@tea-agent/loop-agent 0.39.0-beta.1 → 0.39.0-beta.10

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 (331) hide show
  1. package/AGENTS.md +4 -1
  2. package/CHANGELOG.md +248 -95
  3. package/README.md +9 -5
  4. package/bin/loop-agent.js +7 -3
  5. package/dist/application/task-lifecycle/advance.js +12 -0
  6. package/dist/application/task-lifecycle/recommendations.js +13 -2
  7. package/dist/build-stamp.json +3 -3
  8. package/dist/cli/command-definitions.js +23 -5
  9. package/dist/cli/program.js +21 -2
  10. package/dist/cli/update/notifier.js +2 -2
  11. package/dist/cli/update/runtime-activity.js +1 -29
  12. package/dist/cli.js +2 -1
  13. package/dist/commands/client-recovery.js +3 -0
  14. package/dist/commands/dag-follow-up.js +138 -0
  15. package/dist/commands/dag-rerun.js +55 -1
  16. package/dist/commands/init-model-catalog.js +464 -0
  17. package/dist/commands/init-upgrade.js +265 -97
  18. package/dist/commands/init.js +455 -28
  19. package/dist/commands/inspect-next.js +8 -0
  20. package/dist/commands/task-advance.js +19 -0
  21. package/dist/executors/dag-pi-executor.js +2306 -61
  22. package/dist/executors/pi-executor.js +7 -0
  23. package/dist/executors/pi-extension-resolver.js +233 -0
  24. package/dist/executors/pi-playwright-cli-tool.js +74 -28
  25. package/dist/executors/pi-read-budget-policy.js +239 -0
  26. package/dist/executors/pi-sdk-executor.js +180 -33
  27. package/dist/executors/shell-executor.js +1249 -224
  28. package/dist/executors/shell-write-guard.js +14 -0
  29. package/dist/executors/workspace-write-snapshot.js +68 -0
  30. package/dist/infrastructure/harness/atomic-write.js +23 -0
  31. package/dist/shared/dag-failure-category.js +150 -0
  32. package/dist/shared/dag-prompt-override.js +27 -0
  33. package/dist/shared/openspec-spec.js +70 -4
  34. package/dist/shared/operator/capabilities.js +39 -0
  35. package/dist/shared/playwright-cli-command-policy.js +15 -0
  36. package/dist/shared/update/console-notifier.js +100 -0
  37. package/dist/{cli → shared}/update/npm-client.js +70 -15
  38. package/dist/{cli → shared}/update/state.js +41 -13
  39. package/dist/task/config-types.js +50 -5
  40. package/dist/task/contract/apply.js +11 -1
  41. package/dist/task/contract/constants.js +2 -0
  42. package/dist/task/contract/hash.js +3 -0
  43. package/dist/task/contract/observe.js +25 -4
  44. package/dist/task/contract/paths.js +2 -1
  45. package/dist/task/contract/project.js +18 -1
  46. package/dist/task/contract/schema.js +4 -1
  47. package/dist/task/contract/transaction.js +12 -0
  48. package/dist/task/frontend-project-capability.js +199 -18
  49. package/dist/task/source-prepare/completeness.js +188 -0
  50. package/dist/task/source-prepare/fragment-inventory.js +477 -0
  51. package/dist/task/source-prepare/index.js +3 -0
  52. package/dist/task/source-prepare/ledger-reconciliation.js +127 -0
  53. package/dist/task/source-prepare/ledger-review.js +214 -0
  54. package/dist/task/source-prepare/ledger.js +545 -0
  55. package/dist/task/source-prepare/parse-intent.js +24 -4
  56. package/dist/task/source-prepare/prepare.js +298 -1
  57. package/dist/task/source-prepare/semantic-intake.js +25 -5
  58. package/dist/task/source-prepare/source-fidelity-pi.js +384 -0
  59. package/dist/task/source-prepare/types.js +22 -0
  60. package/dist/task/source-references.js +22 -1
  61. package/dist/worker/cli.js +18 -2
  62. package/dist/worker/console/chat/chat-event-store.js +62 -4
  63. package/dist/worker/console/chat/context-panel.js +7 -0
  64. package/dist/worker/console/chat/deferred-turn.js +12 -0
  65. package/dist/worker/console/chat/operation-card.js +8 -1
  66. package/dist/worker/console/chat/pi-console-config.js +81 -10
  67. package/dist/worker/console/chat/pi-runtime.js +264 -16
  68. package/dist/worker/console/chat/provider-error.js +98 -0
  69. package/dist/worker/console/chat/routes.js +298 -32
  70. package/dist/worker/console/chat/session-stats.js +179 -0
  71. package/dist/worker/console/chat/session-store.js +93 -63
  72. package/dist/worker/console/chat/todos.js +115 -0
  73. package/dist/worker/console/chat/tool-preview.js +42 -0
  74. package/dist/worker/console/chat/turn-process.js +20 -55
  75. package/dist/worker/console/chat/user-questions.js +266 -0
  76. package/dist/worker/console/chat/workspace-landing.js +14 -0
  77. package/dist/worker/console/console-handoff.js +82 -0
  78. package/dist/worker/console/console-update-and-init.js +115 -0
  79. package/dist/worker/console/console-update-runtime.js +84 -0
  80. package/dist/worker/console/dag-execution-receipt.js +33 -0
  81. package/dist/worker/console/draft-store.js +49 -0
  82. package/dist/worker/console/frontend-human-decision-adapter.js +19 -0
  83. package/dist/worker/console/frontend-split-operation-adapter.js +20 -0
  84. package/dist/worker/console/index.js +3 -0
  85. package/dist/worker/console/inspect-split.js +18 -0
  86. package/dist/worker/console/native-directory-picker.js +13 -0
  87. package/dist/worker/console/operation-run-facts.js +19 -3
  88. package/dist/worker/console/operation-store.js +227 -7
  89. package/dist/worker/console/operator-actions.js +388 -0
  90. package/dist/worker/console/operator-surface-health.js +1 -0
  91. package/dist/worker/console/operator-user-error.js +2 -2
  92. package/dist/worker/console/prd-intake-bridge.js +13 -0
  93. package/dist/worker/console/routes.js +355 -3
  94. package/dist/worker/console/security.js +67 -13
  95. package/dist/worker/console/server.js +609 -164
  96. package/dist/worker/console/static/assets/{abnfDiagram-N423BO3Z-C4kVYweA.js → abnfDiagram-N423BO3Z-BAsjg7ew.js} +1 -1
  97. package/dist/worker/console/static/assets/{arc-Bj2M8iO5.js → arc-CyIlrwZw.js} +1 -1
  98. package/dist/worker/console/static/assets/{architectureDiagram-T3A2C74G-CLlXe4-6.js → architectureDiagram-T3A2C74G-Bo4tobhH.js} +1 -1
  99. package/dist/worker/console/static/assets/{blockDiagram-VBNYF7ZC-psEp5xH0.js → blockDiagram-VBNYF7ZC-Dfpow9gu.js} +1 -1
  100. package/dist/worker/console/static/assets/{c4Diagram-5PPSVZJV-DeKcOCGJ.js → c4Diagram-5PPSVZJV-DPmzG_B4.js} +1 -1
  101. package/dist/worker/console/static/assets/channel-DSrJAn0-.js +1 -0
  102. package/dist/worker/console/static/assets/{chunk-2GRJ4B5K-K0C744p1.js → chunk-2GRJ4B5K-pKQpB1UW.js} +1 -1
  103. package/dist/worker/console/static/assets/{chunk-2Q5K7J3B-xbNw4J9-.js → chunk-2Q5K7J3B-5n9YYquZ.js} +1 -1
  104. package/dist/worker/console/static/assets/{chunk-5RXB4S5H-DjLaSDUO.js → chunk-5RXB4S5H-DP3CtA2a.js} +1 -1
  105. package/dist/worker/console/static/assets/{chunk-5VM5RSS4-CtiPlKel.js → chunk-5VM5RSS4-DYxMJPUy.js} +1 -1
  106. package/dist/worker/console/static/assets/{chunk-6Q2QTUOP-DZtolir7.js → chunk-6Q2QTUOP-CS3D50Io.js} +1 -1
  107. package/dist/worker/console/static/assets/{chunk-GF5L2VYU-CPcdNeew.js → chunk-GF5L2VYU-gSdDfo5_.js} +1 -1
  108. package/dist/worker/console/static/assets/{chunk-JWPE2WC7-NgHc6V37.js → chunk-JWPE2WC7-A48q0ZNt.js} +1 -1
  109. package/dist/worker/console/static/assets/{chunk-KBJHAD2P-DQ_T-hg8.js → chunk-KBJHAD2P-ow8Mm3DY.js} +1 -1
  110. package/dist/worker/console/static/assets/{chunk-RYQCIY6F-DgLsxcVP.js → chunk-RYQCIY6F-CMKfLWYE.js} +1 -1
  111. package/dist/worker/console/static/assets/{chunk-XXDRQBXY-UrVoM9z1.js → chunk-XXDRQBXY-BiTBdQPX.js} +1 -1
  112. package/dist/worker/console/static/assets/classDiagram-JCYQIIEL-DeMmPl0_.js +1 -0
  113. package/dist/worker/console/static/assets/classDiagram-v2-OCEON4UE-DeMmPl0_.js +1 -0
  114. package/dist/worker/console/static/assets/{cose-bilkent-JH36ORCC-CLkYvTb3.js → cose-bilkent-JH36ORCC-Cm7exRz_.js} +1 -1
  115. package/dist/worker/console/static/assets/{cynefin-VYW2F7L2-BwT_xKrE.js → cynefin-VYW2F7L2-D9O5qGOu.js} +1 -1
  116. package/dist/worker/console/static/assets/{cynefinDiagram-MW4NZA55-Bi9tuzif.js → cynefinDiagram-MW4NZA55-CkvQnako.js} +1 -1
  117. package/dist/worker/console/static/assets/{dagre-VZM6K2ZE-BCIqkWBV.js → dagre-VZM6K2ZE-B1rzp7CB.js} +1 -1
  118. package/dist/worker/console/static/assets/{diagram-7IWD3JNH-Dip6l9_Z.js → diagram-7IWD3JNH-BP07Ipcc.js} +1 -1
  119. package/dist/worker/console/static/assets/{diagram-B4RE2ZJO-B-xkh_wK.js → diagram-B4RE2ZJO-BlPzweQb.js} +1 -1
  120. package/dist/worker/console/static/assets/{diagram-LBJQPF4R-DZO0kTt_.js → diagram-LBJQPF4R-CxPa7n0S.js} +1 -1
  121. package/dist/worker/console/static/assets/{diagram-Q27KOJAE-D6ZplNbZ.js → diagram-Q27KOJAE-CMTZs53A.js} +1 -1
  122. package/dist/worker/console/static/assets/{diagram-UB23O5K3-BPekbhoS.js → diagram-UB23O5K3-CxG_X52r.js} +1 -1
  123. package/dist/worker/console/static/assets/{ebnfDiagram-BXEA7PRR-BtmaJpK5.js → ebnfDiagram-BXEA7PRR-D8axtJM0.js} +1 -1
  124. package/dist/worker/console/static/assets/{erDiagram-JOGREHBK-6DS0Xc44.js → erDiagram-JOGREHBK-DIsGzn7b.js} +1 -1
  125. package/dist/worker/console/static/assets/{flowDiagram-UKHOOZJN-CihnROSm.js → flowDiagram-UKHOOZJN-CU9zkDG0.js} +1 -1
  126. package/dist/worker/console/static/assets/{ganttDiagram-PKOTCBZU-CZC4zOE2.js → ganttDiagram-PKOTCBZU-DCjTmz5P.js} +1 -1
  127. package/dist/worker/console/static/assets/{gitGraphDiagram-DS77QQ5N-Dq1fhzGN.js → gitGraphDiagram-DS77QQ5N-BksdmFzg.js} +1 -1
  128. package/dist/worker/console/static/assets/index-B-O5ChYq.css +1 -0
  129. package/dist/worker/console/static/assets/index-CfnY8AQx.js +407 -0
  130. package/dist/worker/console/static/assets/{infoDiagram-6WML65LV-DlSl4BGx.js → infoDiagram-6WML65LV-CGzzzJtL.js} +1 -1
  131. package/dist/worker/console/static/assets/{ishikawaDiagram-WSZJBQD7-snnzFl_U.js → ishikawaDiagram-WSZJBQD7-BdseB80N.js} +1 -1
  132. package/dist/worker/console/static/assets/{journeyDiagram-NVQOT4AX-BDE7qpMx.js → journeyDiagram-NVQOT4AX-DM9gsgha.js} +1 -1
  133. package/dist/worker/console/static/assets/{kanban-definition-27J2QSJJ-CD2Ci04G.js → kanban-definition-27J2QSJJ-3q-_9R2b.js} +1 -1
  134. package/dist/worker/console/static/assets/{linear-DdnapKIH.js → linear-B8N-Mxsd.js} +1 -1
  135. package/dist/worker/console/static/assets/{mermaid.core-BVjAT9b8.js → mermaid.core-EPIHUUsu.js} +5 -5
  136. package/dist/worker/console/static/assets/{mindmap-definition-FAOFIHXS-D0yJk-zA.js → mindmap-definition-FAOFIHXS-CxwO12aa.js} +1 -1
  137. package/dist/worker/console/static/assets/{pegDiagram-VL7TDLO6-BVo9tFP-.js → pegDiagram-VL7TDLO6-bPqbVX6Y.js} +1 -1
  138. package/dist/worker/console/static/assets/{pieDiagram-7S7Q4E2Y-DnOV-mZ_.js → pieDiagram-7S7Q4E2Y-KzqIvzeX.js} +1 -1
  139. package/dist/worker/console/static/assets/{quadrantDiagram-CIZ2JOQS-CzUJo58i.js → quadrantDiagram-CIZ2JOQS-CyofpWiV.js} +1 -1
  140. package/dist/worker/console/static/assets/{railroadDiagram-AXF67PYL-Bvikrolh.js → railroadDiagram-AXF67PYL-cTWJvLTP.js} +1 -1
  141. package/dist/worker/console/static/assets/{requirementDiagram-LRYGKXZP-DtaSVCap.js → requirementDiagram-LRYGKXZP-BK--q27B.js} +1 -1
  142. package/dist/worker/console/static/assets/{sankeyDiagram-W5VNT64P-C2c9A0wW.js → sankeyDiagram-W5VNT64P-C0SDOHXK.js} +1 -1
  143. package/dist/worker/console/static/assets/{sequenceDiagram-SI44F4Z6-DmXJcU7r.js → sequenceDiagram-SI44F4Z6-Cong6YIo.js} +1 -1
  144. package/dist/worker/console/static/assets/{sizeCapture-X5ZJPWSS-B4GUFW92.js → sizeCapture-X5ZJPWSS-DUVkgZx8.js} +1 -1
  145. package/dist/worker/console/static/assets/{stateDiagram-OKZ733FA-Dsad2MXf.js → stateDiagram-OKZ733FA-C7XgxuZg.js} +1 -1
  146. package/dist/worker/console/static/assets/stateDiagram-v2-UEYNNEHI-Dzbnd5oX.js +1 -0
  147. package/dist/worker/console/static/assets/{swimlanes-SLNWSIFB-DGq48Fbi.js → swimlanes-SLNWSIFB-D9Ayvwcy.js} +2 -2
  148. package/dist/worker/console/static/assets/swimlanesDiagram-ULZ7WXOC-BY0hHk6G.js +8 -0
  149. package/dist/worker/console/static/assets/{timeline-definition-Z64GVDOM-CSSi6OFf.js → timeline-definition-Z64GVDOM-Byb000cH.js} +1 -1
  150. package/dist/worker/console/static/assets/{vennDiagram-T6HMQDX7-BqyzPrWv.js → vennDiagram-T6HMQDX7-DMLegPt4.js} +1 -1
  151. package/dist/worker/console/static/assets/{wardleyDiagram-T6FBY63Y-DlznVb6S.js → wardleyDiagram-T6FBY63Y-ChMMeOsP.js} +1 -1
  152. package/dist/worker/console/static/assets/{xychartDiagram-ELKLHX3M-CfBcgJ_K.js → xychartDiagram-ELKLHX3M-bEMEPisW.js} +1 -1
  153. package/dist/worker/console/static/index.html +2 -2
  154. package/dist/worker/console/static-src/active-run-badge.js +17 -0
  155. package/dist/worker/console/static-src/app/console-types.js +68 -4
  156. package/dist/worker/console/static-src/app/useConsoleShell.js +30 -12
  157. package/dist/worker/console/static-src/app/useOperatorActions.js +41 -6
  158. package/dist/worker/console/static-src/app/useRecoveryConsole.js +45 -10
  159. package/dist/worker/console/static-src/app/useRunProgress.js +47 -0
  160. package/dist/worker/console/static-src/app/useTaskWizard.js +74 -2
  161. package/dist/worker/console/static-src/night/useNightBoard.js +8 -5
  162. package/dist/worker/console/static-src/operator-chat/cards/failure-category-advice.js +25 -0
  163. package/dist/worker/console/static-src/operator-chat/chat-sse-events.js +140 -8
  164. package/dist/worker/console/static-src/operator-chat/details-dag-actions.js +122 -0
  165. package/dist/worker/console/static-src/operator-chat/input-history.js +76 -0
  166. package/dist/worker/console/static-src/operator-chat/mutation-gate.js +1 -1
  167. package/dist/worker/console/static-src/operator-chat/refs.js +3 -0
  168. package/dist/worker/console/static-src/operator-chat/runtime-selection-labels.js +27 -0
  169. package/dist/worker/console/static-src/operator-chat/turn-submission.js +16 -3
  170. package/dist/worker/console/static-src/operator-chat/useChatSessions.js +168 -72
  171. package/dist/worker/console/static-src/operator-chat/useChatStream.js +102 -21
  172. package/dist/worker/console/static-src/operator-chat/useChatThread.js +184 -4
  173. package/dist/worker/console/static-src/operator-chat/useComposer.js +32 -0
  174. package/dist/worker/console/static-src/operator-chat/useRepoBrowser.js +63 -9
  175. package/dist/worker/console/static-src/operator-chat/useWorkspaceBrowserGate.js +4 -0
  176. package/dist/worker/console/static-src/operator-chat/workspace-layout-mode.js +3 -3
  177. package/dist/worker/console/static-src/pages/tasks/run-id-resolution.js +79 -0
  178. package/dist/worker/console/static-src/pages/tasks/run-ownership-verify.js +23 -0
  179. package/dist/worker/console/static-src/pages/tasks/run-panel-progress.js +110 -0
  180. package/dist/worker/console/static-src/shell/useWorkspaces.js +299 -0
  181. package/dist/worker/console/static-src/shell/workspace-route.js +337 -0
  182. package/dist/worker/console/workspace-context.js +344 -0
  183. package/dist/worker/console/workspace-registry.js +214 -0
  184. package/dist/worker/loop-agent/loop-agent-client.js +7 -3
  185. package/dist/worker/materialize/frontend-split-task-materializer.js +72 -0
  186. package/dist/worker/materialize/harness-task-lineage.js +5 -2
  187. package/dist/worker/observability/init-runtime-activity.js +26 -0
  188. package/dist/worker/observability/read-model.js +22 -0
  189. package/dist/worker/observe/node-input.js +160 -3
  190. package/dist/worker/observe/node-process.js +377 -0
  191. package/dist/worker/observe/routes.js +224 -3
  192. package/dist/worker/observe/server.js +4 -0
  193. package/dist/worker/observe/shell-handler-keys.js +34 -0
  194. package/dist/worker/observe/static/api.js +90 -4
  195. package/dist/worker/observe/static/app.js +18 -70
  196. package/dist/worker/observe/static/constants.js +8 -1
  197. package/dist/worker/observe/static/dag-history-labels.js +4 -0
  198. package/dist/worker/observe/static/dag-node-purpose.d.ts +6 -0
  199. package/dist/worker/observe/static/dag-node-purpose.js +208 -0
  200. package/dist/worker/observe/static/format.js +60 -0
  201. package/dist/worker/observe/static/index.html +6 -1
  202. package/dist/worker/observe/static/inspect-workspace.js +307 -0
  203. package/dist/worker/observe/static/markdown-render.js +20 -1
  204. package/dist/worker/observe/static/operator-chrome.css +145 -20
  205. package/dist/worker/observe/static/operator-chrome.d.ts +20 -3
  206. package/dist/worker/observe/static/operator-chrome.js +330 -96
  207. package/dist/worker/observe/static/prompt-restart-candidates.d.ts +18 -0
  208. package/dist/worker/observe/static/prompt-restart-candidates.js +36 -0
  209. package/dist/worker/observe/static/router.d.ts +46 -0
  210. package/dist/worker/observe/static/router.js +25 -1
  211. package/dist/worker/observe/static/state.d.ts +50 -0
  212. package/dist/worker/observe/static/state.js +184 -1
  213. package/dist/worker/observe/static/styles.css +395 -0
  214. package/dist/worker/observe/static/views/dag-graph.js +4 -2
  215. package/dist/worker/observe/static/views/dag-inspector.js +1067 -9
  216. package/dist/worker/observe/static/views/dag.d.ts +13 -0
  217. package/dist/worker/observe/static/views/dag.js +14 -5
  218. package/dist/worker/observe/static/views/dashboard.js +6 -4
  219. package/dist/worker/observe/static/views/failures.js +5 -2
  220. package/dist/worker/observe/static/views/night.js +2 -2
  221. package/dist/worker/observe/static/views/pool.js +1 -0
  222. package/dist/worker/observe/static/views/run.js +3 -1
  223. package/dist/worker/observe/static/views/session-timeline.js +78 -9
  224. package/dist/worker/observe/static/views/task.js +56 -1
  225. package/dist/workflows/dag/backend-test-case-coverage-analysis.js +49 -3
  226. package/dist/workflows/dag/backend-test-markdown-workflow.js +176 -1
  227. package/dist/workflows/dag/backend-test-pytest-collection.js +98 -14
  228. package/dist/workflows/dag/backend-test-writer-completeness.js +10 -5
  229. package/dist/workflows/dag/contract-output-registry.js +15 -0
  230. package/dist/workflows/dag/contract-validator-registrations.js +1 -2
  231. package/dist/workflows/dag/dynamic-runtime/loop-until.js +1 -1
  232. package/dist/workflows/dag/dynamic-runtime/map.js +1 -1
  233. package/dist/workflows/dag/failure-category.js +7 -116
  234. package/dist/workflows/dag/frontend-closeout.js +221 -0
  235. package/dist/workflows/dag/frontend-design-policy.js +400 -0
  236. package/dist/workflows/dag/frontend-human-decision.js +182 -0
  237. package/dist/workflows/dag/frontend-implementation-contract.js +538 -166
  238. package/dist/workflows/dag/frontend-plan-render.js +24 -0
  239. package/dist/workflows/dag/frontend-prewrite-gate.js +388 -392
  240. package/dist/workflows/dag/frontend-provider-capability-matrix.js +159 -0
  241. package/dist/workflows/dag/frontend-recovery-capsule.js +455 -0
  242. package/dist/workflows/dag/frontend-recovery-controller.js +202 -0
  243. package/dist/workflows/dag/frontend-recovery-lineage.js +178 -0
  244. package/dist/workflows/dag/frontend-recovery-plan.js +17 -10
  245. package/dist/workflows/dag/frontend-recovery-run.js +33 -20
  246. package/dist/workflows/dag/frontend-repair.js +7 -424
  247. package/dist/workflows/dag/frontend-review-context.js +288 -14
  248. package/dist/workflows/dag/frontend-review-findings.js +270 -0
  249. package/dist/workflows/dag/frontend-risk.js +15 -2
  250. package/dist/workflows/dag/frontend-shadow-dual-write.js +814 -0
  251. package/dist/workflows/dag/frontend-shape-capsule-store.js +191 -0
  252. package/dist/workflows/dag/frontend-shape-facts.js +409 -0
  253. package/dist/workflows/dag/frontend-shape.js +427 -0
  254. package/dist/workflows/dag/frontend-source-fidelity-ledger.js +108 -0
  255. package/dist/workflows/dag/frontend-split-application-service.js +203 -0
  256. package/dist/workflows/dag/frontend-split-orchestrator.js +899 -0
  257. package/dist/workflows/dag/frontend-test-case-checklist.js +30 -4
  258. package/dist/workflows/dag/frontend-test-case-manifest.js +11 -4
  259. package/dist/workflows/dag/frontend-test-case-quality.js +11 -16
  260. package/dist/workflows/dag/frontend-test-environment-probe.js +230 -0
  261. package/dist/workflows/dag/frontend-test-html-report.js +3 -1
  262. package/dist/workflows/dag/frontend-test-l5-report.js +3 -1
  263. package/dist/workflows/dag/frontend-test-layout.js +159 -0
  264. package/dist/workflows/dag/frontend-test-markdown.js +61 -0
  265. package/dist/workflows/dag/frontend-test-result-contract.js +188 -42
  266. package/dist/workflows/dag/frontend-test-standard-scenarios.js +70 -0
  267. package/dist/workflows/dag/frontend-typed-event-store.js +452 -0
  268. package/dist/workflows/dag/frontend-typed-event-transaction.js +180 -0
  269. package/dist/workflows/dag/frontend-verification-trace.js +52 -17
  270. package/dist/workflows/dag/frontend-worktree-diff.js +250 -17
  271. package/dist/workflows/dag/frontend-writer-admission.js +272 -0
  272. package/dist/workflows/dag/frontend-writer-status.js +244 -0
  273. package/dist/workflows/dag/init-hybrid.js +984 -613
  274. package/dist/workflows/dag/interrupt-request.js +10 -3
  275. package/dist/workflows/dag/lifecycle.js +3 -3
  276. package/dist/workflows/dag/node-execution.js +468 -58
  277. package/dist/workflows/dag/prompt.js +44 -19
  278. package/dist/workflows/dag/recovery-lease.js +80 -0
  279. package/dist/workflows/dag/report.js +37 -1
  280. package/dist/workflows/dag/rerun-feedback.js +124 -1
  281. package/dist/workflows/dag/rerun-plan.js +144 -5
  282. package/dist/workflows/dag/rerun-run.js +81 -6
  283. package/dist/workflows/dag/rerun-task.js +29 -0
  284. package/dist/workflows/dag/retry-policy.js +318 -8
  285. package/dist/workflows/dag/runner.js +429 -122
  286. package/dist/workflows/dag/scheduler.js +114 -16
  287. package/dist/workflows/dag/structured-output-repair.js +712 -0
  288. package/dist/workflows/dag/types.js +321 -31
  289. package/dist/workflows/dag/upstream-artifacts.js +0 -4
  290. package/dist/workflows/dag/validate.js +92 -32
  291. package/docs/README.md +2 -0
  292. package/docs/architecture/evolution.md +45 -0
  293. package/docs/init-surface.manifest.json +9 -0
  294. package/docs/operations/README.md +1 -1
  295. package/docs/operations/local-development-environment.md +21 -0
  296. package/docs/templates/README.md +2 -1
  297. package/docs/templates/agent-dag-report.schema.json +8 -2
  298. package/docs/templates/agent-dag.schema.json +45 -1
  299. package/docs/templates/frontend-implementation-contract.schema.json +684 -24
  300. package/docs/templates/frontend-test-case-checklist.md +2 -2
  301. package/docs/templates/frontend-test-dag.json +8 -10
  302. package/docs/templates/frontend-test-dag.retrieve-context.prompt.md +1 -1
  303. package/docs/templates/init-managed-agents.md +13 -3
  304. package/docs/templates/spec-registry.schema.json +45 -0
  305. package/harness.json +3 -3
  306. package/package.json +7 -1
  307. package/scripts/next-info.mjs +356 -0
  308. package/scripts/next-publish-gate.mjs +238 -0
  309. package/scripts/release-source-binding.mjs +251 -0
  310. package/skills/codebase-scout/SKILL.md +1 -1
  311. package/skills/fe-test-ui-scout/SKILL.md +65 -0
  312. package/skills/fe-test-ui-scout/references/ledger-schema.md +62 -0
  313. package/skills/fe-test-ui-scout/references/recon-protocol.md +54 -0
  314. package/skills/frontend-bounded-implement/SKILL.md +8 -14
  315. package/skills/frontend-design-review/SKILL.md +20 -41
  316. package/skills/frontend-implementation/SKILL.md +3 -4
  317. package/skills/frontend-implementation/references/design-spec.md +15 -10
  318. package/skills/frontend-implementation/references/node-contracts.md +23 -13
  319. package/skills/frontend-review/SKILL.md +22 -14
  320. package/skills/frontend-review/references/review-findings.md +6 -7
  321. package/skills/loop-agent/references/command-reference.md +12 -4
  322. package/skills/loop-agent/references/hybrid-dag.md +8 -1
  323. package/skills/playwright-cli/SKILL.md +23 -1
  324. package/skills/playwright-cli-case-generator/SKILL.md +29 -8
  325. package/dist/worker/console/static/assets/channel-CPF4N7pf.js +0 -1
  326. package/dist/worker/console/static/assets/classDiagram-JCYQIIEL-C2DwA_Y1.js +0 -1
  327. package/dist/worker/console/static/assets/classDiagram-v2-OCEON4UE-C2DwA_Y1.js +0 -1
  328. package/dist/worker/console/static/assets/index-DTZOKgAn.js +0 -404
  329. package/dist/worker/console/static/assets/index-Ya5FE7cD.css +0 -1
  330. package/dist/worker/console/static/assets/stateDiagram-v2-UEYNNEHI-B8FB_cNk.js +0 -1
  331. package/dist/worker/console/static/assets/swimlanesDiagram-ULZ7WXOC-CrZYaWfx.js +0 -8
@@ -1012,6 +1012,13 @@ export function classifyPiFailure(input) {
1012
1012
  if (input.exitCode === 0 && input.assistantText.trim())
1013
1013
  return "success";
1014
1014
  const combined = `${input.stderr}\n${input.stdout}`.toLowerCase();
1015
+ // Context overflow (400 request-too-large / context window exceeded).
1016
+ // Must be detected BEFORE the generic rate-limit/quota phrases: an overflow
1017
+ // error often contains "too many tokens" / "maximum context" which would
1018
+ // otherwise be misread as rate-limit or quota. Kept in sync with
1019
+ // CONTEXT_OVERFLOW_PHRASES in commands/client-recovery.ts.
1020
+ if (/请求上下文过大|context_length_exceeded|context overflow|context length overflow|too many tokens|maximum context|prompt is too long|request_too_large|context window/.test(combined))
1021
+ return "context-overflow";
1015
1022
  // Prefer explicit rate-limit / HTTP 429 before generic "limit" quota phrases.
1016
1023
  // Chinese gateways often say "请求频率已达到限制" without English "rate limit".
1017
1024
  if (/rate.?limit|too many requests|\b429\b|请求频率|频率.*限制|rpm|tpm/.test(combined))
@@ -0,0 +1,233 @@
1
+ import { existsSync, readdirSync, readFileSync, statSync } from "node:fs";
2
+ import os from "node:os";
3
+ import path from "node:path";
4
+ /**
5
+ * Pi extension allowlist for backend implementation DAGs (plan D2/D3).
6
+ * DAG specs carry these frozen short ids; the executor resolves them against
7
+ * the Pi agent settings package inventory at run time.
8
+ */
9
+ export const DAG_PI_EXTENSION_IDS = ["pi-codegraph", "pi-lens"];
10
+ /**
11
+ * Alias table: one DAG short id may resolve to multiple published package
12
+ * names. The first settings spec whose de-`npm:`-prefixed package name is in
13
+ * the alias set wins (project settings override user settings).
14
+ */
15
+ export const DAG_PI_EXTENSION_PACKAGE_ALIASES = {
16
+ "pi-codegraph": [
17
+ "@estebanforge/pi-codegraph-enhanced",
18
+ "@vndv/pi-codegraph",
19
+ "pi-codegraph",
20
+ ],
21
+ "pi-lens": ["pi-lens"],
22
+ };
23
+ /**
24
+ * Extension tools that must never be activated as extras on any node (D5):
25
+ * they mutate files without the writer path policy. Writer nodes keep only
26
+ * policy-wrapped edit/write; read nodes keep only navigation tools.
27
+ */
28
+ export const DAG_PI_EXTENSION_WRITE_TOOL_DENY = new Set([
29
+ "ast_grep_replace",
30
+ ]);
31
+ /** Builtin tool names extensions must never replace (D5). */
32
+ export const DAG_PI_EXTENSION_PROTECTED_BUILTIN_TOOLS = new Set(["read", "edit", "write", "bash", "grep", "find", "ls"]);
33
+ /** Deny surface extensions must never activate (D5). */
34
+ export const DAG_PI_EXTENSION_DENIED_TOOL_IDS = new Set([
35
+ "apply_patch",
36
+ "apply-patch",
37
+ "full-tools",
38
+ "full_tools",
39
+ "coding-chat",
40
+ "coding_chat",
41
+ "shell",
42
+ ]);
43
+ export function isDagPiExtensionId(value) {
44
+ return DAG_PI_EXTENSION_IDS.includes(value);
45
+ }
46
+ function getPiAgentDir() {
47
+ const envDir = process.env.PI_CODING_AGENT_DIR;
48
+ if (envDir)
49
+ return envDir;
50
+ return path.join(os.homedir(), ".pi", "agent");
51
+ }
52
+ function readSettingsPackages(settingsPath) {
53
+ try {
54
+ if (!existsSync(settingsPath))
55
+ return undefined;
56
+ const parsed = JSON.parse(readFileSync(settingsPath, "utf8"));
57
+ if (!Array.isArray(parsed.packages))
58
+ return undefined;
59
+ return parsed.packages.filter((spec) => typeof spec === "string");
60
+ }
61
+ catch {
62
+ return undefined;
63
+ }
64
+ }
65
+ /** Strip an `npm:` prefix from a settings package spec, e.g. `npm:pi-lens`. */
66
+ export function packageNameFromSpec(spec) {
67
+ return spec.startsWith("npm:") ? spec.slice(4) : spec;
68
+ }
69
+ function loadPackageInventories(repoRoot) {
70
+ const agentDir = getPiAgentDir();
71
+ const userSpecs = readSettingsPackages(path.join(agentDir, "settings.json"));
72
+ const projectSpecs = readSettingsPackages(path.join(repoRoot, ".pi", "settings.json"));
73
+ const inventories = [];
74
+ if (userSpecs) {
75
+ inventories.push({
76
+ specs: userSpecs,
77
+ nodeModulesRoot: path.join(agentDir, "npm", "node_modules"),
78
+ });
79
+ }
80
+ if (projectSpecs) {
81
+ inventories.push({
82
+ specs: projectSpecs,
83
+ nodeModulesRoot: path.join(repoRoot, "node_modules"),
84
+ });
85
+ }
86
+ return inventories;
87
+ }
88
+ const EXTENSION_FILE_RE = /\.(?:ts|js|mjs|cjs)$/;
89
+ /**
90
+ * Resolve a package's `pi.extensions` entries to absolute entry paths.
91
+ * - A file entry maps to itself.
92
+ * - A directory entry expands to every extension file directly inside it
93
+ * (e.g. `@estebanforge/pi-codegraph-enhanced` ships `./extensions` holding
94
+ * `codegraph.ts`).
95
+ * No `pi.extensions` declaration, or a declaration whose paths do not exist,
96
+ * yields no entries (caller records `no-entry`) — no heuristic scanning
97
+ * outside declared paths.
98
+ */
99
+ export function resolvePackageEntryPaths(packageDir) {
100
+ let manifest;
101
+ try {
102
+ manifest = JSON.parse(readFileSync(path.join(packageDir, "package.json"), "utf8"));
103
+ }
104
+ catch {
105
+ return { entryPaths: [], reason: "no-entry" };
106
+ }
107
+ const declared = manifest.pi?.extensions;
108
+ if (!Array.isArray(declared)) {
109
+ return { entryPaths: [], reason: "no-entry" };
110
+ }
111
+ const entryPaths = [];
112
+ for (const entry of declared) {
113
+ if (typeof entry !== "string" || entry.length === 0)
114
+ continue;
115
+ const resolved = path.resolve(packageDir, entry);
116
+ if (!existsSync(resolved))
117
+ continue;
118
+ let isDirectory = false;
119
+ try {
120
+ isDirectory = statSync(resolved).isDirectory();
121
+ }
122
+ catch {
123
+ continue;
124
+ }
125
+ if (!isDirectory) {
126
+ entryPaths.push(resolved);
127
+ continue;
128
+ }
129
+ let names;
130
+ try {
131
+ names = readdirSync(resolved);
132
+ }
133
+ catch {
134
+ continue;
135
+ }
136
+ for (const name of names.sort()) {
137
+ if (!EXTENSION_FILE_RE.test(name))
138
+ continue;
139
+ entryPaths.push(path.join(resolved, name));
140
+ }
141
+ }
142
+ if (entryPaths.length === 0) {
143
+ return { entryPaths: [], reason: "no-entry" };
144
+ }
145
+ return { entryPaths };
146
+ }
147
+ /**
148
+ * Resolve frozen DAG short ids against the Pi settings package inventory
149
+ * (user `~/.pi/agent/settings.json`, then project `.pi/settings.json` — a
150
+ * later match on the same alias wins). Pure synchronous read; never throws.
151
+ */
152
+ export function resolveDagPiExtensions(input) {
153
+ const inventories = loadPackageInventories(input.repoRoot);
154
+ const resolved = [];
155
+ const missing = [];
156
+ for (const id of input.ids) {
157
+ if (!isDagPiExtensionId(id)) {
158
+ missing.push({
159
+ id,
160
+ reason: "package-unreadable",
161
+ detail: `unknown pi extension id: ${id}`,
162
+ });
163
+ continue;
164
+ }
165
+ const aliases = DAG_PI_EXTENSION_PACKAGE_ALIASES[id];
166
+ let matched;
167
+ // Later inventories (project) override earlier (user).
168
+ for (const inventory of inventories) {
169
+ for (const spec of inventory.specs) {
170
+ const packageName = packageNameFromSpec(spec);
171
+ if (!aliases.includes(packageName))
172
+ continue;
173
+ const packageDir = path.join(inventory.nodeModulesRoot, ...packageName.split("/"));
174
+ if (!existsSync(packageDir))
175
+ continue;
176
+ try {
177
+ const manifest = JSON.parse(readFileSync(path.join(packageDir, "package.json"), "utf8"));
178
+ if (typeof manifest.version !== "string")
179
+ continue;
180
+ matched = { packageDir, packageName, version: manifest.version };
181
+ }
182
+ catch {
183
+ // unreadable manifest: skip this candidate
184
+ }
185
+ }
186
+ }
187
+ if (!matched) {
188
+ missing.push({ id, reason: "not-installed" });
189
+ continue;
190
+ }
191
+ const entries = resolvePackageEntryPaths(matched.packageDir);
192
+ if (entries.entryPaths.length === 0) {
193
+ missing.push({
194
+ id,
195
+ reason: entries.reason === "no-entry" ? "no-entry" : "package-unreadable",
196
+ });
197
+ continue;
198
+ }
199
+ resolved.push({
200
+ id,
201
+ packageName: matched.packageName,
202
+ version: matched.version,
203
+ entryPaths: entries.entryPaths,
204
+ });
205
+ }
206
+ return { resolved, missing };
207
+ }
208
+ /**
209
+ * Compute extension tool names that may join a node's active tool allowlist
210
+ * after session creation (D5): new tool names registered by the loaded
211
+ * extensions, minus protected builtins, the deny surface, and write-capable
212
+ * extension tools (both buckets — writer writes stay policy-wrapped only).
213
+ */
214
+ export function computeExtensionToolAllowlistExtras(sessionTools, baseline) {
215
+ const baselineLower = new Set(baseline.map((name) => name.toLowerCase()));
216
+ const extras = [];
217
+ for (const tool of sessionTools) {
218
+ const name = tool && typeof tool.name === "string" ? tool.name : undefined;
219
+ if (!name)
220
+ continue;
221
+ const lower = name.toLowerCase();
222
+ if (baselineLower.has(lower))
223
+ continue;
224
+ if (DAG_PI_EXTENSION_PROTECTED_BUILTIN_TOOLS.has(lower))
225
+ continue;
226
+ if (DAG_PI_EXTENSION_DENIED_TOOL_IDS.has(lower))
227
+ continue;
228
+ if (DAG_PI_EXTENSION_WRITE_TOOL_DENY.has(lower))
229
+ continue;
230
+ extras.push(name);
231
+ }
232
+ return extras;
233
+ }
@@ -4,7 +4,8 @@ import { appendFile, lstat, mkdir, readFile, realpath, stat, } from "node:fs/pro
4
4
  import path from "node:path";
5
5
  import { processTreeSpawnOptions, terminateProcessTree, } from "./process-tree.js";
6
6
  import { resolvePlaywrightCliLauncher } from "./playwright-cli-launcher.js";
7
- import { PLAYWRIGHT_CLI_ALLOWED_COMMANDS, isPlaywrightCliCommand, } from "../shared/playwright-cli-command-policy.js";
7
+ import { PLAYWRIGHT_CLI_ALLOWED_COMMANDS, isPlaywrightCliCommand, parsePlaywrightCliFindMatchCount, } from "../shared/playwright-cli-command-policy.js";
8
+ import { defaultFrontendTestLayout, escapeRegexLiteral, } from "../workflows/dag/frontend-test-layout.js";
8
9
  export { PLAYWRIGHT_CLI_ALLOWED_COMMANDS, isPlaywrightCliCommand, };
9
10
  /** Logical capability name only; execution always uses the verified JS launcher. */
10
11
  export const PLAYWRIGHT_CLI_EXECUTABLE = "playwright-cli";
@@ -111,6 +112,13 @@ export async function readPlaywrightCliReceipts(runDir, nodeId) {
111
112
  }
112
113
  }
113
114
  export function summarizePlaywrightCliReceipts(receipts) {
115
+ // Sequence numbers are assigned at dispatch time, which is the controller's
116
+ // causal order. Receipt lines are appended when each command COMPLETES, so
117
+ // concurrent commands (e.g. parallel find calls issued in one turn) persist
118
+ // out of sequence order. Judge chain contiguity on the dispatch order by
119
+ // sorting on `sequence` first; a genuinely missing receipt (gap) or a
120
+ // duplicated sequence still fails the contiguity check below.
121
+ const orderedReceipts = [...receipts].sort((a, b) => a.sequence - b.sequence);
114
122
  let hasSuccessfulOpen = false;
115
123
  let hasInteractionOrAssertion = false;
116
124
  let hasMeaningfulAssertion = false;
@@ -122,7 +130,7 @@ export function summarizePlaywrightCliReceipts(receipts) {
122
130
  let hasContiguousControllerSequence = true;
123
131
  let previousSequence = 0;
124
132
  const caseIds = new Set();
125
- for (const receipt of receipts) {
133
+ for (const receipt of orderedReceipts) {
126
134
  caseIds.add(receipt.caseId);
127
135
  const succeeded = receipt.exitCode === 0 && !receipt.timedOut;
128
136
  const ordered = receipt.sequence === previousSequence + 1;
@@ -137,8 +145,10 @@ export function summarizePlaywrightCliReceipts(receipts) {
137
145
  if (PLAYWRIGHT_CLI_INTERACTION_COMMANDS.has(receipt.command) && succeeded) {
138
146
  hasInteractionOrAssertion = true;
139
147
  }
140
- // A successful find is the only model-visible semantic assertion in v1.
141
- if (receipt.command === "find" && succeeded) {
148
+ // A successful find (or find-unique, its semantic wrapper) is the only
149
+ // model-visible semantic assertion in v1.
150
+ if ((receipt.command === "find" || receipt.command === "find-unique") &&
151
+ succeeded) {
142
152
  hasMeaningfulAssertion = true;
143
153
  if (ordered && sawOpen)
144
154
  sawMeaningfulAssertionAfterOpen = true;
@@ -258,7 +268,8 @@ function validateOpenArgs(args, baseUrl) {
258
268
  }
259
269
  function resolveEvidenceRelative(rawPath, ctx) {
260
270
  const evidenceDir = normalizePosix(ctx.evidenceDir).replace(/\/$/, "");
261
- const evidencePrefix = `testcase/frontend/evidence/${ctx.caseId}`;
271
+ const layout = ctx.layout ?? defaultFrontendTestLayout();
272
+ const evidencePrefix = `${layout.evidenceDir}/${ctx.caseId}`;
262
273
  if (!(evidenceDir === evidencePrefix ||
263
274
  evidenceDir.startsWith(`${evidencePrefix}/`))) {
264
275
  throw new PlaywrightCliPolicyError("evidence-dir-invalid", `evidenceDir must be under ${evidencePrefix}`);
@@ -282,7 +293,7 @@ function resolveEvidenceRelative(rawPath, ctx) {
282
293
  if (candidate.startsWith(`${evidenceDir}/`) || candidate === evidenceDir) {
283
294
  return candidate;
284
295
  }
285
- if (candidate.startsWith("testcase/frontend/evidence/") &&
296
+ if (candidate.startsWith(`${layout.evidenceDir}/`) &&
286
297
  !candidate.startsWith(`${evidencePrefix}/`)) {
287
298
  throw new PlaywrightCliPolicyError("path-cross-case", "output path escapes current case evidenceDir");
288
299
  }
@@ -463,11 +474,12 @@ function assertInputPathAllowed(rawPath, ctx) {
463
474
  if (!isSafeRelativePosix(candidate)) {
464
475
  throw new PlaywrightCliPolicyError("input-path-unsafe", `unsafe input path: ${rawPath}`);
465
476
  }
477
+ const layout = ctx.layout ?? defaultFrontendTestLayout();
466
478
  const allowedPrefixes = [
467
- `testcase/frontend/evidence/${ctx.caseId}/`,
468
- "testcase/frontend/cases/",
469
- "testcase/frontend/rag/",
470
- "testcase/frontend/fixtures/",
479
+ `${layout.evidenceDir}/${ctx.caseId}/`,
480
+ `${layout.casesDir}/`,
481
+ `${layout.ragDir}/`,
482
+ `${layout.fixturesDir}/`,
471
483
  ];
472
484
  if (!allowedPrefixes.some((prefix) => candidate === prefix.slice(0, -1) || candidate.startsWith(prefix)) &&
473
485
  !candidate.startsWith(normalizePosix(ctx.evidenceDir) + "/")) {
@@ -523,6 +535,17 @@ export function preparePlaywrightCliArgv(input, ctx) {
523
535
  else if (command === "pdf" || command === "snapshot") {
524
536
  args = normalizePdfOrSnapshotArgs(command, rawArgs, ctx);
525
537
  }
538
+ else if (command === "find-unique") {
539
+ // find-unique is a controller-side semantic wrapper: the underlying CLI
540
+ // only knows `find`; uniqueness (matchCount === 1) is judged by the
541
+ // executor after parsing the find output. Require exactly one text
542
+ // literal — a unique-control assertion is meaningless with multiple.
543
+ const literals = rawArgs.filter((arg) => !arg.startsWith("-"));
544
+ if (rawArgs.length !== 1 || literals.length !== 1) {
545
+ throw new PlaywrightCliPolicyError("find-unique-args", "find-unique requires exactly one text literal and no flags");
546
+ }
547
+ args = literals;
548
+ }
526
549
  else {
527
550
  args = rewriteArgsForPaths(command, rawArgs, ctx);
528
551
  }
@@ -609,8 +632,8 @@ async function validatePlaywrightCliOutputPath(command, args, ctx) {
609
632
  if (!outputFilename)
610
633
  return;
611
634
  const repoRoot = path.resolve(ctx.repoRoot);
612
- const caseEvidenceRoot = path.resolve(repoRoot, "testcase", "frontend", "evidence", ctx.caseId);
613
- const evidenceDir = path.resolve(repoRoot, ctx.evidenceDir);
635
+ const caseEvidenceRoot = path.resolve(repoRoot, ctx.evidenceDir);
636
+ const evidenceDir = caseEvidenceRoot;
614
637
  const target = path.resolve(repoRoot, outputFilename);
615
638
  if (!pathIsContained(repoRoot, caseEvidenceRoot) ||
616
639
  !pathIsContained(caseEvidenceRoot, evidenceDir) ||
@@ -818,7 +841,10 @@ export async function executePlaywrightCliCommand(input, ctx, signal) {
818
841
  };
819
842
  }
820
843
  let executable = PLAYWRIGHT_CLI_EXECUTABLE;
821
- let argv = [command, ...args];
844
+ // find-unique executes as the underlying CLI `find` command; uniqueness is
845
+ // a controller-side judgment on the parsed match count.
846
+ const cliCommand = command === "find-unique" ? "find" : command;
847
+ let argv = [cliCommand, ...args];
822
848
  try {
823
849
  await validatePlaywrightCliOutputPath(command, args, ctx);
824
850
  await validatePlaywrightCliInputFiles(command, args, ctx);
@@ -827,7 +853,7 @@ export async function executePlaywrightCliCommand(input, ctx, signal) {
827
853
  if (!ctx.runCommand) {
828
854
  const launcher = resolvePlaywrightCliLauncher(ctx.repoRoot);
829
855
  executable = launcher.executable;
830
- argv = [...launcher.argvPrefix, command, ...args];
856
+ argv = [...launcher.argvPrefix, cliCommand, ...args];
831
857
  }
832
858
  }
833
859
  catch (error) {
@@ -887,6 +913,19 @@ export async function executePlaywrightCliCommand(input, ctx, signal) {
887
913
  }
888
914
  const finishedAt = new Date();
889
915
  const combined = `${runResult.stdout}\n${runResult.stderr}`;
916
+ // find-family commands report their match count in stdout with exit code
917
+ // 0 even on zero matches — parse the count so receipts carry the real
918
+ // assertion signal for finalize-side audits.
919
+ const matchCount = command === "find" || command === "find-unique"
920
+ ? parsePlaywrightCliFindMatchCount(runResult.stdout || "")
921
+ : undefined;
922
+ // find-unique fails closed unless exactly one match was found (including
923
+ // zero and unparseable outputs).
924
+ const findUniqueFailed = command === "find-unique" && matchCount !== 1 && !runResult.timedOut;
925
+ const effectiveExitCode = findUniqueFailed && runResult.exitCode === 0 ? 1 : runResult.exitCode;
926
+ const effectiveStderr = findUniqueFailed
927
+ ? `${runResult.stderr}${runResult.stderr ? "\n" : ""}find-unique: expected exactly 1 match, got ${matchCount === undefined ? "unparseable output" : `${matchCount}`}`
928
+ : runResult.stderr;
890
929
  const receipt = {
891
930
  schemaVersion: 1,
892
931
  caseId: ctx.caseId,
@@ -895,25 +934,28 @@ export async function executePlaywrightCliCommand(input, ctx, signal) {
895
934
  argsRedacted: redactPlaywrightCliArgs(command, args),
896
935
  startedAt: startedAt.toISOString(),
897
936
  finishedAt: finishedAt.toISOString(),
898
- exitCode: runResult.exitCode,
937
+ exitCode: effectiveExitCode,
899
938
  durationMs: finishedAt.getTime() - startedAt.getTime(),
900
939
  outputSha256: createHash("sha256").update(combined).digest("hex"),
901
940
  toolVersion: ctx.toolVersion ?? "unknown",
902
941
  timedOut: runResult.timedOut || undefined,
942
+ ...(matchCount !== undefined ? { matchCount } : {}),
903
943
  errorClass: runResult.timedOut
904
944
  ? "timeout"
905
945
  : runResult.signal === "abort"
906
946
  ? "abort"
907
- : runResult.exitCode === 0
947
+ : effectiveExitCode === 0
908
948
  ? undefined
909
- : "nonzero-exit",
949
+ : findUniqueFailed
950
+ ? "find-unique-count-mismatch"
951
+ : "nonzero-exit",
910
952
  };
911
953
  await appendReceipt(ctx, receipt);
912
954
  return {
913
- ok: runResult.exitCode === 0 && !runResult.timedOut,
914
- exitCode: runResult.exitCode,
955
+ ok: effectiveExitCode === 0 && !runResult.timedOut,
956
+ exitCode: effectiveExitCode,
915
957
  stdout: runResult.stdout,
916
- stderr: runResult.stderr,
958
+ stderr: effectiveStderr,
917
959
  timedOut: runResult.timedOut,
918
960
  receipt,
919
961
  errorClass: receipt.errorClass,
@@ -985,22 +1027,26 @@ export async function createPlaywrightCliTool(ctx) {
985
1027
  export const PI_COMMAND_CAPABILITY_REGISTRY = {
986
1028
  "playwright-cli": createPlaywrightCliTool,
987
1029
  };
988
- export function resolveCaseIdFromWriteSet(writeSet) {
1030
+ export function resolveCaseIdFromWriteSet(writeSet, layout = defaultFrontendTestLayout()) {
1031
+ const prefix = layout.evidenceDir.replace(/\/$/, "");
1032
+ const escaped = escapeRegexLiteral(prefix);
1033
+ const exact = new RegExp(`^${escaped}/([^/]+)(?:/\\*\\*)?$`);
1034
+ const nested = new RegExp(`^${escaped}/([^/]+)/`);
989
1035
  for (const entry of writeSet ?? []) {
990
1036
  const normalized = normalizePosix(entry);
991
- const match = normalized.match(/^testcase\/frontend\/evidence\/([^/]+)(?:\/\*\*)?$/);
1037
+ const match = normalized.match(exact);
992
1038
  if (match?.[1] && match[1] !== "**" && match[1] !== "*") {
993
1039
  return match[1];
994
1040
  }
995
- const nested = normalized.match(/^testcase\/frontend\/evidence\/([^/]+)\//);
996
- if (nested?.[1])
997
- return nested[1];
1041
+ const nestedMatch = normalized.match(nested);
1042
+ if (nestedMatch?.[1])
1043
+ return nestedMatch[1];
998
1044
  }
999
1045
  return undefined;
1000
1046
  }
1001
- export function resolveEvidenceDirFromWriteSet(writeSet) {
1002
- const caseId = resolveCaseIdFromWriteSet(writeSet);
1047
+ export function resolveEvidenceDirFromWriteSet(writeSet, layout = defaultFrontendTestLayout()) {
1048
+ const caseId = resolveCaseIdFromWriteSet(writeSet, layout);
1003
1049
  if (!caseId)
1004
1050
  return undefined;
1005
- return `testcase/frontend/evidence/${caseId}`;
1051
+ return `${layout.evidenceDir}/${caseId}`;
1006
1052
  }