1784
2019
  "frontend-design-gate-pi",
1785
2020
  "frontend-mock-assess-pi",
1786
- ...(requirementIds.length > 0 ? ["frontend-requirement-coverage-shell"] : []),
2021
+ "frontend-implementation-contract-shell",
2022
+ ...(requirementIds.length > 0
2023
+ ? ["frontend-requirement-coverage-shell"]
2024
+ : []),
1787
2025
  ],
1788
2026
  role: "reviewer",
1789
2027
  executor: "pi",
@@ -1833,6 +2071,7 @@ function buildFrontendHybridDagFromTask(sources) {
1833
2071
  id: implementId,
1834
2072
  depends_on: [
1835
2073
  "frontend-final-design-gate-shell",
2074
+ "frontend-implementation-contract-shell",
1836
2075
  "frontend-plan-revision-pi",
1837
2076
  "frontend-final-design-review-pi",
1838
2077
  "frontend-plan-pi",
@@ -1847,22 +2086,28 @@ function buildFrontendHybridDagFromTask(sources) {
1847
2086
  allowedPaths: implementPaths.allowedPaths,
1848
2087
  forbiddenPaths,
1849
2088
  skills: FRONTEND_IMPLEMENTATION_SKILLS,
1850
- outputContract: "Markdown summary with Changed Files, Implemented Behavior, UI States Covered, Styling / Component Notes, Verification Attempted, and Residual Risks.",
2089
+ outputContract: "Markdown delivery summary with Contract Ref (path/schema/hash), Changed Files, Requirements Implemented, UI States, Tests Changed, Verification Attempts, Deviations, and Residual Risks. Follow fixed stages: contract confirm → tests → component/state → API/Mock → focused checks → diff cleanup.",
1851
2090
  subtask_prompt: [
1852
- "Implement the final approved frontend plan (from frontend-plan-revision-pi) with minimal focused changes.",
1853
- "Implement only the approved Mock strategy from frontend-mock-assess-pi as carried through the approved plan. Preserve the real request path as the default, require explicit test/dev activation, and never comment out or replace the real request with inline data.",
2091
+ "Implement against the validated run-owned Frontend Implementation Contract from frontend-implementation-contract-shell (path/schema/hash). Do not rebuild the contract from Markdown alone.",
2092
+ "Execute in fixed stages and report each in the delivery summary: (1) Contract confirm, (2) Tests sync, (3) Component/UI state implementation, (4) API/Mock wiring per contract.mockApi, (5) Focused checks behind frozen entrypoints only, (6) Diff cleanup.",
2093
+ "Map every requirement id and applicable UI state from the contract to concrete files. Do not invent shell verification commands; only frozen static/behavior entrypoints will run.",
2094
+ "Implement only the approved Mock strategy from frontend-mock-assess-pi as carried in the contract. Preserve the real request path as the default, require explicit test/dev activation, and never comment out or replace the real request with inline data.",
1854
2095
  "The frontend-final-design-review-pi verdict confirmed the plan is ready. Stay within writeSet and preserve unrelated files.",
1855
2096
  "For native, browser-intercept, or request-adapter, implement contract-aligned fixtures/states and a dev/test-only activation boundary in this same writer. For not-needed, do not add Mock files or a framework and state the positive reason.",
1856
- "Do not write root artifacts/** unless explicitly included in writeSet.",
2097
+ "Do not write root artifacts/** unless explicitly included in writeSet. Do not claim Browser/visual verification.",
1857
2098
  writerDeliveryContract(taskConfig),
1858
2099
  sourceContext,
1859
2100
  mockContextBlock,
1860
- ].filter((value) => Boolean(value)).join("\n\n"),
2101
+ ]
2102
+ .filter((value) => Boolean(value))
2103
+ .join("\n\n"),
1861
2104
  },
1862
2105
  // Optional dedicated Mock verification exists only when trusted commands
1863
2106
  // were frozen at generation time. Behavior verification remains required.
1864
2107
  ...(mockMode === "required" && hasMockVerifyCommands
1865
- ? [buildFrontendMockVerifyNode(frontendSources, implementId, readOnlyPaths, forbiddenPaths)]
2108
+ ? [
2109
+ buildFrontendMockVerifyNode(frontendSources, implementId, readOnlyPaths, forbiddenPaths),
2110
+ ]
1866
2111
  : []),
1867
2112
  {
1868
2113
  id: "frontend-static-verify-shell",
@@ -1882,6 +2127,7 @@ function buildFrontendHybridDagFromTask(sources) {
1882
2127
  verifyEvidence: staticVerifyEvidence,
1883
2128
  cwd: ".",
1884
2129
  timeoutMs: 300000,
2130
+ nonZeroExitPolicy: "record",
1885
2131
  },
1886
2132
  },
1887
2133
  {
@@ -1900,14 +2146,166 @@ function buildFrontendHybridDagFromTask(sources) {
1900
2146
  verifyEvidence: behaviorVerifyEvidence,
1901
2147
  cwd: ".",
1902
2148
  timeoutMs: 300000,
2149
+ nonZeroExitPolicy: "record",
1903
2150
  },
1904
2151
  },
1905
2152
  {
1906
- id: "frontend-review-pi",
2153
+ id: "frontend-verification-trace-shell",
1907
2154
  depends_on: [
2155
+ "frontend-behavior-verify-shell",
1908
2156
  "frontend-static-verify-shell",
2157
+ "frontend-implementation-contract-shell",
2158
+ ],
2159
+ role: "verifier",
2160
+ executor: "shell",
2161
+ complexity: "LOW",
2162
+ writePolicy: "read-only",
2163
+ allowedPaths: readOnlyPaths,
2164
+ forbiddenPaths,
2165
+ outputContract: "Deterministic verification trace: contract verificationTargets bound to current-run static/behavior commandLabels; target files/symbols exist; Browser/visual not-run. No worktree writes.",
2166
+ subtask_prompt: "Validate AC/UI/verification targets against frozen shell evidence and workspace files. Do not invent commands. Does not prove semantic test quality.",
2167
+ shell: {
2168
+ commands: ["frontend-verification-trace-gate"],
2169
+ cwd: ".",
2170
+ timeoutMs: 120000,
2171
+ nonZeroExitPolicy: "record",
2172
+ },
2173
+ },
2174
+ {
2175
+ id: "frontend-failure-assess-shell",
2176
+ depends_on: [
2177
+ "frontend-verification-trace-shell",
1909
2178
  "frontend-behavior-verify-shell",
2179
+ "frontend-static-verify-shell",
2180
+ "frontend-implementation-contract-shell",
2181
+ ],
2182
+ role: "verifier",
2183
+ executor: "shell",
2184
+ complexity: "LOW",
2185
+ writePolicy: "read-only",
2186
+ allowedPaths: readOnlyPaths,
2187
+ forbiddenPaths,
2188
+ outputContract: "Run-owned frontend-repair-assessment-v1 at contracts/frontend-repair-assessment.json classifying verify/trace failures as repairable or not.",
2189
+ subtask_prompt: "Assess current-run static/behavior/trace failure facts against the validated contract. Do not repair code.",
2190
+ shell: {
2191
+ commands: ["frontend-failure-assess-gate"],
2192
+ cwd: ".",
2193
+ timeoutMs: 60000,
2194
+ },
2195
+ },
2196
+ {
2197
+ id: "frontend-repair-contract-shell",
2198
+ depends_on: ["frontend-failure-assess-shell"],
2199
+ role: "verifier",
2200
+ executor: "shell",
2201
+ complexity: "LOW",
2202
+ writePolicy: "read-only",
2203
+ allowedPaths: readOnlyPaths,
2204
+ forbiddenPaths,
2205
+ outputContract: "Deterministic repair eligibility gate: pass when no failure or repairable assessment; fail-closed on non-repairable classes.",
2206
+ subtask_prompt: "Validate frontend-repair-assessment writeSet subset, attempt limit, and eligibility before repair writer.",
2207
+ shell: {
2208
+ commands: ["frontend-repair-contract-gate"],
2209
+ cwd: ".",
2210
+ timeoutMs: 60000,
2211
+ },
2212
+ },
2213
+ {
2214
+ id: "frontend-repair-pi",
2215
+ depends_on: [
2216
+ "frontend-repair-contract-shell",
2217
+ "frontend-failure-assess-shell",
1910
2218
  implementId,
2219
+ ],
2220
+ role: "implementer",
2221
+ executor: "pi",
2222
+ toolProfile: "write",
2223
+ complexity: resolveWriterComplexity(taskConfig),
2224
+ writePolicy: "exclusive",
2225
+ writeSet: implementPaths.writeSet,
2226
+ allowedPaths: implementPaths.allowedPaths,
2227
+ forbiddenPaths,
2228
+ skills: FRONTEND_IMPLEMENTATION_SKILLS,
2229
+ outputContract: "Repair summary or explicit no-op when assessment has no failure/eligible=false path already gated. Must not expand writeSet, re-interpret requirements, skip tests, or enable Mock by default.",
2230
+ subtask_prompt: [
2231
+ "Read contracts/frontend-repair-assessment.json and the validated frontend implementation contract.",
2232
+ "If failedNodeIds is empty, return no-op with evidence. If eligible repairable failure, apply the smallest fix inside the original implement writeSet only.",
2233
+ "Do not change lint/type/test config, do not add .skip/.only, do not comment out real requests, do not default-enable Mock, do not add dependencies.",
2234
+ "Do not re-plan requirements or expand allowed paths. Browser/visual remain not-run.",
2235
+ writerDeliveryContract(taskConfig),
2236
+ sourceContext,
2237
+ mockContextBlock,
2238
+ ]
2239
+ .filter((value) => Boolean(value))
2240
+ .join("\n\n"),
2241
+ },
2242
+ {
2243
+ id: "frontend-static-reverify-shell",
2244
+ depends_on: ["frontend-repair-pi"],
2245
+ role: "verifier",
2246
+ executor: "shell",
2247
+ complexity: "LOW",
2248
+ writePolicy: "read-only",
2249
+ allowedPaths: readOnlyPaths,
2250
+ forbiddenPaths,
2251
+ outputContract: "Archived static re-verification after repair using the same frozen commands; fail on nonzero.",
2252
+ subtask_prompt: "Re-run frozen static entrypoints after repair. Fresh evidence only; do not rewrite prior failure artifacts.",
2253
+ shell: {
2254
+ commands: staticShellCommands,
2255
+ verifyEvidence: staticVerifyEvidence,
2256
+ cwd: ".",
2257
+ timeoutMs: 300000,
2258
+ },
2259
+ },
2260
+ {
2261
+ id: "frontend-behavior-reverify-shell",
2262
+ depends_on: ["frontend-static-reverify-shell"],
2263
+ role: "verifier",
2264
+ executor: "shell",
2265
+ complexity: "LOW",
2266
+ writePolicy: "read-only",
2267
+ allowedPaths: behaviorPaths,
2268
+ forbiddenPaths,
2269
+ outputContract: "Archived behavior re-verification after repair using the same frozen commands; fail on nonzero.",
2270
+ subtask_prompt: "Re-run frozen behavior entrypoints after repair.",
2271
+ shell: {
2272
+ commands: behaviorShellCommands,
2273
+ verifyEvidence: behaviorVerifyEvidence,
2274
+ cwd: ".",
2275
+ timeoutMs: 300000,
2276
+ },
2277
+ },
2278
+ {
2279
+ id: "frontend-verification-retrace-shell",
2280
+ depends_on: [
2281
+ "frontend-behavior-reverify-shell",
2282
+ "frontend-static-reverify-shell",
2283
+ "frontend-implementation-contract-shell",
2284
+ ],
2285
+ role: "verifier",
2286
+ executor: "shell",
2287
+ complexity: "LOW",
2288
+ writePolicy: "read-only",
2289
+ allowedPaths: readOnlyPaths,
2290
+ forbiddenPaths,
2291
+ outputContract: "Re-run verification trace against contract and reverify shell evidence after repair.",
2292
+ subtask_prompt: "Trace AC/UI/verification targets against post-repair static/behavior evidence.",
2293
+ shell: {
2294
+ commands: ["frontend-verification-trace-gate"],
2295
+ cwd: ".",
2296
+ timeoutMs: 120000,
2297
+ },
2298
+ },
2299
+ {
2300
+ id: "frontend-review-pi",
2301
+ depends_on: [
2302
+ "frontend-verification-retrace-shell",
2303
+ "frontend-static-reverify-shell",
2304
+ "frontend-behavior-reverify-shell",
2305
+ "frontend-repair-pi",
2306
+ "frontend-failure-assess-shell",
2307
+ implementId,
2308
+ "frontend-implementation-contract-shell",
1911
2309
  "frontend-contract-pi",
1912
2310
  "frontend-plan-pi",
1913
2311
  "frontend-plan-revision-pi",
@@ -1929,6 +2327,8 @@ function buildFrontendHybridDagFromTask(sources) {
1929
2327
  "Review the frontend implementation and verification evidence.",
1930
2328
  "First non-empty line must be exactly VERDICT: pass or VERDICT: request-revision.",
1931
2329
  "Any Critical or Important finding must force VERDICT: request-revision.",
2330
+ "Read the validated frontend-implementation-contract, frontend-verification-trace evidence, static/behavior shell facts, and actual diff. Trace proves command/file/symbol binding only—not semantic correctness.",
2331
+ "Flag .skip/.only, deleted or weakened tests, unauthorized config changes, Mock-only evidence claimed as real integration, and Browser/visual claims (always not-run in this workflow).",
1932
2332
  "Use the direct contract, original plan, revision/no-op result, and final design review to reconstruct the approved plan and design verdict; do not infer them from the implementation summary.",
1933
2333
  "Treat a commented-out real request, default-enabled Mock, production entrypoint importing test mocks, API/fixture contract drift, unauthorized Mock dependency/path, or missing behavior evidence for the selected strategy as at least Important. Mock strategies require Mock-backed evidence; not-needed requires applicable real or no-remote behavior evidence. Verify that the real request remains the default when Mock activation is absent.",
1934
2334
  "Inspect the production/default-real-path static evidence directly and require Mock activation to be off for that check.",
@@ -1966,8 +2366,12 @@ function buildFrontendHybridDagFromTask(sources) {
1966
2366
  depends_on: [
1967
2367
  "frontend-review-gate-shell",
1968
2368
  "frontend-review-pi",
1969
- "frontend-static-verify-shell",
1970
- "frontend-behavior-verify-shell",
2369
+ "frontend-verification-retrace-shell",
2370
+ "frontend-static-reverify-shell",
2371
+ "frontend-behavior-reverify-shell",
2372
+ "frontend-repair-pi",
2373
+ "frontend-failure-assess-shell",
2374
+ "frontend-implementation-contract-shell",
1971
2375
  "frontend-mock-assess-pi",
1972
2376
  ...(mockMode === "required" && hasMockVerifyCommands
1973
2377
  ? ["frontend-mock-verify-shell"]
@@ -1985,6 +2389,7 @@ function buildFrontendHybridDagFromTask(sources) {
1985
2389
  outputContract: "Markdown closeout summary with Changes, Mock Decision / Strategy / Files / Verification / Production Boundary, Verification Evidence, Review Result, Frontend Status, Real Integration Status, Known Risks, and Follow-up. No file writes.",
1986
2390
  subtask_prompt: [
1987
2391
  "Return a frontend closeout summary covering Mock decision/strategy/files/verification/production boundary, changes, verification evidence, review result, known risks, and follow-up.",
2392
+ "Include a coverage matrix for each requirement id, applicable UI state, and verification target/check with status passed|failed|not-run|blocked|unavailable. Always state Browser accessibility verification: not-run and Visual regression: not-run. Use frontend-verification-trace facts; do not invent Browser evidence from component tests.",
1988
2393
  `When only Mock-backed evidence passed, state exactly Frontend status: mock-validated and Real integration: pending, summarize the Real Integration Gap, and name ${taskConfig.taskId}-real-api-integration-verify as the explicit follow-up task to create/run after backend readiness. This follow-up is not auto-created or auto-executed. Never describe Mock evidence as real API integration.`,
1989
2394
  "Read-only: do not modify code, docs, artifacts, or .harness/dag-runs/.",
1990
2395
  sourceContext,
@@ -1993,6 +2398,7 @@ function buildFrontendHybridDagFromTask(sources) {
1993
2398
  },
1994
2399
  ],
1995
2400
  };
2401
+ spec.tasks = pruneFrontendTasksForRisk(spec.tasks, frontendRisk);
1996
2402
  applyDefaultReadOnlyRetryPolicy(spec);
1997
2403
  parseDagSpec(spec);
1998
2404
  assertValidDagSpec(spec);
@@ -2049,10 +2455,64 @@ function buildBackendTestAnalysisContractGateNode(sources) {
2049
2455
  },
2050
2456
  };
2051
2457
  }
2458
+ function buildBackendTestEnvironmentScoutNode(sources) {
2459
+ return {
2460
+ id: "backend-test-environment-scout-pi",
2461
+ depends_on: ["backend-test-analysis-contract-shell"],
2462
+ role: "scout",
2463
+ executor: "pi",
2464
+ complexity: "MED",
2465
+ writePolicy: "read-only",
2466
+ allowedPaths: commonReadOnlyPaths(sources),
2467
+ forbiddenPaths: commonForbiddenPaths(sources),
2468
+ outputContract: "Pure Backend Test Execution Contract v1 JSON object matching docs/templates/backend-test-execution.schema.json. No Markdown prose and no file writes.",
2469
+ subtask_prompt: [
2470
+ "Read-only environment scout for backend-test pytest MVP.",
2471
+ "Return exactly one JSON object matching Backend Test Execution Contract v1 (schema docs/templates/backend-test-execution.schema.json).",
2472
+ "Prefer pure JSON; a single fenced json block is tolerated; no trailing prose.",
2473
+ "Discover only non-secret evidence: pytest config files (pytest.ini / pyproject.toml / setup.cfg test paths), candidate test roots, existing fixtures/clients, documented run commands, and env *names* (not values).",
2474
+ "Do NOT search the whole repo for secrets, .env values, tokens, private keys, or production credentials.",
2475
+ 'framework must be "pytest". Default targetMode to "in-process" unless evidence clearly shows an external service base URL env name or documented managed start/stop with sourceRef.',
2476
+ 'Do NOT select targetMode "managed-command" unless task source documents a safe start/stop command with an explicit sourceRef; otherwise leave managedCommand absent and record the gap in evidenceGaps.',
2477
+ "testRoot and workingDirectory must be repo-relative posix paths without .. or absolute form. Adapter default testRoot is testcase when evidence is incomplete.",
2478
+ "runner must not include secret values. report.format must be junit with a relativeHint under the run (e.g. reports/backend-test-junit.xml).",
2479
+ "requiredEnvNames lists env NAMES only. baseUrlEnvName is required only for external-running-service and must match ^[A-Z_][A-Z0-9_]*$.",
2480
+ "Record incomplete discovery in evidenceGaps. Populate evidenceRefs with repo-relative paths actually read.",
2481
+ "Required top-level keys: schemaVersion, framework, runner, testRoot, workingDirectory, report, targetMode, existingFixtures, authenticationMode, requiredEnvNames, dataIsolation, evidenceGaps, evidenceRefs.",
2482
+ "Read-only: do not modify code, docs, artifacts, or repository files.",
2483
+ buildSourceContextBlock(sources),
2484
+ ].join("\n\n"),
2485
+ };
2486
+ }
2487
+ function buildBackendTestExecutionContractGateNode(sources) {
2488
+ return {
2489
+ id: "backend-test-execution-contract-shell",
2490
+ depends_on: ["backend-test-environment-scout-pi"],
2491
+ role: "verifier",
2492
+ executor: "shell",
2493
+ complexity: "LOW",
2494
+ writePolicy: "read-only",
2495
+ allowedPaths: commonReadOnlyPaths(sources),
2496
+ forbiddenPaths: commonForbiddenPaths(sources),
2497
+ outputContract: "Validated run-owned Backend Test Execution Contract v1 artifact pointer, schema ID, and SHA-256.",
2498
+ subtask_prompt: "Materialize and validate the backend-test execution contract under the current DAG run.",
2499
+ shell: {
2500
+ commands: [],
2501
+ jsonArtifactGate: {
2502
+ fromNodeId: "backend-test-environment-scout-pi",
2503
+ schemaId: "backend-test-execution-v1",
2504
+ artifactName: "backend-test-execution.json",
2505
+ outputDir: "contracts",
2506
+ },
2507
+ cwd: ".",
2508
+ timeoutMs: 60000,
2509
+ },
2510
+ };
2511
+ }
2052
2512
  function buildGenerateBackendFunctionalCasesNode(sources) {
2053
2513
  return {
2054
2514
  id: "generate-backend-functional-cases-pi",
2055
- depends_on: ["backend-test-analysis-contract-shell"],
2515
+ depends_on: ["backend-test-execution-contract-shell"],
2056
2516
  role: "implementer",
2057
2517
  executor: "pi",
2058
2518
  toolProfile: "write",
@@ -2064,7 +2524,8 @@ function buildGenerateBackendFunctionalCasesNode(sources) {
2064
2524
  // 注意:Pi 节点超时由 executor 层控制(默认 30 分钟)
2065
2525
  // 如需调整,在 harness.json 的 executors.pi 中配置 modelConfig.timeoutMs
2066
2526
  subtask_prompt: [
2067
- "Read the validated structured artifact pointer from backend-test-analysis-contract-shell and generate cases only from that JSON contract.", ,
2527
+ "Read the validated structured artifact pointer from backend-test-analysis-contract-shell and generate cases only from that JSON contract.",
2528
+ ,
2068
2529
  "",
2069
2530
  "## Output Steps (do in order):",
2070
2531
  "1. First, output a brief summary: how many modules, how many cases planned per module",
@@ -2090,15 +2551,92 @@ function buildGenerateBackendFunctionalCasesNode(sources) {
2090
2551
  "",
2091
2552
  "## Constraints:",
2092
2553
  "- Stay within writeSet: testcase/md/**",
2093
- "- Do NOT re-read source documents or fall back to free-form analysis — use the validated structured artifact only", ,
2554
+ "- Do NOT re-read source documents or fall back to free-form analysis — use the validated structured artifact only",
2555
+ ,
2094
2556
  "- Do not write root artifacts/**",
2095
2557
  ].join("\n\n"),
2096
2558
  };
2097
2559
  }
2560
+ function buildEmitBackendCaseManifestNode(sources) {
2561
+ return {
2562
+ id: "emit-backend-case-manifest-pi",
2563
+ depends_on: [
2564
+ "generate-backend-functional-cases-pi",
2565
+ "backend-test-analysis-contract-shell",
2566
+ ],
2567
+ role: "scout",
2568
+ executor: "pi",
2569
+ complexity: "MED",
2570
+ writePolicy: "read-only",
2571
+ allowedPaths: commonReadOnlyPaths(sources),
2572
+ forbiddenPaths: commonForbiddenPaths(sources),
2573
+ outputContract: "Pure Backend Test Case Manifest v1 JSON (schema docs/templates/backend-test-case-manifest.schema.json). No file writes; model must not write .harness/**.",
2574
+ subtask_prompt: [
2575
+ "Emit Backend Test Case Manifest v1 as pure JSON (or one fenced json block with no trailing text).",
2576
+ "Read-only: use validated contracts/backend-test-analysis.json pointer + testcase/md/** only. Do not write repository files or .harness/**.",
2577
+ "sourceBinding must match the analysis contract / DAG source binding exactly (taskId, requirementPath, requirementSha256, referencePaths, requirementIds).",
2578
+ "For each functional case under testcase/md/: caseId BE-<MODULE>-<NNN>, acIds[], title, category, automationStatus.",
2579
+ "After case generation (pre-pytest), default automationStatus=planned. Use skipped/unsupported only with gapReason. Use generated only when file+symbol already exist.",
2580
+ "evidenceGaps: structured gaps for explicit AC-* that cannot be mapped to a case.",
2581
+ "Do NOT invent coverage percentages. Optional coverageSummary must match deterministic counts (gate recomputes/validates).",
2582
+ "No secrets or credential-shaped fields.",
2583
+ ].join("\n\n"),
2584
+ };
2585
+ }
2586
+ function buildBackendTestCaseManifestGateNode(sources) {
2587
+ return {
2588
+ id: "backend-test-case-manifest-shell",
2589
+ depends_on: ["emit-backend-case-manifest-pi"],
2590
+ role: "verifier",
2591
+ executor: "shell",
2592
+ complexity: "LOW",
2593
+ writePolicy: "read-only",
2594
+ allowedPaths: commonReadOnlyPaths(sources),
2595
+ forbiddenPaths: commonForbiddenPaths(sources),
2596
+ outputContract: "Validated run-owned Backend Test Case Manifest v1 at contracts/backend-test-case-manifest.json (schemaId backend-test-case-manifest-v1) with deterministic AC coverage.",
2597
+ subtask_prompt: "Materialize and validate Backend Test Case Manifest v1; fail closed on duplicate IDs, unknown AC, missing AC coverage without gap, or skipped without gapReason.",
2598
+ shell: {
2599
+ commands: [],
2600
+ jsonArtifactGate: {
2601
+ fromNodeId: "emit-backend-case-manifest-pi",
2602
+ schemaId: "backend-test-case-manifest-v1",
2603
+ artifactName: "backend-test-case-manifest.json",
2604
+ outputDir: "contracts",
2605
+ },
2606
+ cwd: ".",
2607
+ timeoutMs: 60000,
2608
+ },
2609
+ };
2610
+ }
2611
+ function buildBackendTestTraceabilityGateNode(sources) {
2612
+ return {
2613
+ id: "backend-test-traceability-gate-shell",
2614
+ depends_on: [
2615
+ "generate-backend-pytest-pi",
2616
+ "backend-test-case-manifest-shell",
2617
+ ],
2618
+ role: "verifier",
2619
+ executor: "shell",
2620
+ complexity: "LOW",
2621
+ writePolicy: "read-only",
2622
+ allowedPaths: commonReadOnlyPaths(sources),
2623
+ forbiddenPaths: commonForbiddenPaths(sources),
2624
+ outputContract: "Deterministic traceability: generated cases have real file/symbol; skipped/unsupported have gapReason; convention symbols scanned under testcase/**/test_*.py.",
2625
+ subtask_prompt: "Fail closed when generated automation claims do not resolve to workspace pytest symbols, or skip/unsupported lacks gapReason.",
2626
+ shell: {
2627
+ commands: ["backend-test-traceability-gate"],
2628
+ cwd: ".",
2629
+ timeoutMs: 60000,
2630
+ },
2631
+ };
2632
+ }
2098
2633
  function buildReviewBackendCasesNode(sources) {
2099
2634
  return {
2100
2635
  id: "review-backend-cases-pi",
2101
- depends_on: ["generate-backend-functional-cases-pi", "backend-test-analysis-contract-shell"],
2636
+ depends_on: [
2637
+ "backend-test-case-manifest-shell",
2638
+ "backend-test-analysis-contract-shell",
2639
+ ],
2102
2640
  role: "reviewer",
2103
2641
  executor: "pi",
2104
2642
  complexity: "HIGH",
@@ -2107,7 +2645,7 @@ function buildReviewBackendCasesNode(sources) {
2107
2645
  forbiddenPaths: commonForbiddenPaths(sources),
2108
2646
  outputContract: "Plain Markdown whose first non-empty line is VERDICT: pass or VERDICT: request-revision; followed by Findings and Coverage Assessment. No file writes.",
2109
2647
  subtask_prompt: [
2110
- "Review the generated backend functional test cases under testcase/md/.",
2648
+ "Review the generated backend functional test cases under testcase/md/ and the validated Case Manifest v1.",
2111
2649
  "",
2112
2650
  "## Mandatory First Line:",
2113
2651
  "First non-empty line must be exactly: VERDICT: pass or VERDICT: request-revision",
@@ -2116,9 +2654,10 @@ function buildReviewBackendCasesNode(sources) {
2116
2654
  "- ID format: every case uses BE-<MODULE>-<NNN>",
2117
2655
  "- Positive coverage: each acceptance criterion (AC-xxx) has happy-path case",
2118
2656
  "- Negative coverage: error scenarios (invalid input, not found, state violations)",
2119
- "- Traceability: each AC maps to at least one case ID",
2657
+ "- Traceability: each AC maps to at least one case ID (prefer contracts/backend-test-case-manifest.json coverageSummary)",
2120
2658
  "- Case structure: ID, Title, Precondition, Steps, Expected Result",
2121
2659
  "- No duplicate IDs across files",
2660
+ "- Manifest consistency: MD cases align with manifest caseId/acIds; do not invent coverage %",
2122
2661
  "",
2123
2662
  "## Conditional Coverage (check ONLY if mentioned in upstream analysis):",
2124
2663
  "- Boundary coverage: check ONLY if analyze-inputs-pi mentions value ranges, length limits, numeric bounds, or format constraints",
@@ -2133,13 +2672,13 @@ function buildReviewBackendCasesNode(sources) {
2133
2672
  "- Any Critical fails OR Important > 2 → VERDICT: request-revision",
2134
2673
  "",
2135
2674
  "## Output After Verdict:",
2136
- "1. Coverage Assessment table (AC → case IDs)",
2675
+ "1. Coverage Assessment table (AC → case IDs) using manifest + MD",
2137
2676
  "2. Findings list (Critical/Important/Informational)",
2138
2677
  "3. Statistics (total cases, positive/negative/boundary breakdown)",
2139
2678
  "",
2140
2679
  "## Constraints:",
2141
2680
  "- Read-only: do not modify files",
2142
- "- Read the validated backend-test analysis artifact pointer from upstream and use it for AC/source-binding coverage checks", ,
2681
+ "- Read validated analysis + case manifest artifacts; do not recompute coverage percentages",
2143
2682
  "- Use testcase/md/ files for case review",
2144
2683
  ].join("\n\n"),
2145
2684
  };
@@ -2172,7 +2711,10 @@ function buildReviewBackendCasesGateNode(sources) {
2172
2711
  function buildGenerateBackendPytestNode(sources) {
2173
2712
  return {
2174
2713
  id: "generate-backend-pytest-pi",
2175
- depends_on: ["review-backend-cases-gate-shell"],
2714
+ depends_on: [
2715
+ "review-backend-cases-gate-shell",
2716
+ "backend-test-execution-contract-shell",
2717
+ ],
2176
2718
  role: "implementer",
2177
2719
  executor: "pi",
2178
2720
  toolProfile: "write",
@@ -2193,6 +2735,12 @@ function buildGenerateBackendPytestNode(sources) {
2193
2735
  subtask_prompt: [
2194
2736
  "Convert the reviewed test cases under testcase/md/ into pytest automation code.",
2195
2737
  "",
2738
+ "## Inputs (MUST use validated contracts):",
2739
+ "- Reviewed cases under testcase/md/ (after review-backend-cases-gate-shell).",
2740
+ "- Validated Backend Test Analysis v1 under the current run contracts/ (analysis gate).",
2741
+ "- Validated Backend Test Execution Contract v1 under contracts/backend-test-execution.json (execution gate).",
2742
+ "Use only fixture names, env NAMES, testRoot, targetMode, and field/API facts already present in those contracts or reviewed cases. Do not invent production credentials or secret values.",
2743
+ "",
2196
2744
  "## Output Steps (do in order):",
2197
2745
  "1. First, output a brief summary: how many files, how many test functions planned",
2198
2746
  "2. Then write each test file under testcase/",
@@ -2262,8 +2810,8 @@ function buildGenerateBackendPytestNode(sources) {
2262
2810
  "",
2263
2811
  "### Field Name Resolution",
2264
2812
  "Field names MUST come from the upstream analyze-inputs-pi output (API Endpoints section) or reviewed cases, NOT guessed. For example:",
2265
- "- If API spec defines {\"ret\": 0, \"msg\": \"success\"}, assert response.json()['ret'] and response.json()['msg']",
2266
- "- If API spec defines {\"code\": 4001, \"message\": \"error\"}, assert response.json()['code'] and response.json()['message']",
2813
+ '- If API spec defines {"ret": 0, "msg": "success"}, assert response.json()[\'ret\'] and response.json()[\'msg\']',
2814
+ '- If API spec defines {"code": 4001, "message": "error"}, assert response.json()[\'code\'] and response.json()[\'message\']',
2267
2815
  "",
2268
2816
  "## Conditional Implementation (include ONLY if test cases exist):",
2269
2817
  "- Authentication tests: implement ONLY if testcase/md/ contains auth-related cases",
@@ -2284,26 +2832,41 @@ function buildGenerateBackendPytestNode(sources) {
2284
2832
  function buildExecuteBackendPytestNode(sources) {
2285
2833
  // Keep the target worktree read-only: JUnit is runner-owned evidence under
2286
2834
  // the current DAG run and moves with active → completed/paused lifecycle.
2835
+ // Adapter default testRoot is frozen at DAG generation time (auditable) and
2836
+ // cross-checked against the materialized execution contract in preflight.
2837
+ const frozenTestRoot = BACKEND_TEST_EXECUTION_DEFAULT_TEST_ROOT;
2838
+ const preflightCommand = buildBackendTestExecutionPreflightShellSnippet({
2839
+ expectedTestRoot: frozenTestRoot,
2840
+ });
2841
+ // Map pytest exit 0/1 → node success ONLY when JUnit exists (assertion-fail is a
2842
+ // legal result). Do not change global shell ok semantics. Persist raw exit for parse.
2287
2843
  const pytestCommand = [
2288
- 'test -n "${HARNESS_DAG_RUN_DIR:-}" || { echo "missing HARNESS_DAG_RUN_DIR for backend pytest report" >&2; exit 2; }',
2844
+ preflightCommand,
2289
2845
  'REPORT="${HARNESS_DAG_RUN_DIR}/reports/backend-test-junit.xml"',
2846
+ 'EXIT_FILE="${HARNESS_DAG_RUN_DIR}/reports/backend-test-pytest-exit.txt"',
2290
2847
  'mkdir -p "$(dirname "${REPORT}")"',
2291
- 'PYTHONDONTWRITEBYTECODE=1 python -m pytest testcase/ -v -p no:cacheprovider --junitxml="${REPORT}"',
2292
- 'STATUS=$?',
2848
+ `PYTHONDONTWRITEBYTECODE=1 python -m pytest ${frozenTestRoot}/ -v -p no:cacheprovider --junitxml="\${REPORT}"`,
2849
+ "STATUS=$?",
2850
+ 'printf "%s" "${STATUS}" > "${EXIT_FILE}"',
2293
2851
  'printf "JUnit report: %s\\n" "${REPORT}"',
2852
+ 'printf "pytestExitCode=%s\\n" "${STATUS}"',
2853
+ 'if { [ "${STATUS}" -eq 0 ] || [ "${STATUS}" -eq 1 ]; } && [ -s "${REPORT}" ]; then exit 0; fi',
2294
2854
  'exit "${STATUS}"',
2295
2855
  ].join("; ");
2296
2856
  return {
2297
2857
  id: "execute-backend-pytest-shell",
2298
- depends_on: ["generate-backend-pytest-pi"],
2858
+ depends_on: [
2859
+ "backend-test-traceability-gate-shell",
2860
+ "backend-test-execution-contract-shell",
2861
+ ],
2299
2862
  role: "verifier",
2300
2863
  executor: "shell",
2301
2864
  complexity: "LOW",
2302
2865
  writePolicy: "read-only",
2303
2866
  allowedPaths: commonReadOnlyPaths(sources),
2304
2867
  forbiddenPaths: commonForbiddenPaths(sources),
2305
- outputContract: "Archived pytest stdout/stderr with exit codes; JUnit XML is runner-owned evidence at $HARNESS_DAG_RUN_DIR/reports/backend-test-junit.xml. Must not modify worktree files, testcase sources, production code, or assertions.",
2306
- subtask_prompt: "Run pytest for the backend test suite; write JUnit evidence only under the current HARNESS_DAG_RUN_DIR/reports/.",
2868
+ outputContract: "Archived pytest stdout/stderr; raw pytestExitCode side-channel + JUnit at $HARNESS_DAG_RUN_DIR/reports/**. Exit 0/1 with non-empty JUnit finishes the node so parse/classify/retrospect can run; assertion failures remain recorded in exit file.",
2869
+ subtask_prompt: "Run pytest for the backend test suite; write JUnit + pytestExitCode evidence only under the current HARNESS_DAG_RUN_DIR/reports/.",
2307
2870
  shell: {
2308
2871
  commands: [pytestCommand],
2309
2872
  verifyEvidence: buildVerifyEvidence({
@@ -2318,10 +2881,61 @@ function buildExecuteBackendPytestNode(sources) {
2318
2881
  },
2319
2882
  };
2320
2883
  }
2884
+ function buildParseBackendTestResultNode(sources) {
2885
+ return {
2886
+ id: "parse-backend-test-result-shell",
2887
+ depends_on: ["execute-backend-pytest-shell"],
2888
+ role: "verifier",
2889
+ executor: "shell",
2890
+ complexity: "LOW",
2891
+ writePolicy: "read-only",
2892
+ allowedPaths: commonReadOnlyPaths(sources),
2893
+ forbiddenPaths: commonForbiddenPaths(sources),
2894
+ outputContract: "Validated run-owned Backend Test Result v1 at contracts/backend-test-result.json (schemaId backend-test-result-v1) with outcome/counts/failures from deterministic JUnit parse.",
2895
+ subtask_prompt: "Materialize Backend Test Result v1 from JUnit + pytestExitCode under the current DAG run (fail-closed on missing/corrupt report).",
2896
+ shell: {
2897
+ commands: [],
2898
+ jsonArtifactGate: {
2899
+ fromNodeId: "execute-backend-pytest-shell",
2900
+ schemaId: "backend-test-result-v1",
2901
+ artifactName: "backend-test-result.json",
2902
+ outputDir: "contracts",
2903
+ },
2904
+ cwd: ".",
2905
+ timeoutMs: 60000,
2906
+ },
2907
+ };
2908
+ }
2909
+ function buildClassifyBackendTestResultNode(sources) {
2910
+ return {
2911
+ id: "classify-backend-test-result-pi",
2912
+ depends_on: ["parse-backend-test-result-shell"],
2913
+ role: "reviewer",
2914
+ executor: "pi",
2915
+ complexity: "MED",
2916
+ writePolicy: "read-only",
2917
+ allowedPaths: commonReadOnlyPaths(sources),
2918
+ forbiddenPaths: commonForbiddenPaths(sources),
2919
+ outputContract: "Pure JSON classification: category in {ProductBug,TestBug,EnvFailure,ContractMismatch,FlakyTest,Unknown}, evidence[], confidence (capped), notes. No file writes.",
2920
+ subtask_prompt: [
2921
+ "Read-only classifier for Backend Test Result v1.",
2922
+ "Return exactly one JSON object (prefer pure JSON; single fenced json block tolerated; no trailing prose).",
2923
+ "Read contracts/backend-test-result.json (run-owned Result v1). Do NOT invent pass rates from raw logs.",
2924
+ "category must be one of: ProductBug, TestBug, EnvFailure, ContractMismatch, FlakyTest, Unknown.",
2925
+ "Hard constraints:",
2926
+ "- Single-run failure MUST NOT use FlakyTest (use Unknown, TestBug, or ProductBug).",
2927
+ "- executionStatus/outcome collection-error, command-error, or report-error MUST NOT use ProductBug.",
2928
+ "- Prefer EnvFailure/Unknown/TestBug for env, import, collection, and missing-report cases.",
2929
+ "- confidence must respect deterministic caps (≤0.75 for assertion failures; ≤0.6 for env/collection).",
2930
+ "Include evidence[] referencing result fields (outcome, failed, failures[].name, executionStatus).",
2931
+ "Read-only: do not modify code, docs, artifacts, or repository files.",
2932
+ ].join("\n\n"),
2933
+ };
2934
+ }
2321
2935
  function buildTestRetrospectNode(sources) {
2322
2936
  return {
2323
2937
  id: "test-retrospect-pi",
2324
- depends_on: ["execute-backend-pytest-shell"],
2938
+ depends_on: ["classify-backend-test-result-pi"],
2325
2939
  role: "closeout",
2326
2940
  executor: "pi",
2327
2941
  toolProfile: "write",
@@ -2331,23 +2945,29 @@ function buildTestRetrospectNode(sources) {
2331
2945
  allowedPaths: ["docs/test-reports/**"],
2332
2946
  forbiddenPaths: commonForbiddenPaths(sources),
2333
2947
  subtask_prompt: [
2334
- "Read upstream outputs (review report + pytest results) and generate a test retrospective report.",
2948
+ "Read upstream Result v1 + Case Manifest coverageSummary + classification and generate a test retrospective report.",
2335
2949
  "",
2336
2950
  "## Output Steps (do in order):",
2337
2951
  "1. First, output the maturity rating on the first line: Rating: A/B/C/D",
2338
2952
  "2. Then write the full report under docs/test-reports/",
2339
2953
  "",
2954
+ "## Stats authority (deterministic only):",
2955
+ "- Pass rate, failed/error/skipped counts, and failure list MUST come from contracts/backend-test-result.json only.",
2956
+ "- AC coverage ratio / case counts MUST come from contracts/backend-test-case-manifest.json coverageSummary (or gate-derived fields). Do NOT invent coverage %.",
2957
+ "- Use classify-backend-test-result-pi JSON as interpretive evidence only.",
2958
+ "- NEVER rewrite a failed result as passed. Outcome gate (not this report) is authoritative for task success.",
2959
+ "",
2340
2960
  "## Report Structure:",
2341
2961
  "1. Maturity Rating with rationale",
2342
- "2. Test Coverage Summary (total cases, pass rate, failed case analysis)",
2962
+ "2. Test Coverage Summary (manifest coverageSummary + Result v1 pass rate)",
2343
2963
  "3. Review Findings and resolution status",
2344
- "4. Failed Test Analysis (if any)",
2964
+ "4. Failed Test Analysis (if any) + classification category",
2345
2965
  "5. Recommendations for improvement",
2346
2966
  "",
2347
2967
  "## Rating Criteria:",
2348
- "- A: 100% acceptance criteria covered + 100% pytest pass + no Critical findings",
2349
- "- B: ≥80% coverage + ≥90% pass + Low findings only",
2350
- "- C: ≥60% coverage + ≥70% pass + no Critical findings",
2968
+ "- A: coverageSummary.acCoverageRatio=1 + 100% pytest pass + no Critical findings",
2969
+ "- B: acCoverageRatio0.8 + ≥90% pass + Low findings only",
2970
+ "- C: acCoverageRatio0.6 + ≥70% pass + no Critical findings",
2351
2971
  "- D: below C thresholds",
2352
2972
  "",
2353
2973
  "## Constraints:",
@@ -2357,6 +2977,33 @@ function buildTestRetrospectNode(sources) {
2357
2977
  ].join("\n\n"),
2358
2978
  };
2359
2979
  }
2980
+ function buildBackendTestOutcomeGateNode(sources) {
2981
+ const gateCommand = buildBackendTestOutcomeGateShellSnippet();
2982
+ return {
2983
+ id: "backend-test-outcome-gate-shell",
2984
+ depends_on: ["test-retrospect-pi"],
2985
+ role: "verifier",
2986
+ executor: "shell",
2987
+ complexity: "LOW",
2988
+ writePolicy: "read-only",
2989
+ allowedPaths: commonReadOnlyPaths(sources),
2990
+ forbiddenPaths: commonForbiddenPaths(sources),
2991
+ outputContract: "Shell exit 0 only when Result v1 outcome=passed with failed=0 and error=0; non-zero otherwise. Ignores retrospective Markdown.",
2992
+ subtask_prompt: "Gate the backend-test DAG on run-owned Result v1 shell facts only (not retrospective prose).",
2993
+ shell: {
2994
+ commands: [gateCommand],
2995
+ verifyEvidence: buildVerifyEvidence({
2996
+ phase: "final",
2997
+ quota: "full",
2998
+ commandSource: "inline",
2999
+ fallbackCommands: [gateCommand],
3000
+ finalFullRequired: true,
3001
+ }),
3002
+ cwd: ".",
3003
+ timeoutMs: 60000,
3004
+ },
3005
+ };
3006
+ }
2360
3007
  const BACKEND_TEST_DEFAULTS = {
2361
3008
  ...HYBRID_DEFAULTS,
2362
3009
  writePolicy: "read-only",
@@ -2388,6 +3035,9 @@ function buildBackendTestHybridDag(sources) {
2388
3035
  "review-backend-cases-gate-shell must block pytest generation unless the review verdict is exactly VERDICT: pass.",
2389
3036
  "If a target test filename already exists under testcase/, add a numeric suffix (_01, _02, ...); never overwrite or append to existing files.",
2390
3037
  "execute-backend-pytest-shell must not modify test assertions or production code to make tests pass; test failures indicate potential implementation issues and must be reported honestly.",
3038
+ "parse-backend-test-result-shell materializes Backend Test Result v1 from JUnit + pytestExitCode; classify/retrospect run on pass and assertion-fail; backend-test-outcome-gate-shell uses result.outcome only.",
3039
+ "backend-test-case-manifest-shell validates schemaId backend-test-case-manifest-v1 and materializes contracts/backend-test-case-manifest.json; AC coverage is fail-closed and deterministic.",
3040
+ "backend-test-traceability-gate-shell verifies generated file/symbol existence after pytest generation and before execute; models must not invent coverage percentages.",
2391
3041
  ];
2392
3042
  const spec = {
2393
3043
  version: 3,
@@ -2409,12 +3059,20 @@ function buildBackendTestHybridDag(sources) {
2409
3059
  tasks: [
2410
3060
  buildAnalyzeInputsNode(sources),
2411
3061
  buildBackendTestAnalysisContractGateNode(sources),
3062
+ buildBackendTestEnvironmentScoutNode(sources),
3063
+ buildBackendTestExecutionContractGateNode(sources),
2412
3064
  buildGenerateBackendFunctionalCasesNode(sources),
3065
+ buildEmitBackendCaseManifestNode(sources),
3066
+ buildBackendTestCaseManifestGateNode(sources),
2413
3067
  buildReviewBackendCasesNode(sources),
2414
3068
  buildReviewBackendCasesGateNode(sources),
2415
3069
  buildGenerateBackendPytestNode(sources),
3070
+ buildBackendTestTraceabilityGateNode(sources),
2416
3071
  buildExecuteBackendPytestNode(sources),
3072
+ buildParseBackendTestResultNode(sources),
3073
+ buildClassifyBackendTestResultNode(sources),
2417
3074
  buildTestRetrospectNode(sources),
3075
+ buildBackendTestOutcomeGateNode(sources),
2418
3076
  ],
2419
3077
  };
2420
3078
  applyDefaultReadOnlyRetryPolicy(spec);
@@ -2427,8 +3085,12 @@ function buildBackendTestHybridDag(sources) {
2427
3085
  // ---------------------------------------------------------------------------
2428
3086
  function buildFrontendTestHybridDag(sources) {
2429
3087
  const config = sources.taskConfig.frontendTest ?? { maxCasesPerBatch: 20 };
2430
- const hasFrontendTestWriteScope = sources.taskConfig.allowedPaths.some((pattern) => pattern === "testcase/frontend/**" || pattern === "testcase/**" || pattern === "**");
2431
- const hasReportWriteScope = sources.taskConfig.allowedPaths.some((pattern) => pattern === "docs/test-reports/**" || pattern === "docs/**" || pattern === "**");
3088
+ const hasFrontendTestWriteScope = sources.taskConfig.allowedPaths.some((pattern) => pattern === "testcase/frontend/**" ||
3089
+ pattern === "testcase/**" ||
3090
+ pattern === "**");
3091
+ const hasReportWriteScope = sources.taskConfig.allowedPaths.some((pattern) => pattern === "docs/test-reports/**" ||
3092
+ pattern === "docs/**" ||
3093
+ pattern === "**");
2432
3094
  if (!hasFrontendTestWriteScope || !hasReportWriteScope) {
2433
3095
  throw new Error('frontend-test requires task.json allowedPaths to include both "testcase/frontend/**" and "docs/test-reports/**" (or explicit containing globs).');
2434
3096
  }
@@ -2474,44 +3136,163 @@ function buildFrontendTestHybridDag(sources) {
2474
3136
  "Browser startup for generated cases must be playwright-cli open --browser=chrome --headed <base-url>.",
2475
3137
  "Token settings are post-case stop thresholds, never a hard provider token cap. Unstarted cases after a threshold are blocked: token-budget-exhausted.",
2476
3138
  ],
2477
- defaults: { ...HYBRID_DEFAULTS, writePolicy: "read-only", contextProfile: sources.taskConfig.contextProfile },
3139
+ defaults: {
3140
+ ...HYBRID_DEFAULTS,
3141
+ writePolicy: "read-only",
3142
+ contextProfile: sources.taskConfig.contextProfile,
3143
+ },
2478
3144
  skillsByRole: {
2479
- planner: ["loop-agent"], scout: ["playwright-cli"], implementer: ["playwright-cli-case-generator", "playwright-cli", "webapp-testing"], reviewer: ["requesting-code-review"], verifier: ["playwright-cli", "webapp-testing"], closeout: ["loop-agent", "verification-before-completion"],
3145
+ planner: ["loop-agent"],
3146
+ scout: ["playwright-cli"],
3147
+ implementer: [
3148
+ "playwright-cli-case-generator",
3149
+ "playwright-cli",
3150
+ "webapp-testing",
3151
+ ],
3152
+ reviewer: ["requesting-code-review"],
3153
+ verifier: ["playwright-cli", "webapp-testing"],
3154
+ closeout: ["loop-agent", "verification-before-completion"],
2480
3155
  },
2481
3156
  executorModels: sources.executorModelMatrix ?? DEFAULT_DAG_EXECUTOR_MODELS,
2482
3157
  tasks: [
2483
3158
  {
2484
- id: "retrieve-frontend-test-context-pi", depends_on: [], role: "planner", executor: "pi", toolProfile: "write", complexity: "HIGH", writePolicy: "exclusive", writeSet: ragWriteSet, allowedPaths: [...commonReadOnlyPaths(sources), ...ragWriteSet], forbiddenPaths: forbidden,
3159
+ id: "retrieve-frontend-test-context-pi",
3160
+ depends_on: [],
3161
+ role: "planner",
3162
+ executor: "pi",
3163
+ toolProfile: "write",
3164
+ complexity: "HIGH",
3165
+ writePolicy: "exclusive",
3166
+ writeSet: ragWriteSet,
3167
+ allowedPaths: [...commonReadOnlyPaths(sources), ...ragWriteSet],
3168
+ forbiddenPaths: forbidden,
2485
3169
  outputContract: "Write testcase/frontend/rag/context.md and coverage-map.md with traceable UI/API/test-environment facts.",
2486
- subtask_prompt: ["Build the frontend test RAG package.", "Read task source, relevant routes/components/API or Mock facts, existing tests, and execution contract. Write only testcase/frontend/rag/context.md and coverage-map.md.", "Record AC IDs, source paths, routes, states, roles, fixture/data prerequisites, API mapping status, risks, and isolated execution contract. Do not guess unavailable facts.", buildSourceContextBlock(sources)].join("\n\n"),
3170
+ subtask_prompt: [
3171
+ "Build the frontend test RAG package.",
3172
+ "Read task source, relevant routes/components/API or Mock facts, existing tests, and execution contract. Write only testcase/frontend/rag/context.md and coverage-map.md.",
3173
+ "Record AC IDs, source paths, routes, states, roles, fixture/data prerequisites, API mapping status, risks, and isolated execution contract. Do not guess unavailable facts.",
3174
+ buildSourceContextBlock(sources),
3175
+ ].join("\n\n"),
2487
3176
  },
2488
3177
  {
2489
- id: "generate-frontend-functional-cases-pi", depends_on: ["retrieve-frontend-test-context-pi"], role: "implementer", executor: "pi", toolProfile: "write", complexity: "HIGH", writePolicy: "exclusive", writeSet: casesWriteSet, allowedPaths: [...ragWriteSet, ...casesWriteSet], forbiddenPaths: forbidden,
3178
+ id: "generate-frontend-functional-cases-pi",
3179
+ depends_on: ["retrieve-frontend-test-context-pi"],
3180
+ role: "implementer",
3181
+ executor: "pi",
3182
+ toolProfile: "write",
3183
+ complexity: "HIGH",
3184
+ writePolicy: "exclusive",
3185
+ writeSet: casesWriteSet,
3186
+ allowedPaths: [...ragWriteSet, ...casesWriteSet],
3187
+ forbiddenPaths: forbidden,
2490
3188
  outputContract: "Write executable Markdown frontend cases, index.md, and manifest.json schemaVersion 1; no test source code.",
2491
- subtask_prompt: ["Use skill playwright-cli-case-generator.", "Read only testcase/frontend/rag/context.md, testcase/frontend/rag/coverage-map.md, and existing testcase/frontend/cases/. Write only testcase/frontend/cases/**.", "Generate Markdown cases, index.md and manifest.json (schemaVersion 1; cases[] with caseId, casePath, dimension, acIds, evidenceDir). IDs use FE-<FEATURE>-<NNN>-<dimension>; dimensions core|boundary|flow|backend.", "Never infer API fields, constraints, SLA, credentials, or unrecorded test data. Do not create pytest or Playwright source. Every browser start command is: playwright-cli open --browser=chrome --headed <base-url>.", "Each case must be independent, declare its session/preconditions/data cleanup, UI assertions, evidence paths under testcase/frontend/evidence/<case-id>/, and mark unsafe/missing dependencies blocked."].join("\n\n"),
3189
+ subtask_prompt: [
3190
+ "Use skill playwright-cli-case-generator.",
3191
+ "Read only testcase/frontend/rag/context.md, testcase/frontend/rag/coverage-map.md, and existing testcase/frontend/cases/. Write only testcase/frontend/cases/**.",
3192
+ "Generate Markdown cases, index.md and manifest.json (schemaVersion 1; cases[] with caseId, casePath, dimension, acIds, evidenceDir). IDs use FE-<FEATURE>-<NNN>-<dimension>; dimensions core|boundary|flow|backend.",
3193
+ "Never infer API fields, constraints, SLA, credentials, or unrecorded test data. Do not create pytest or Playwright source. Every browser start command is: playwright-cli open --browser=chrome --headed <base-url>.",
3194
+ "Each case must be independent, declare its session/preconditions/data cleanup, UI assertions, evidence paths under testcase/frontend/evidence/<case-id>/, and mark unsafe/missing dependencies blocked.",
3195
+ ].join("\n\n"),
2492
3196
  },
2493
3197
  {
2494
- id: "review-frontend-cases-pi", depends_on: ["generate-frontend-functional-cases-pi"], role: "reviewer", executor: "pi", complexity: "HIGH", writePolicy: "read-only", allowedPaths: [...ragWriteSet, ...casesWriteSet], forbiddenPaths: forbidden,
2495
- outputContract: "First line VERDICT: pass or VERDICT: request-revision, followed by AC-to-case coverage and execution risk findings; no writes.", subtask_prompt: "Review only the RAG package and frontend Markdown cases. Verify traceability, independent execution, safe data/environment handling, manifest correctness, and evidence requirements. The verdict is advisory and does not block case execution.",
3198
+ id: "review-frontend-cases-pi",
3199
+ depends_on: ["generate-frontend-functional-cases-pi"],
3200
+ role: "reviewer",
3201
+ executor: "pi",
3202
+ complexity: "HIGH",
3203
+ writePolicy: "read-only",
3204
+ allowedPaths: [...ragWriteSet, ...casesWriteSet],
3205
+ forbiddenPaths: forbidden,
3206
+ outputContract: "First line VERDICT: pass or VERDICT: request-revision, followed by AC-to-case coverage and execution risk findings; no writes.",
3207
+ subtask_prompt: "Review only the RAG package and frontend Markdown cases. Verify traceability, independent execution, safe data/environment handling, manifest correctness, and evidence requirements. The verdict is advisory and does not block case execution.",
2496
3208
  },
2497
3209
  {
2498
- id: "materialize-frontend-case-manifest-shell", depends_on: ["review-frontend-cases-pi"], role: "verifier", executor: "shell", complexity: "LOW", writePolicy: "read-only", allowedPaths: casesWriteSet, forbiddenPaths: forbidden,
2499
- outputContract: "stdout is exactly JSON { cases: [...] } after deterministic frontend manifest validation.", subtask_prompt: "Validate and materialize the generated frontend case manifest.", shell: { commands: [manifestValidation], cwd: ".", timeoutMs: 120000 },
3210
+ id: "materialize-frontend-case-manifest-shell",
3211
+ depends_on: ["review-frontend-cases-pi"],
3212
+ role: "verifier",
3213
+ executor: "shell",
3214
+ complexity: "LOW",
3215
+ writePolicy: "read-only",
3216
+ allowedPaths: casesWriteSet,
3217
+ forbiddenPaths: forbidden,
3218
+ outputContract: "stdout is exactly JSON { cases: [...] } after deterministic frontend manifest validation.",
3219
+ subtask_prompt: "Validate and materialize the generated frontend case manifest.",
3220
+ shell: { commands: [manifestValidation], cwd: ".", timeoutMs: 120000 },
2500
3221
  },
2501
3222
  {
2502
- id: "execute-frontend-cases-map", depends_on: ["materialize-frontend-case-manifest-shell"], role: "verifier", executor: "static", complexity: "LOW", writePolicy: "none", allowedPaths: [], forbiddenPaths: forbidden,
2503
- outputContract: "Serial aggregate of case execution summaries, evidence paths, tokens, and token-budget blocked cases.", subtask_prompt: "Expand and execute the validated frontend case manifest serially.", static: { resultMarkdown: "Frontend case map expansion barrier." },
2504
- dynamicExpansion: { type: "map_agent", workflowNodeId: "execute-frontend-cases-map", itemsFrom: "$.nodes['materialize-frontend-case-manifest-shell'].output.cases", itemName: "case", maxItems: config.maxCasesPerBatch, maxExpandedNodes: config.maxCasesPerBatch, childIdPrefix: "execute-frontend-case", tokenBudget: { maxTokensPerCase: config.maxTokensPerCase, maxTotalTokens: config.maxTotalTokens }, childTask: {
2505
- executor: "pi", role: "verifier", skills: ["playwright-cli", "webapp-testing"], toolProfile: "write", complexity: "MED", writePolicy: "exclusive", allowedPaths: ["testcase/frontend/cases/{{case.caseId}}.md", "testcase/frontend/rag/context.md", "testcase/frontend/rag/coverage-map.md", `${evidenceRoot}/{{case.caseId}}/**`], forbiddenPaths: forbidden, writeSet: [`${evidenceRoot}/{{case.caseId}}/**`], outputContract: "Compact JSON <=1200 characters with case status, evidence paths, error summary, and tokens.", subtaskPromptTemplate: ["Execute exactly case {{case.caseId}} from {{case.casePath}} using playwright-cli and webapp-testing. This is a fresh Pi session; do not use /new.", "Use only the declared isolated test environment. If CLI/browser/base URL/credentials/fixture isolation is missing, record blocked rather than installing tools or guessing.", "Use playwright-cli open --browser=chrome --headed <base-url>. Persist execution.md, case-result.json, screenshots/trace/video/logs under {{case.evidenceDir}} before returning.", "A business failed or blocked case is a recorded result, not a node failure. Close the session and return only compact JSON (<=1200 chars): {caseId,status,evidencePaths,errorSummary,tokens}."].join("\n\n")
2506
- } },
3223
+ id: "execute-frontend-cases-map",
3224
+ depends_on: ["materialize-frontend-case-manifest-shell"],
3225
+ role: "verifier",
3226
+ executor: "static",
3227
+ complexity: "LOW",
3228
+ writePolicy: "none",
3229
+ allowedPaths: [],
3230
+ forbiddenPaths: forbidden,
3231
+ outputContract: "Serial aggregate of case execution summaries, evidence paths, tokens, and token-budget blocked cases.",
3232
+ subtask_prompt: "Expand and execute the validated frontend case manifest serially.",
3233
+ static: { resultMarkdown: "Frontend case map expansion barrier." },
3234
+ dynamicExpansion: {
3235
+ type: "map_agent",
3236
+ workflowNodeId: "execute-frontend-cases-map",
3237
+ itemsFrom: "$.nodes['materialize-frontend-case-manifest-shell'].output.cases",
3238
+ itemName: "case",
3239
+ maxItems: config.maxCasesPerBatch,
3240
+ maxExpandedNodes: config.maxCasesPerBatch,
3241
+ childIdPrefix: "execute-frontend-case",
3242
+ tokenBudget: {
3243
+ maxTokensPerCase: config.maxTokensPerCase,
3244
+ maxTotalTokens: config.maxTotalTokens,
3245
+ },
3246
+ childTask: {
3247
+ executor: "pi",
3248
+ role: "verifier",
3249
+ skills: ["playwright-cli", "webapp-testing"],
3250
+ toolProfile: "write",
3251
+ complexity: "MED",
3252
+ writePolicy: "exclusive",
3253
+ allowedPaths: [
3254
+ "testcase/frontend/cases/{{case.caseId}}.md",
3255
+ "testcase/frontend/rag/context.md",
3256
+ "testcase/frontend/rag/coverage-map.md",
3257
+ `${evidenceRoot}/{{case.caseId}}/**`,
3258
+ ],
3259
+ forbiddenPaths: forbidden,
3260
+ writeSet: [`${evidenceRoot}/{{case.caseId}}/**`],
3261
+ outputContract: "Compact JSON <=1200 characters with case status, evidence paths, error summary, and tokens.",
3262
+ subtaskPromptTemplate: [
3263
+ "Execute exactly case {{case.caseId}} from {{case.casePath}} using playwright-cli and webapp-testing. This is a fresh Pi session; do not use /new.",
3264
+ "Use only the declared isolated test environment. If CLI/browser/base URL/credentials/fixture isolation is missing, record blocked rather than installing tools or guessing.",
3265
+ "Use playwright-cli open --browser=chrome --headed <base-url>. Persist execution.md, case-result.json, screenshots/trace/video/logs under {{case.evidenceDir}} before returning.",
3266
+ "A business failed or blocked case is a recorded result, not a node failure. Close the session and return only compact JSON (<=1200 chars): {caseId,status,evidencePaths,errorSummary,tokens}.",
3267
+ ].join("\n\n"),
3268
+ },
3269
+ },
2507
3270
  },
2508
3271
  {
2509
- id: "review-frontend-execution-pi", depends_on: ["execute-frontend-cases-map"], role: "reviewer", executor: "pi", complexity: "HIGH", writePolicy: "read-only", allowedPaths: ["testcase/frontend/**"], forbiddenPaths: forbidden,
2510
- outputContract: "Read-only AC-to-case-to-browser-evidence review, including failed, blocked and token-budget-exhausted cases.", subtask_prompt: "Review the frontend case aggregate and on-disk case/evidence artifacts. A passed case requires assertion plus screenshot or equivalent browser evidence; failed/blocked cases require reasons. Do not replace browser evidence with model conclusions.",
3272
+ id: "review-frontend-execution-pi",
3273
+ depends_on: ["execute-frontend-cases-map"],
3274
+ role: "reviewer",
3275
+ executor: "pi",
3276
+ complexity: "HIGH",
3277
+ writePolicy: "read-only",
3278
+ allowedPaths: ["testcase/frontend/**"],
3279
+ forbiddenPaths: forbidden,
3280
+ outputContract: "Read-only AC-to-case-to-browser-evidence review, including failed, blocked and token-budget-exhausted cases.",
3281
+ subtask_prompt: "Review the frontend case aggregate and on-disk case/evidence artifacts. A passed case requires assertion plus screenshot or equivalent browser evidence; failed/blocked cases require reasons. Do not replace browser evidence with model conclusions.",
2511
3282
  },
2512
3283
  {
2513
- id: "frontend-test-retrospect-pi", depends_on: ["review-frontend-execution-pi"], role: "closeout", executor: "pi", toolProfile: "write", complexity: "MED", writePolicy: "exclusive", writeSet: ["docs/test-reports/**"], allowedPaths: ["testcase/frontend/**", "docs/test-reports/**"], forbiddenPaths: forbidden,
2514
- outputContract: "Write frontend-test-retrospect-<date>.md with coverage, pass/fail/blocked, risks, findings, and A/B/C/D rating.", subtask_prompt: "Write the frontend test retrospective under docs/test-reports/. Summarize coverage, passed/failed/blocked cases (including token-budget-exhausted), review findings, browser anomalies, residual risks, and A/B/C/D rating. Blocked cases never count as passed.",
3284
+ id: "frontend-test-retrospect-pi",
3285
+ depends_on: ["review-frontend-execution-pi"],
3286
+ role: "closeout",
3287
+ executor: "pi",
3288
+ toolProfile: "write",
3289
+ complexity: "MED",
3290
+ writePolicy: "exclusive",
3291
+ writeSet: ["docs/test-reports/**"],
3292
+ allowedPaths: ["testcase/frontend/**", "docs/test-reports/**"],
3293
+ forbiddenPaths: forbidden,
3294
+ outputContract: "Write frontend-test-retrospect-<date>.md with coverage, pass/fail/blocked, risks, findings, and A/B/C/D rating.",
3295
+ subtask_prompt: "Write the frontend test retrospective under docs/test-reports/. Summarize coverage, passed/failed/blocked cases (including token-budget-exhausted), review findings, browser anomalies, residual risks, and A/B/C/D rating. Blocked cases never count as passed.",
2515
3296
  },
2516
3297
  ],
2517
3298
  };
@@ -2640,7 +3421,9 @@ export function assertSafeKnowledgeSyncFeatureId(featureId) {
2640
3421
  if (!KNOWLEDGE_SYNC_FEATURE_ID_RE.test(trimmed)) {
2641
3422
  throw new Error(`knowledge-sync featureId must match F-<id> (letters/digits/._- only); got ${JSON.stringify(featureId)}`);
2642
3423
  }
2643
- if (trimmed.includes("..") || trimmed.includes("/") || trimmed.includes("\\")) {
3424
+ if (trimmed.includes("..") ||
3425
+ trimmed.includes("/") ||
3426
+ trimmed.includes("\\")) {
2644
3427
  throw new Error(`knowledge-sync featureId must not contain path segments: ${featureId}`);
2645
3428
  }
2646
3429
  return trimmed;
@@ -3060,7 +3843,10 @@ function buildKgBootstrapInventoryNode(sources) {
3060
3843
  executor: "shell",
3061
3844
  complexity: "LOW",
3062
3845
  writePolicy: "exclusive",
3063
- writeSet: ["knowledge/bootstrap/inventory.json", "knowledge/bootstrap/status.yaml"],
3846
+ writeSet: [
3847
+ "knowledge/bootstrap/inventory.json",
3848
+ "knowledge/bootstrap/status.yaml",
3849
+ ],
3064
3850
  allowedPaths: ["knowledge/bootstrap/**", "./**"],
3065
3851
  forbiddenPaths: commonForbiddenPaths(sources),
3066
3852
  outputContract: "Deterministic inventory.json under knowledge/bootstrap/ from directory and feature signals.",
@@ -3087,10 +3873,7 @@ function buildKgBootstrapProposeNode(sources) {
3087
3873
  toolProfile: "write",
3088
3874
  complexity: "HIGH",
3089
3875
  writePolicy: "exclusive",
3090
- writeSet: [
3091
- "knowledge/bootstrap/staging/**",
3092
- "knowledge/bootstrap/runs/**",
3093
- ],
3876
+ writeSet: ["knowledge/bootstrap/staging/**", "knowledge/bootstrap/runs/**"],
3094
3877
  allowedPaths: [
3095
3878
  "knowledge/bootstrap/**",
3096
3879
  "knowledge/**",
@@ -3285,11 +4068,7 @@ function buildKgBootstrapPromoteNode(sources) {
3285
4068
  "knowledge/graph/**",
3286
4069
  "features/**",
3287
4070
  ],
3288
- forbiddenPaths: [
3289
- ...commonForbiddenPaths(sources),
3290
- "src/**",
3291
- "testcase/**",
3292
- ],
4071
+ forbiddenPaths: [...commonForbiddenPaths(sources), "src/**", "testcase/**"],
3293
4072
  outputContract: "Promote staging → formal knowledge trees without overwriting existing files; copy edges.manual.yaml if absent.",
3294
4073
  subtask_prompt: "B5 promote: merge new files only (no overwrite of existing asserted content).",
3295
4074
  shell: {
@@ -3581,22 +4360,117 @@ function buildWriteSetAuditNode(sources) {
3581
4360
  ].join("\n\n"),
3582
4361
  };
3583
4362
  }
4363
+ function buildWriteSetAuditFormatRepairNode(sources, options) {
4364
+ return {
4365
+ id: options.id,
4366
+ depends_on: [options.auditNodeId],
4367
+ role: "reviewer",
4368
+ executor: "pi",
4369
+ complexity: "LOW",
4370
+ writePolicy: "read-only",
4371
+ allowedPaths: commonReadOnlyPaths(sources),
4372
+ forbiddenPaths: commonForbiddenPaths(sources),
4373
+ outputContract: "Plain Markdown whose first non-empty line is exactly VERDICT: pass or VERDICT: request-revision, followed by the original audit findings without substantive changes. No file writes.",
4374
+ subtask_prompt: [
4375
+ `Normalize the output format of ${options.auditNodeId}; this is the single read-only format-repair attempt for that audit.`,
4376
+ "First non-empty line must be exactly VERDICT: pass or VERDICT: request-revision.",
4377
+ "If the upstream audit already contains a valid verdict, preserve it exactly. If it omitted or malformed the verdict but states an unambiguous audit conclusion, add only the matching canonical verdict and preserve the findings.",
4378
+ "Do not add, remove, or reclassify substantive findings. If the upstream conclusion is ambiguous or cannot be preserved safely, emit VERDICT: request-revision and report the format ambiguity.",
4379
+ "Do not infer a pass from general prose, expand task allowedPaths, or edit files.",
4380
+ buildSourceContextBlock(sources),
4381
+ ].join("\n\n"),
4382
+ };
4383
+ }
4384
+ function buildWriteSetFormatGateNode(sources) {
4385
+ return {
4386
+ id: "write-set-format-gate-shell",
4387
+ depends_on: ["write-set-audit-format-repair-pi"],
4388
+ role: "verifier",
4389
+ executor: "shell",
4390
+ complexity: "LOW",
4391
+ writePolicy: "read-only",
4392
+ allowedPaths: commonReadOnlyPaths(sources),
4393
+ forbiddenPaths: commonForbiddenPaths(sources),
4394
+ outputContract: "Deterministic initial write-set verdict format gate: accept pass or request-revision so the bounded plan-revision stage can run; reject missing or unexpected verdicts.",
4395
+ subtask_prompt: "Validate the normalized initial write-set audit verdict before the bounded plan-revision stage. This gate does not authorize implementation writes.",
4396
+ shell: {
4397
+ commands: [],
4398
+ verdictGate: {
4399
+ fromNodeId: "write-set-audit-format-repair-pi",
4400
+ accept: ["VERDICT: pass", "VERDICT: request-revision"],
4401
+ label: "initial write-set audit format",
4402
+ lineMode: "first-verdict-line",
4403
+ },
4404
+ cwd: ".",
4405
+ timeoutMs: 60000,
4406
+ },
4407
+ };
4408
+ }
4409
+ function buildPlanRevisionNode(sources) {
4410
+ return {
4411
+ id: "plan-revision-pi",
4412
+ depends_on: [
4413
+ "write-set-format-gate-shell",
4414
+ "plan-pi",
4415
+ "write-set-audit-format-repair-pi",
4416
+ ],
4417
+ role: "planner",
4418
+ executor: "pi",
4419
+ complexity: "MED",
4420
+ writePolicy: "read-only",
4421
+ allowedPaths: commonReadOnlyPaths(sources),
4422
+ forbiddenPaths: commonForbiddenPaths(sources),
4423
+ outputContract: "PASS_NO_REVISION_NEEDED when the normalized initial audit passed, otherwise a complete revised implementation plan with a corrected WriteSet Coverage Matrix. No file writes.",
4424
+ subtask_prompt: [
4425
+ "Perform the single bounded plan-revision round after the normalized initial write-set audit.",
4426
+ "If the normalized verdict is VERDICT: pass, output PASS_NO_REVISION_NEEDED and do not change the original plan.",
4427
+ "If it is VERDICT: request-revision, return a complete revised plan that resolves every audit finding and includes a corrected WriteSet Coverage Matrix.",
4428
+ "Do not expand task.json.allowedPaths, weaken forbiddenPaths, or edit files.",
4429
+ buildSourceContextBlock(sources),
4430
+ ].join("\n\n"),
4431
+ };
4432
+ }
4433
+ function buildFinalWriteSetAuditNode(sources) {
4434
+ return {
4435
+ id: "final-write-set-audit-pi",
4436
+ depends_on: [
4437
+ "plan-pi",
4438
+ "plan-revision-pi",
4439
+ "write-set-audit-format-repair-pi",
4440
+ ],
4441
+ role: "reviewer",
4442
+ executor: "pi",
4443
+ complexity: "MED",
4444
+ writePolicy: "read-only",
4445
+ allowedPaths: commonReadOnlyPaths(sources),
4446
+ forbiddenPaths: commonForbiddenPaths(sources),
4447
+ outputContract: "Plain Markdown whose first non-empty line is VERDICT: pass or VERDICT: request-revision; includes final writeSet coverage findings after the single plan-revision round. No file writes.",
4448
+ subtask_prompt: [
4449
+ "Perform the final write-set audit after the single bounded plan-revision round.",
4450
+ "When plan-revision-pi returned PASS_NO_REVISION_NEEDED, audit the original plan-pi output. Otherwise audit the complete revised plan.",
4451
+ "First non-empty line must be exactly VERDICT: pass or VERDICT: request-revision.",
4452
+ "Request revision if required files still lack a single exclusive owner, writeSet is broad/placeholder, forbidden paths overlap, or any initial finding remains unresolved.",
4453
+ "Do not expand task allowedPaths or edit files.",
4454
+ buildSourceContextBlock(sources),
4455
+ ].join("\n\n"),
4456
+ };
4457
+ }
3584
4458
  function buildWriteSetGateNode(sources) {
3585
4459
  return {
3586
4460
  id: "write-set-gate-shell",
3587
- depends_on: ["write-set-audit-pi"],
4461
+ depends_on: ["final-write-set-audit-format-repair-pi"],
3588
4462
  role: "verifier",
3589
4463
  executor: "shell",
3590
4464
  complexity: "LOW",
3591
4465
  writePolicy: "read-only",
3592
4466
  allowedPaths: commonReadOnlyPaths(sources),
3593
4467
  forbiddenPaths: commonForbiddenPaths(sources),
3594
- outputContract: "Deterministic write-set audit verdict gate: exit 0 only when write-set-audit-pi first non-empty assistant output line is pass.",
3595
- subtask_prompt: "Deterministic gate: block the implementation writer unless write-set-audit-pi emitted VERDICT: pass.",
4468
+ outputContract: "Deterministic final write-set audit verdict gate: exit 0 only when final-write-set-audit-format-repair-pi emits VERDICT: pass after the bounded revision round.",
4469
+ subtask_prompt: "Deterministic gate: block the implementation writer unless the normalized final write-set audit emitted VERDICT: pass.",
3596
4470
  shell: {
3597
4471
  commands: [],
3598
4472
  verdictGate: {
3599
- fromNodeId: "write-set-audit-pi",
4473
+ fromNodeId: "final-write-set-audit-format-repair-pi",
3600
4474
  accept: ["VERDICT: pass"],
3601
4475
  label: "write-set audit",
3602
4476
  lineMode: "first-verdict-line",
@@ -3717,7 +4591,9 @@ function buildRepairNode(sources) {
3717
4591
  "If VERDICT: pass, return no-op with evidence. Re-run focused tests when you change code.",
3718
4592
  writerDeliveryContract(sources.taskConfig),
3719
4593
  buildSourceContextBlock(sources),
3720
- ].filter((value) => Boolean(value)).join("\n\n"),
4594
+ ]
4595
+ .filter((value) => Boolean(value))
4596
+ .join("\n\n"),
3721
4597
  };
3722
4598
  }
3723
4599
  function buildHardVerifyNode(sources) {
@@ -3769,7 +4645,8 @@ function buildDecisionNode(sources) {
3769
4645
  outputContract: "Markdown with exactly one DECISION_ENVELOPE_JSON fenced block plus evidence summary. No file writes.",
3770
4646
  subtask_prompt: [
3771
4647
  "Return an advisory Decision Gate envelope for the supervised DAG outcome.",
3772
- "Include exactly one DECISION_ENVELOPE_JSON fenced block and concise evidence. Read-only: do not modify files.",
4648
+ "Read-only: do not modify files. Review deterministic verification, review findings, write boundaries, and risks before deciding.",
4649
+ buildDecisionEnvelopePromptContract(),
3773
4650
  buildSourceContextBlock(sources),
3774
4651
  ].join("\n\n"),
3775
4652
  decisionGate: { enabled: true, schemaVersion: 1, mode: "record-only" },
@@ -3807,8 +4684,26 @@ function buildSupervisedHybridDag(standard, sources) {
3807
4684
  cloneTask(scoutTests),
3808
4685
  cloneTask(plan),
3809
4686
  buildWriteSetAuditNode(sources),
4687
+ buildWriteSetAuditFormatRepairNode(sources, {
4688
+ id: "write-set-audit-format-repair-pi",
4689
+ auditNodeId: "write-set-audit-pi",
4690
+ }),
4691
+ buildWriteSetFormatGateNode(sources),
4692
+ buildPlanRevisionNode(sources),
4693
+ buildFinalWriteSetAuditNode(sources),
4694
+ buildWriteSetAuditFormatRepairNode(sources, {
4695
+ id: "final-write-set-audit-format-repair-pi",
4696
+ auditNodeId: "final-write-set-audit-pi",
4697
+ }),
3810
4698
  buildWriteSetGateNode(sources),
3811
- cloneTask(implement, { depends_on: ["write-set-gate-shell"] }),
4699
+ cloneTask(implement, {
4700
+ depends_on: [
4701
+ "write-set-gate-shell",
4702
+ "plan-pi",
4703
+ "plan-revision-pi",
4704
+ "final-write-set-audit-format-repair-pi",
4705
+ ],
4706
+ }),
3812
4707
  buildSoftVerifyNode(sources),
3813
4708
  buildProcessSupervisorNode(sources),
3814
4709
  buildProcessGateNode(sources),