@claude-flow/cli 3.32.8 → 3.32.10

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 (448) hide show
  1. package/.claude/agents/analysis/analyze-code-quality.md +178 -178
  2. package/.claude/agents/analysis/code-analyzer.md +209 -209
  3. package/.claude/agents/analysis/code-review/analyze-code-quality.md +178 -178
  4. package/.claude/agents/architecture/arch-system-design.md +156 -156
  5. package/.claude/agents/architecture/system-design/arch-system-design.md +154 -154
  6. package/.claude/agents/browser/browser-agent.yaml +182 -182
  7. package/.claude/agents/consensus/byzantine-coordinator.md +62 -62
  8. package/.claude/agents/consensus/crdt-synchronizer.md +996 -996
  9. package/.claude/agents/consensus/gossip-coordinator.md +62 -62
  10. package/.claude/agents/consensus/performance-benchmarker.md +850 -850
  11. package/.claude/agents/consensus/quorum-manager.md +822 -822
  12. package/.claude/agents/consensus/raft-manager.md +62 -62
  13. package/.claude/agents/consensus/security-manager.md +621 -621
  14. package/.claude/agents/core/planner.md +374 -374
  15. package/.claude/agents/custom/test-long-runner.md +44 -44
  16. package/.claude/agents/data/data-ml-model.md +444 -444
  17. package/.claude/agents/data/ml/data-ml-model.md +192 -192
  18. package/.claude/agents/development/backend/dev-backend-api.md +141 -141
  19. package/.claude/agents/development/dev-backend-api.md +344 -344
  20. package/.claude/agents/devops/ci-cd/ops-cicd-github.md +163 -163
  21. package/.claude/agents/devops/ops-cicd-github.md +164 -164
  22. package/.claude/agents/documentation/api-docs/docs-api-openapi.md +173 -173
  23. package/.claude/agents/documentation/docs-api-openapi.md +354 -354
  24. package/.claude/agents/flow-nexus/app-store.md +87 -87
  25. package/.claude/agents/flow-nexus/authentication.md +68 -68
  26. package/.claude/agents/flow-nexus/challenges.md +80 -80
  27. package/.claude/agents/flow-nexus/neural-network.md +87 -87
  28. package/.claude/agents/flow-nexus/payments.md +82 -82
  29. package/.claude/agents/flow-nexus/sandbox.md +75 -75
  30. package/.claude/agents/flow-nexus/swarm.md +75 -75
  31. package/.claude/agents/flow-nexus/user-tools.md +95 -95
  32. package/.claude/agents/flow-nexus/workflow.md +83 -83
  33. package/.claude/agents/github/code-review-swarm.md +377 -377
  34. package/.claude/agents/github/github-modes.md +172 -172
  35. package/.claude/agents/github/issue-tracker.md +575 -575
  36. package/.claude/agents/github/multi-repo-swarm.md +552 -552
  37. package/.claude/agents/github/pr-manager.md +437 -437
  38. package/.claude/agents/github/project-board-sync.md +508 -508
  39. package/.claude/agents/github/release-manager.md +604 -604
  40. package/.claude/agents/github/release-swarm.md +582 -582
  41. package/.claude/agents/github/repo-architect.md +397 -397
  42. package/.claude/agents/github/swarm-issue.md +572 -572
  43. package/.claude/agents/github/swarm-pr.md +427 -427
  44. package/.claude/agents/github/sync-coordinator.md +451 -451
  45. package/.claude/agents/github/workflow-automation.md +902 -902
  46. package/.claude/agents/goal/agent.md +815 -815
  47. package/.claude/agents/optimization/benchmark-suite.md +664 -664
  48. package/.claude/agents/optimization/load-balancer.md +430 -430
  49. package/.claude/agents/optimization/performance-monitor.md +671 -671
  50. package/.claude/agents/optimization/resource-allocator.md +673 -673
  51. package/.claude/agents/optimization/topology-optimizer.md +807 -807
  52. package/.claude/agents/payments/agentic-payments.md +126 -126
  53. package/.claude/agents/sona/sona-learning-optimizer.md +74 -74
  54. package/.claude/agents/sparc/architecture.md +698 -698
  55. package/.claude/agents/sparc/pseudocode.md +519 -519
  56. package/.claude/agents/sparc/refinement.md +801 -801
  57. package/.claude/agents/sparc/specification.md +477 -477
  58. package/.claude/agents/specialized/mobile/spec-mobile-react-native.md +224 -224
  59. package/.claude/agents/specialized/spec-mobile-react-native.md +226 -226
  60. package/.claude/agents/sublinear/consensus-coordinator.md +337 -337
  61. package/.claude/agents/sublinear/matrix-optimizer.md +184 -184
  62. package/.claude/agents/sublinear/pagerank-analyzer.md +298 -298
  63. package/.claude/agents/sublinear/performance-optimizer.md +367 -367
  64. package/.claude/agents/sublinear/trading-predictor.md +245 -245
  65. package/.claude/agents/swarm/adaptive-coordinator.md +1126 -1126
  66. package/.claude/agents/swarm/hierarchical-coordinator.md +709 -709
  67. package/.claude/agents/swarm/mesh-coordinator.md +962 -962
  68. package/.claude/agents/templates/automation-smart-agent.md +204 -204
  69. package/.claude/agents/templates/base-template-generator.md +289 -289
  70. package/.claude/agents/templates/coordinator-swarm-init.md +89 -89
  71. package/.claude/agents/templates/github-pr-manager.md +176 -176
  72. package/.claude/agents/templates/implementer-sparc-coder.md +258 -258
  73. package/.claude/agents/templates/memory-coordinator.md +186 -186
  74. package/.claude/agents/templates/orchestrator-task.md +138 -138
  75. package/.claude/agents/templates/performance-analyzer.md +198 -198
  76. package/.claude/agents/templates/sparc-coordinator.md +513 -513
  77. package/.claude/agents/testing/production-validator.md +394 -394
  78. package/.claude/agents/testing/tdd-london-swarm.md +243 -243
  79. package/.claude/agents/v3/aidefence-guardian.md +282 -282
  80. package/.claude/agents/v3/claims-authorizer.md +208 -208
  81. package/.claude/agents/v3/collective-intelligence-coordinator.md +993 -993
  82. package/.claude/agents/v3/ddd-domain-expert.md +220 -220
  83. package/.claude/agents/v3/injection-analyst.md +236 -236
  84. package/.claude/agents/v3/performance-engineer.md +1233 -1233
  85. package/.claude/agents/v3/pii-detector.md +151 -151
  86. package/.claude/agents/v3/reasoningbank-learner.md +213 -213
  87. package/.claude/agents/v3/security-architect-aidefence.md +410 -410
  88. package/.claude/agents/v3/security-architect.md +867 -867
  89. package/.claude/agents/v3/swarm-memory-manager.md +157 -157
  90. package/.claude/agents/v3/v3-integration-architect.md +205 -205
  91. package/.claude/commands/agents/README.md +50 -50
  92. package/.claude/commands/agents/agent-capabilities.md +140 -140
  93. package/.claude/commands/agents/agent-coordination.md +28 -28
  94. package/.claude/commands/agents/agent-spawning.md +28 -28
  95. package/.claude/commands/agents/agent-types.md +216 -216
  96. package/.claude/commands/agents/health.md +139 -139
  97. package/.claude/commands/agents/list.md +100 -100
  98. package/.claude/commands/agents/logs.md +130 -130
  99. package/.claude/commands/agents/metrics.md +122 -122
  100. package/.claude/commands/agents/pool.md +127 -127
  101. package/.claude/commands/agents/spawn.md +140 -140
  102. package/.claude/commands/agents/status.md +115 -115
  103. package/.claude/commands/agents/stop.md +102 -102
  104. package/.claude/commands/analysis/COMMAND_COMPLIANCE_REPORT.md +53 -53
  105. package/.claude/commands/analysis/README.md +9 -9
  106. package/.claude/commands/analysis/bottleneck-detect.md +162 -162
  107. package/.claude/commands/analysis/performance-bottlenecks.md +58 -58
  108. package/.claude/commands/analysis/performance-report.md +25 -25
  109. package/.claude/commands/analysis/token-efficiency.md +44 -44
  110. package/.claude/commands/analysis/token-usage.md +25 -25
  111. package/.claude/commands/automation/README.md +9 -9
  112. package/.claude/commands/automation/auto-agent.md +122 -122
  113. package/.claude/commands/automation/self-healing.md +105 -105
  114. package/.claude/commands/automation/session-memory.md +89 -89
  115. package/.claude/commands/automation/smart-agents.md +72 -72
  116. package/.claude/commands/automation/smart-spawn.md +25 -25
  117. package/.claude/commands/automation/workflow-select.md +25 -25
  118. package/.claude/commands/claude-flow-help.md +103 -103
  119. package/.claude/commands/claude-flow-memory.md +107 -107
  120. package/.claude/commands/claude-flow-swarm.md +205 -205
  121. package/.claude/commands/coordination/README.md +9 -9
  122. package/.claude/commands/coordination/agent-spawn.md +25 -25
  123. package/.claude/commands/coordination/init.md +44 -44
  124. package/.claude/commands/coordination/orchestrate.md +43 -43
  125. package/.claude/commands/coordination/spawn.md +45 -45
  126. package/.claude/commands/coordination/swarm-init.md +85 -85
  127. package/.claude/commands/coordination/task-orchestrate.md +25 -25
  128. package/.claude/commands/github/README.md +11 -11
  129. package/.claude/commands/github/code-review-swarm.md +513 -513
  130. package/.claude/commands/github/code-review.md +25 -25
  131. package/.claude/commands/github/github-modes.md +146 -146
  132. package/.claude/commands/github/github-swarm.md +121 -121
  133. package/.claude/commands/github/issue-tracker.md +291 -291
  134. package/.claude/commands/github/issue-triage.md +25 -25
  135. package/.claude/commands/github/multi-repo-swarm.md +518 -518
  136. package/.claude/commands/github/pr-enhance.md +26 -26
  137. package/.claude/commands/github/pr-manager.md +169 -169
  138. package/.claude/commands/github/project-board-sync.md +470 -470
  139. package/.claude/commands/github/release-manager.md +339 -339
  140. package/.claude/commands/github/release-swarm.md +543 -543
  141. package/.claude/commands/github/repo-analyze.md +25 -25
  142. package/.claude/commands/github/repo-architect.md +366 -366
  143. package/.claude/commands/github/swarm-issue.md +484 -484
  144. package/.claude/commands/github/swarm-pr.md +287 -287
  145. package/.claude/commands/github/sync-coordinator.md +302 -302
  146. package/.claude/commands/github/workflow-automation.md +441 -441
  147. package/.claude/commands/hive-mind/README.md +17 -17
  148. package/.claude/commands/hive-mind/hive-mind-consensus.md +8 -8
  149. package/.claude/commands/hive-mind/hive-mind-init.md +18 -18
  150. package/.claude/commands/hive-mind/hive-mind-memory.md +8 -8
  151. package/.claude/commands/hive-mind/hive-mind-metrics.md +8 -8
  152. package/.claude/commands/hive-mind/hive-mind-resume.md +8 -8
  153. package/.claude/commands/hive-mind/hive-mind-sessions.md +8 -8
  154. package/.claude/commands/hive-mind/hive-mind-spawn.md +21 -21
  155. package/.claude/commands/hive-mind/hive-mind-status.md +8 -8
  156. package/.claude/commands/hive-mind/hive-mind-stop.md +8 -8
  157. package/.claude/commands/hive-mind/hive-mind-wizard.md +8 -8
  158. package/.claude/commands/hive-mind/hive-mind.md +27 -27
  159. package/.claude/commands/hooks/README.md +11 -11
  160. package/.claude/commands/hooks/overview.md +57 -57
  161. package/.claude/commands/hooks/post-edit.md +117 -117
  162. package/.claude/commands/hooks/post-task.md +112 -112
  163. package/.claude/commands/hooks/pre-edit.md +113 -113
  164. package/.claude/commands/hooks/pre-task.md +111 -111
  165. package/.claude/commands/hooks/session-end.md +118 -118
  166. package/.claude/commands/hooks/setup.md +102 -102
  167. package/.claude/commands/memory/README.md +9 -9
  168. package/.claude/commands/memory/memory-persist.md +25 -25
  169. package/.claude/commands/memory/memory-search.md +25 -25
  170. package/.claude/commands/memory/memory-usage.md +25 -25
  171. package/.claude/commands/memory/neural.md +47 -47
  172. package/.claude/commands/monitoring/README.md +9 -9
  173. package/.claude/commands/monitoring/agent-metrics.md +25 -25
  174. package/.claude/commands/monitoring/agents.md +44 -44
  175. package/.claude/commands/monitoring/real-time-view.md +25 -25
  176. package/.claude/commands/monitoring/status.md +46 -46
  177. package/.claude/commands/monitoring/swarm-monitor.md +25 -25
  178. package/.claude/commands/optimization/README.md +9 -9
  179. package/.claude/commands/optimization/auto-topology.md +61 -61
  180. package/.claude/commands/optimization/cache-manage.md +25 -25
  181. package/.claude/commands/optimization/parallel-execute.md +25 -25
  182. package/.claude/commands/optimization/parallel-execution.md +49 -49
  183. package/.claude/commands/optimization/topology-optimize.md +25 -25
  184. package/.claude/commands/pair/README.md +260 -260
  185. package/.claude/commands/pair/commands.md +545 -545
  186. package/.claude/commands/pair/config.md +509 -509
  187. package/.claude/commands/pair/examples.md +511 -511
  188. package/.claude/commands/pair/modes.md +347 -347
  189. package/.claude/commands/pair/session.md +406 -406
  190. package/.claude/commands/pair/start.md +208 -208
  191. package/.claude/commands/sparc/analyzer.md +51 -51
  192. package/.claude/commands/sparc/architect.md +53 -53
  193. package/.claude/commands/sparc/ask.md +97 -97
  194. package/.claude/commands/sparc/batch-executor.md +54 -54
  195. package/.claude/commands/sparc/code.md +89 -89
  196. package/.claude/commands/sparc/coder.md +54 -54
  197. package/.claude/commands/sparc/debug.md +83 -83
  198. package/.claude/commands/sparc/debugger.md +54 -54
  199. package/.claude/commands/sparc/designer.md +53 -53
  200. package/.claude/commands/sparc/devops.md +109 -109
  201. package/.claude/commands/sparc/docs-writer.md +80 -80
  202. package/.claude/commands/sparc/documenter.md +54 -54
  203. package/.claude/commands/sparc/innovator.md +54 -54
  204. package/.claude/commands/sparc/integration.md +83 -83
  205. package/.claude/commands/sparc/mcp.md +117 -117
  206. package/.claude/commands/sparc/memory-manager.md +54 -54
  207. package/.claude/commands/sparc/optimizer.md +54 -54
  208. package/.claude/commands/sparc/orchestrator.md +131 -131
  209. package/.claude/commands/sparc/post-deployment-monitoring-mode.md +83 -83
  210. package/.claude/commands/sparc/refinement-optimization-mode.md +83 -83
  211. package/.claude/commands/sparc/researcher.md +54 -54
  212. package/.claude/commands/sparc/reviewer.md +54 -54
  213. package/.claude/commands/sparc/security-review.md +80 -80
  214. package/.claude/commands/sparc/sparc-modes.md +174 -174
  215. package/.claude/commands/sparc/sparc.md +111 -111
  216. package/.claude/commands/sparc/spec-pseudocode.md +80 -80
  217. package/.claude/commands/sparc/supabase-admin.md +348 -348
  218. package/.claude/commands/sparc/swarm-coordinator.md +54 -54
  219. package/.claude/commands/sparc/tdd.md +54 -54
  220. package/.claude/commands/sparc/tester.md +54 -54
  221. package/.claude/commands/sparc/tutorial.md +79 -79
  222. package/.claude/commands/sparc/workflow-manager.md +54 -54
  223. package/.claude/commands/sparc.md +166 -166
  224. package/.claude/commands/stream-chain/pipeline.md +120 -120
  225. package/.claude/commands/stream-chain/run.md +69 -69
  226. package/.claude/commands/swarm/README.md +15 -15
  227. package/.claude/commands/swarm/analysis.md +95 -95
  228. package/.claude/commands/swarm/development.md +96 -96
  229. package/.claude/commands/swarm/examples.md +168 -168
  230. package/.claude/commands/swarm/maintenance.md +102 -102
  231. package/.claude/commands/swarm/optimization.md +117 -117
  232. package/.claude/commands/swarm/research.md +136 -136
  233. package/.claude/commands/swarm/swarm-analysis.md +8 -8
  234. package/.claude/commands/swarm/swarm-background.md +8 -8
  235. package/.claude/commands/swarm/swarm-init.md +19 -19
  236. package/.claude/commands/swarm/swarm-modes.md +8 -8
  237. package/.claude/commands/swarm/swarm-monitor.md +8 -8
  238. package/.claude/commands/swarm/swarm-spawn.md +19 -19
  239. package/.claude/commands/swarm/swarm-status.md +8 -8
  240. package/.claude/commands/swarm/swarm-strategies.md +8 -8
  241. package/.claude/commands/swarm/swarm.md +87 -87
  242. package/.claude/commands/swarm/testing.md +131 -131
  243. package/.claude/commands/training/README.md +9 -9
  244. package/.claude/commands/training/model-update.md +25 -25
  245. package/.claude/commands/training/neural-patterns.md +107 -107
  246. package/.claude/commands/training/neural-train.md +75 -75
  247. package/.claude/commands/training/pattern-learn.md +25 -25
  248. package/.claude/commands/training/specialization.md +62 -62
  249. package/.claude/commands/truth/start.md +142 -142
  250. package/.claude/commands/verify/check.md +49 -49
  251. package/.claude/commands/verify/start.md +127 -127
  252. package/.claude/commands/workflows/README.md +9 -9
  253. package/.claude/commands/workflows/development.md +77 -77
  254. package/.claude/commands/workflows/research.md +62 -62
  255. package/.claude/commands/workflows/workflow-create.md +25 -25
  256. package/.claude/commands/workflows/workflow-execute.md +25 -25
  257. package/.claude/commands/workflows/workflow-export.md +25 -25
  258. package/.claude/eval/human-relevance-frozen-v1.json +17 -17
  259. package/.claude/evolve-proof/generation-0.json +211 -211
  260. package/.claude/evolve-proof/real-generation-0.json +406 -406
  261. package/.claude/evolve-proof/real-generation-1.json +406 -406
  262. package/.claude/helpers/README.md +96 -96
  263. package/.claude/helpers/adr-compliance.sh +186 -186
  264. package/.claude/helpers/auto-commit.sh +178 -178
  265. package/.claude/helpers/auto-memory-hook.mjs +430 -430
  266. package/.claude/helpers/checkpoint-manager.sh +251 -251
  267. package/.claude/helpers/daemon-manager.sh +252 -252
  268. package/.claude/helpers/ddd-tracker.sh +144 -144
  269. package/.claude/helpers/github-safe.js +156 -156
  270. package/.claude/helpers/github-setup.sh +45 -45
  271. package/.claude/helpers/guidance-hook.sh +13 -13
  272. package/.claude/helpers/guidance-hooks.sh +102 -102
  273. package/.claude/helpers/health-monitor.sh +108 -108
  274. package/.claude/helpers/helpers.manifest.json +6 -6
  275. package/.claude/helpers/hook-handler.cjs +565 -565
  276. package/.claude/helpers/intelligence.cjs +1058 -1058
  277. package/.claude/helpers/learning-hooks.sh +329 -329
  278. package/.claude/helpers/learning-optimizer.sh +127 -127
  279. package/.claude/helpers/learning-service.mjs +1144 -1144
  280. package/.claude/helpers/memory.js +83 -83
  281. package/.claude/helpers/metrics-db.mjs +503 -503
  282. package/.claude/helpers/pattern-consolidator.sh +86 -86
  283. package/.claude/helpers/perf-worker.sh +160 -160
  284. package/.claude/helpers/post-commit +16 -16
  285. package/.claude/helpers/pre-commit +26 -26
  286. package/.claude/helpers/quick-start.sh +19 -19
  287. package/.claude/helpers/router.js +105 -105
  288. package/.claude/helpers/security-scanner.sh +127 -127
  289. package/.claude/helpers/session.js +157 -157
  290. package/.claude/helpers/setup-mcp.sh +18 -18
  291. package/.claude/helpers/standard-checkpoint-hooks.sh +189 -189
  292. package/.claude/helpers/statusline-hook.sh +21 -21
  293. package/.claude/helpers/statusline.cjs +1060 -1006
  294. package/.claude/helpers/statusline.js +340 -340
  295. package/.claude/helpers/swarm-comms.sh +353 -353
  296. package/.claude/helpers/swarm-hooks.sh +761 -761
  297. package/.claude/helpers/swarm-monitor.sh +210 -210
  298. package/.claude/helpers/sync-v3-metrics.sh +245 -245
  299. package/.claude/helpers/update-v3-progress.sh +165 -165
  300. package/.claude/helpers/v3-quick-status.sh +57 -57
  301. package/.claude/helpers/v3.sh +110 -110
  302. package/.claude/helpers/validate-v3-config.sh +215 -215
  303. package/.claude/helpers/worker-manager.sh +170 -170
  304. package/.claude/proven-config.manifest.json +37 -37
  305. package/.claude/proven-config.signed.json +41 -41
  306. package/.claude/settings.json +182 -182
  307. package/.claude/skills/agentdb-advanced/SKILL.md +550 -550
  308. package/.claude/skills/agentdb-learning/SKILL.md +545 -545
  309. package/.claude/skills/agentdb-memory-patterns/SKILL.md +339 -339
  310. package/.claude/skills/agentdb-optimization/SKILL.md +509 -509
  311. package/.claude/skills/agentdb-vector-search/SKILL.md +339 -339
  312. package/.claude/skills/browser/SKILL.md +204 -204
  313. package/.claude/skills/dual-mode/README.md +71 -71
  314. package/.claude/skills/dual-mode/dual-collect.md +103 -103
  315. package/.claude/skills/dual-mode/dual-coordinate.md +85 -85
  316. package/.claude/skills/dual-mode/dual-spawn.md +81 -81
  317. package/.claude/skills/flow-nexus-neural/SKILL.md +727 -727
  318. package/.claude/skills/flow-nexus-platform/SKILL.md +1154 -1154
  319. package/.claude/skills/flow-nexus-swarm/SKILL.md +604 -604
  320. package/.claude/skills/github-code-review/SKILL.md +1125 -1125
  321. package/.claude/skills/github-multi-repo/SKILL.md +862 -862
  322. package/.claude/skills/github-project-management/SKILL.md +1262 -1262
  323. package/.claude/skills/github-release-management/SKILL.md +1064 -1064
  324. package/.claude/skills/github-workflow-automation/SKILL.md +1047 -1047
  325. package/.claude/skills/hooks-automation/SKILL.md +1201 -1201
  326. package/.claude/skills/pair-programming/SKILL.md +1202 -1202
  327. package/.claude/skills/reasoningbank-agentdb/SKILL.md +446 -446
  328. package/.claude/skills/reasoningbank-intelligence/SKILL.md +201 -201
  329. package/.claude/skills/skill-builder/SKILL.md +910 -910
  330. package/.claude/skills/sparc-methodology/SKILL.md +1106 -1106
  331. package/.claude/skills/stream-chain/SKILL.md +560 -560
  332. package/.claude/skills/swarm-advanced/SKILL.md +970 -970
  333. package/.claude/skills/swarm-orchestration/SKILL.md +179 -179
  334. package/.claude/skills/v3-cli-modernization/SKILL.md +871 -871
  335. package/.claude/skills/v3-core-implementation/SKILL.md +796 -796
  336. package/.claude/skills/v3-ddd-architecture/SKILL.md +441 -441
  337. package/.claude/skills/v3-integration-deep/SKILL.md +240 -240
  338. package/.claude/skills/v3-mcp-optimization/SKILL.md +776 -776
  339. package/.claude/skills/v3-memory-unification/SKILL.md +173 -173
  340. package/.claude/skills/v3-performance-optimization/SKILL.md +389 -389
  341. package/.claude/skills/v3-security-overhaul/SKILL.md +81 -81
  342. package/.claude/skills/v3-swarm-coordination/SKILL.md +339 -339
  343. package/.claude/skills/verification-quality/SKILL.md +691 -691
  344. package/README.md +419 -419
  345. package/bin/cli.js +314 -314
  346. package/bin/mcp-server.js +224 -224
  347. package/bin/preinstall.cjs +2 -2
  348. package/catalog-manifest.json +2 -2
  349. package/dist/src/autopilot-state.js +24 -7
  350. package/dist/src/benchmarks/gaia-critic.js +24 -24
  351. package/dist/src/business-pods/bbs-budget-tracker.js +53 -53
  352. package/dist/src/commands/completions.js +409 -409
  353. package/dist/src/commands/daemon.js +44 -44
  354. package/dist/src/commands/doctor.js +267 -20
  355. package/dist/src/commands/embeddings.js +26 -26
  356. package/dist/src/commands/hive-mind.js +97 -97
  357. package/dist/src/commands/hooks.js +74 -11
  358. package/dist/src/commands/init.js +202 -34
  359. package/dist/src/commands/memory.js +12 -1
  360. package/dist/src/commands/ruvector/backup.js +23 -23
  361. package/dist/src/commands/ruvector/benchmark.js +31 -31
  362. package/dist/src/commands/ruvector/import.js +14 -14
  363. package/dist/src/commands/ruvector/init.js +115 -115
  364. package/dist/src/commands/ruvector/migrate.js +99 -99
  365. package/dist/src/commands/ruvector/optimize.js +51 -51
  366. package/dist/src/commands/ruvector/setup.js +624 -624
  367. package/dist/src/commands/ruvector/status.js +38 -38
  368. package/dist/src/config/proven-config.js +2 -2
  369. package/dist/src/funnel/disclosure.js +13 -2
  370. package/dist/src/funnel/message-transport.d.ts +11 -4
  371. package/dist/src/funnel/message-transport.js +11 -4
  372. package/dist/src/funnel/messages.d.ts +12 -10
  373. package/dist/src/funnel/messages.js +83 -11
  374. package/dist/src/init/claudemd-generator.js +231 -231
  375. package/dist/src/init/executor.js +453 -453
  376. package/dist/src/init/helper-signing.js +2 -2
  377. package/dist/src/init/helpers-generator.js +751 -751
  378. package/dist/src/init/statusline-generator.js +24 -24
  379. package/dist/src/mcp-tools/agentdb-tools.js +15 -15
  380. package/dist/src/mcp-tools/browser-intent-tools.js +19 -19
  381. package/dist/src/mcp-tools/browser-tools.js +8 -0
  382. package/dist/src/mcp-tools/hooks-tools.js +21 -0
  383. package/dist/src/mcp-tools/memory-tools.js +4 -3
  384. package/dist/src/memory/graph-edge-writer.d.ts +9 -0
  385. package/dist/src/memory/graph-edge-writer.js +35 -22
  386. package/dist/src/memory/memory-bridge.js +189 -93
  387. package/dist/src/memory/memory-initializer.js +479 -407
  388. package/dist/src/memory/rabitq-index.js +5 -5
  389. package/dist/src/parser.js +25 -9
  390. package/dist/src/proxy/verify.js +2 -2
  391. package/dist/src/runtime/headless.js +28 -28
  392. package/dist/src/services/distill-tuning.js +7 -7
  393. package/dist/src/services/headless-worker-executor.js +84 -84
  394. package/dist/src/services/memory-distillation.js +4 -4
  395. package/dist/src/services/worker-daemon.js +7 -4
  396. package/dist/src/transfer/deploy-seraphine.js +23 -23
  397. package/package.json +137 -135
  398. package/plugins/ruflo-metaharness/.claude-plugin/plugin.json +32 -32
  399. package/plugins/ruflo-metaharness/README.md +72 -72
  400. package/plugins/ruflo-metaharness/agents/metaharness-architect.md +58 -58
  401. package/plugins/ruflo-metaharness/commands/ruflo-metaharness.md +48 -48
  402. package/plugins/ruflo-metaharness/scripts/_darwin.mjs +210 -210
  403. package/plugins/ruflo-metaharness/scripts/_harness.mjs +330 -330
  404. package/plugins/ruflo-metaharness/scripts/_invoke.mjs +231 -231
  405. package/plugins/ruflo-metaharness/scripts/_redblue.mjs +143 -143
  406. package/plugins/ruflo-metaharness/scripts/_similarity.mjs +161 -161
  407. package/plugins/ruflo-metaharness/scripts/_spike-similarity.mjs +223 -223
  408. package/plugins/ruflo-metaharness/scripts/audit-list.mjs +158 -158
  409. package/plugins/ruflo-metaharness/scripts/audit-trend.mjs +272 -272
  410. package/plugins/ruflo-metaharness/scripts/bench-parse-mcp-scan.mjs +146 -146
  411. package/plugins/ruflo-metaharness/scripts/bench-recordpair-overhead.mjs +186 -186
  412. package/plugins/ruflo-metaharness/scripts/bench-similarity.mjs +177 -177
  413. package/plugins/ruflo-metaharness/scripts/bench.mjs +95 -95
  414. package/plugins/ruflo-metaharness/scripts/drift-from-history.mjs +363 -363
  415. package/plugins/ruflo-metaharness/scripts/evolve.mjs +404 -404
  416. package/plugins/ruflo-metaharness/scripts/genome.mjs +80 -80
  417. package/plugins/ruflo-metaharness/scripts/gepa.mjs +153 -153
  418. package/plugins/ruflo-metaharness/scripts/learn.mjs +127 -127
  419. package/plugins/ruflo-metaharness/scripts/mcp-scan.mjs +111 -111
  420. package/plugins/ruflo-metaharness/scripts/mint.mjs +126 -126
  421. package/plugins/ruflo-metaharness/scripts/oia-audit.mjs +228 -228
  422. package/plugins/ruflo-metaharness/scripts/redblue.mjs +286 -286
  423. package/plugins/ruflo-metaharness/scripts/router-parallel-analyze.mjs +250 -250
  424. package/plugins/ruflo-metaharness/scripts/score.mjs +92 -92
  425. package/plugins/ruflo-metaharness/scripts/security-bench.mjs +174 -174
  426. package/plugins/ruflo-metaharness/scripts/similarity.mjs +158 -158
  427. package/plugins/ruflo-metaharness/scripts/smoke.sh +2356 -2356
  428. package/plugins/ruflo-metaharness/scripts/test-graceful-degradation.mjs +165 -165
  429. package/plugins/ruflo-metaharness/scripts/test-mcp-tools.mjs +472 -472
  430. package/plugins/ruflo-metaharness/scripts/test-parallel-pipeline.mjs +204 -204
  431. package/plugins/ruflo-metaharness/scripts/test-pipeline-roundtrip.mjs +586 -586
  432. package/plugins/ruflo-metaharness/scripts/test-similarity.mjs +334 -334
  433. package/plugins/ruflo-metaharness/scripts/test-with-openrouter.mjs +229 -229
  434. package/plugins/ruflo-metaharness/scripts/threat-model.mjs +59 -59
  435. package/plugins/ruflo-metaharness/skills/harness-bench/SKILL.md +64 -64
  436. package/plugins/ruflo-metaharness/skills/harness-drift-from-history/SKILL.md +65 -65
  437. package/plugins/ruflo-metaharness/skills/harness-evolve/SKILL.md +131 -131
  438. package/plugins/ruflo-metaharness/skills/harness-genome/SKILL.md +54 -54
  439. package/plugins/ruflo-metaharness/skills/harness-gepa/SKILL.md +65 -65
  440. package/plugins/ruflo-metaharness/skills/harness-learn/SKILL.md +65 -65
  441. package/plugins/ruflo-metaharness/skills/harness-mcp-scan/SKILL.md +49 -49
  442. package/plugins/ruflo-metaharness/skills/harness-mint/SKILL.md +72 -72
  443. package/plugins/ruflo-metaharness/skills/harness-oia-audit/SKILL.md +79 -79
  444. package/plugins/ruflo-metaharness/skills/harness-score/SKILL.md +66 -66
  445. package/plugins/ruflo-metaharness/skills/harness-security-bench/SKILL.md +101 -101
  446. package/plugins/ruflo-metaharness/skills/harness-similarity/SKILL.md +67 -67
  447. package/plugins/ruflo-metaharness/skills/harness-threat-model/SKILL.md +41 -41
  448. package/scripts/postinstall.cjs +153 -153
@@ -1,204 +1,204 @@
1
- #!/usr/bin/env node
2
- // test-parallel-pipeline — end-to-end integration test for ADR-150's
3
- // SelfEvolvingRouter parallel-logging pipeline.
4
- //
5
- // Composition under test:
6
- // recordPair() / recordPairOutcome() [iter 11 — TS module]
7
- // → .swarm/router-parallel.jsonl [JSONL trajectory]
8
- // → router-parallel-analyze.mjs [iter 10 — analyzer]
9
- // → 3-criteria AND-gate verdict [ADR-150 review-round-1]
10
- //
11
- // This script bypasses model-router.ts dispatch and calls the recorder
12
- // directly, so it tests the recorder + analyzer composition WITHOUT
13
- // requiring a built CLI dist. Iter-12 dispatch wiring is covered by
14
- // the smoke step 17f source-grep.
15
- //
16
- // USAGE
17
- // node scripts/test-parallel-pipeline.mjs # default
18
- // TEST_PIPELINE_KEEP_FIXTURE=1 node scripts/test-parallel-pipeline.mjs
19
- // # leave temp files for debugging
20
- //
21
- // EXIT CODES
22
- // 0 all assertions passed
23
- // 1 at least one assertion failed
24
- // 2 setup error (e.g. tsc dist not buildable, recorder module missing)
25
-
26
- import { spawnSync } from 'node:child_process';
27
- import { mkdtempSync, rmSync, mkdirSync, existsSync, readFileSync, writeFileSync } from 'node:fs';
28
- import { tmpdir } from 'node:os';
29
- import { join, dirname } from 'node:path';
30
- import { fileURLToPath } from 'node:url';
31
-
32
- const SCRIPTS_DIR = dirname(fileURLToPath(import.meta.url));
33
-
34
- let passed = 0, failed = 0;
35
- const failures = [];
36
-
37
- function assert(cond, label) {
38
- if (cond) {
39
- console.log(` ✓ ${label}`);
40
- passed++;
41
- } else {
42
- console.log(` ✗ ${label}`);
43
- failures.push(label);
44
- failed++;
45
- }
46
- }
47
-
48
- async function main() {
49
- const fixture = mkdtempSync(join(tmpdir(), 'router-parallel-e2e-'));
50
- const jsonlPath = join(fixture, 'router-parallel.jsonl');
51
- console.log(`# router-parallel pipeline e2e (fixture: ${fixture})\n`);
52
-
53
- // ──────────────────────────────────────────────────────────────────
54
- // PHASE 1 — Synthesize a realistic JSONL trajectory by hand.
55
- //
56
- // We can't easily import the TS recorder without a build step, so we
57
- // emit the same JSONL shape the recorder produces. This isolates the
58
- // analyzer pipeline test from compile state.
59
- // ──────────────────────────────────────────────────────────────────
60
- console.log('Phase 1 — synthesize trajectory');
61
-
62
- const rows = [];
63
- // 40 paired decisions: SER beats bandit on quality (+5%) at +0.5% cost
64
- // and +1% latency — should pass the 3-criteria AND-gate.
65
- for (let i = 0; i < 40; i++) {
66
- const q_actual = 0.78 + Math.random() * 0.04;
67
- const u_actual = 0.005 + Math.random() * 0.001;
68
- const l_actual = 1500 + Math.random() * 200;
69
- const disagree = i % 2 === 0;
70
- rows.push({
71
- v: 1, type: 'pair',
72
- ts: new Date(Date.now() - (40 - i) * 1000).toISOString(),
73
- task_hash: `task-${i.toString(16).padStart(8, '0')}`,
74
- bandit: {
75
- pick: 'sonnet',
76
- predictedQuality: q_actual - 0.02,
77
- predictedCostUsd: u_actual,
78
- backend: 'thompson-bandit',
79
- },
80
- ser: {
81
- pick: disagree ? 'opus' : 'sonnet',
82
- predictedQuality: disagree ? q_actual + 0.10 : q_actual + 0.01,
83
- predictedCostUsd: disagree ? u_actual * 1.005 : u_actual,
84
- backend: disagree ? 'metaharness-router-hybrid' : 'bandit-only',
85
- },
86
- outcome: {
87
- actualModel: 'sonnet',
88
- actualQuality: q_actual,
89
- actualUsd: u_actual,
90
- actualLatencyMs: Math.round(l_actual),
91
- },
92
- });
93
- }
94
-
95
- writeFileSync(jsonlPath, rows.map((r) => JSON.stringify(r)).join('\n') + '\n');
96
- assert(existsSync(jsonlPath), 'fixture JSONL written');
97
- assert(rows.length === 40, '40 pair rows synthesized');
98
-
99
- // ──────────────────────────────────────────────────────────────────
100
- // PHASE 2 — Verify recorder TS module compiles and exports the API.
101
- // ──────────────────────────────────────────────────────────────────
102
- console.log('\nPhase 2 — verify recorder TS module exports');
103
-
104
- const recorderTs = join(SCRIPTS_DIR, '..', '..', '..', 'v3', '@claude-flow', 'cli', 'src', 'ruvector', 'router-parallel-recorder.ts');
105
- const recorderSrc = readFileSync(recorderTs, 'utf-8');
106
-
107
- assert(/export function recordPair\b/.test(recorderSrc), 'recordPair exported');
108
- assert(/export function recordPairOutcome\b/.test(recorderSrc), 'recordPairOutcome exported');
109
- assert(/export function parallelRecorderStatus\b/.test(recorderSrc), 'parallelRecorderStatus exported');
110
- assert(/CLAUDE_FLOW_ROUTER_PARALLEL_LOG/.test(recorderSrc), 'env-gated by CLAUDE_FLOW_ROUTER_PARALLEL_LOG');
111
- // ADR-150 rule #3 — never throw
112
- assert(/never throws|never (throw|block)/i.test(recorderSrc), 'no-throw doc invariant present');
113
- // Try/catch around fs operations
114
- assert(/try \{[\s\S]*?appendFileSync[\s\S]*?\} catch/.test(recorderSrc), 'appendFileSync wrapped in try/catch');
115
-
116
- // ──────────────────────────────────────────────────────────────────
117
- // PHASE 3 — Run the analyzer against the synthesized JSONL.
118
- // ──────────────────────────────────────────────────────────────────
119
- console.log('\nPhase 3 — run analyzer against fixture');
120
-
121
- const analyzer = join(SCRIPTS_DIR, 'router-parallel-analyze.mjs');
122
- const r = spawnSync('node', [analyzer, '--input', jsonlPath, '--format', 'json'], {
123
- encoding: 'utf-8',
124
- timeout: 30_000,
125
- });
126
- assert(r.status === 0, `analyzer exit 0 (got ${r.status})`);
127
-
128
- let payload;
129
- try {
130
- payload = JSON.parse(r.stdout);
131
- } catch (e) {
132
- console.error(' stdout:', r.stdout.slice(0, 400));
133
- console.error(' stderr:', r.stderr.slice(0, 400));
134
- assert(false, `analyzer output parseable as JSON (${e.message})`);
135
- process.exit(2);
136
- }
137
-
138
- assert(payload.ok === true, 'payload.ok === true');
139
- assert(payload.sufficient === true, 'payload.sufficient === true (n>=30)');
140
- assert(payload.sampleSize === 40, `sampleSize === 40 (got ${payload.sampleSize})`);
141
- assert(typeof payload.disagreement?.pct === 'number', 'disagreement.pct numeric');
142
-
143
- // The 3 criteria must each be computed.
144
- assert(typeof payload.criteria?.qualityImprovementPct === 'number', 'criteria.qualityImprovementPct present');
145
- assert(typeof payload.criteria?.usdIncreasePct === 'number', 'criteria.usdIncreasePct present');
146
- assert(typeof payload.criteria?.latencyIncreasePct === 'number', 'criteria.latencyIncreasePct present');
147
-
148
- // Thresholds match ADR-150 review-round-1 exactly.
149
- assert(payload.criteria?.qualityThresholdPct === 2, 'quality threshold = 2%');
150
- assert(payload.criteria?.usdThresholdPct === 1, 'cost threshold = 1%');
151
- assert(payload.criteria?.latencyThresholdPct === 5, 'latency threshold = 5%');
152
-
153
- // For the promotable fixture, all three criteria should pass.
154
- assert(payload.criteria?.qualityPasses === true, 'quality > 2% passes');
155
- assert(payload.criteria?.costPasses === true, 'cost < 1% passes');
156
- assert(payload.criteria?.latencyPasses === true, 'latency < 5% passes');
157
- assert(payload.verdict?.promotable === true, 'verdict.promotable === true');
158
-
159
- // ──────────────────────────────────────────────────────────────────
160
- // PHASE 4 — Strict mode behavior.
161
- // ──────────────────────────────────────────────────────────────────
162
- console.log('\nPhase 4 — verify --strict semantics');
163
-
164
- const strictOk = spawnSync('node', [analyzer, '--input', jsonlPath, '--strict'], { encoding: 'utf-8', timeout: 30_000 });
165
- assert(strictOk.status === 0, '--strict on promotable fixture exits 0');
166
-
167
- // Synthesize a non-promotable fixture (quality gain only +1%, below 2% threshold).
168
- const badPath = join(fixture, 'non-promotable.jsonl');
169
- const badRows = [];
170
- for (let i = 0; i < 40; i++) {
171
- const q = 0.80;
172
- badRows.push({
173
- v: 1, type: 'pair',
174
- ts: new Date().toISOString(),
175
- task_hash: `bad-${i.toString(16).padStart(8, '0')}`,
176
- bandit: { pick: 'sonnet', predictedQuality: q, predictedCostUsd: 0.005, backend: 'thompson-bandit' },
177
- ser: { pick: 'sonnet', predictedQuality: q + 0.005, predictedCostUsd: 0.005, backend: 'bandit-only' },
178
- outcome: { actualModel: 'sonnet', actualQuality: q, actualUsd: 0.005, actualLatencyMs: 1500 },
179
- });
180
- }
181
- writeFileSync(badPath, badRows.map((r) => JSON.stringify(r)).join('\n') + '\n');
182
- const strictBad = spawnSync('node', [analyzer, '--input', badPath, '--strict'], { encoding: 'utf-8', timeout: 30_000 });
183
- assert(strictBad.status === 1, '--strict on NON-promotable fixture exits 1 (got ' + strictBad.status + ')');
184
-
185
- // Cleanup
186
- if (process.env.TEST_PIPELINE_KEEP_FIXTURE !== '1') {
187
- rmSync(fixture, { recursive: true, force: true });
188
- } else {
189
- console.log(`\n(fixture kept at ${fixture})`);
190
- }
191
-
192
- console.log(`\n${passed} passed, ${failed} failed`);
193
- if (failed > 0) {
194
- console.log('\nFailures:');
195
- for (const f of failures) console.log(` - ${f}`);
196
- process.exit(1);
197
- }
198
- console.log('\n✓ End-to-end pipeline test PASSED.');
199
- }
200
-
201
- main().catch((e) => {
202
- console.error('test-parallel-pipeline crashed:', e.message || e);
203
- process.exit(2);
204
- });
1
+ #!/usr/bin/env node
2
+ // test-parallel-pipeline — end-to-end integration test for ADR-150's
3
+ // SelfEvolvingRouter parallel-logging pipeline.
4
+ //
5
+ // Composition under test:
6
+ // recordPair() / recordPairOutcome() [iter 11 — TS module]
7
+ // → .swarm/router-parallel.jsonl [JSONL trajectory]
8
+ // → router-parallel-analyze.mjs [iter 10 — analyzer]
9
+ // → 3-criteria AND-gate verdict [ADR-150 review-round-1]
10
+ //
11
+ // This script bypasses model-router.ts dispatch and calls the recorder
12
+ // directly, so it tests the recorder + analyzer composition WITHOUT
13
+ // requiring a built CLI dist. Iter-12 dispatch wiring is covered by
14
+ // the smoke step 17f source-grep.
15
+ //
16
+ // USAGE
17
+ // node scripts/test-parallel-pipeline.mjs # default
18
+ // TEST_PIPELINE_KEEP_FIXTURE=1 node scripts/test-parallel-pipeline.mjs
19
+ // # leave temp files for debugging
20
+ //
21
+ // EXIT CODES
22
+ // 0 all assertions passed
23
+ // 1 at least one assertion failed
24
+ // 2 setup error (e.g. tsc dist not buildable, recorder module missing)
25
+
26
+ import { spawnSync } from 'node:child_process';
27
+ import { mkdtempSync, rmSync, mkdirSync, existsSync, readFileSync, writeFileSync } from 'node:fs';
28
+ import { tmpdir } from 'node:os';
29
+ import { join, dirname } from 'node:path';
30
+ import { fileURLToPath } from 'node:url';
31
+
32
+ const SCRIPTS_DIR = dirname(fileURLToPath(import.meta.url));
33
+
34
+ let passed = 0, failed = 0;
35
+ const failures = [];
36
+
37
+ function assert(cond, label) {
38
+ if (cond) {
39
+ console.log(` ✓ ${label}`);
40
+ passed++;
41
+ } else {
42
+ console.log(` ✗ ${label}`);
43
+ failures.push(label);
44
+ failed++;
45
+ }
46
+ }
47
+
48
+ async function main() {
49
+ const fixture = mkdtempSync(join(tmpdir(), 'router-parallel-e2e-'));
50
+ const jsonlPath = join(fixture, 'router-parallel.jsonl');
51
+ console.log(`# router-parallel pipeline e2e (fixture: ${fixture})\n`);
52
+
53
+ // ──────────────────────────────────────────────────────────────────
54
+ // PHASE 1 — Synthesize a realistic JSONL trajectory by hand.
55
+ //
56
+ // We can't easily import the TS recorder without a build step, so we
57
+ // emit the same JSONL shape the recorder produces. This isolates the
58
+ // analyzer pipeline test from compile state.
59
+ // ──────────────────────────────────────────────────────────────────
60
+ console.log('Phase 1 — synthesize trajectory');
61
+
62
+ const rows = [];
63
+ // 40 paired decisions: SER beats bandit on quality (+5%) at +0.5% cost
64
+ // and +1% latency — should pass the 3-criteria AND-gate.
65
+ for (let i = 0; i < 40; i++) {
66
+ const q_actual = 0.78 + Math.random() * 0.04;
67
+ const u_actual = 0.005 + Math.random() * 0.001;
68
+ const l_actual = 1500 + Math.random() * 200;
69
+ const disagree = i % 2 === 0;
70
+ rows.push({
71
+ v: 1, type: 'pair',
72
+ ts: new Date(Date.now() - (40 - i) * 1000).toISOString(),
73
+ task_hash: `task-${i.toString(16).padStart(8, '0')}`,
74
+ bandit: {
75
+ pick: 'sonnet',
76
+ predictedQuality: q_actual - 0.02,
77
+ predictedCostUsd: u_actual,
78
+ backend: 'thompson-bandit',
79
+ },
80
+ ser: {
81
+ pick: disagree ? 'opus' : 'sonnet',
82
+ predictedQuality: disagree ? q_actual + 0.10 : q_actual + 0.01,
83
+ predictedCostUsd: disagree ? u_actual * 1.005 : u_actual,
84
+ backend: disagree ? 'metaharness-router-hybrid' : 'bandit-only',
85
+ },
86
+ outcome: {
87
+ actualModel: 'sonnet',
88
+ actualQuality: q_actual,
89
+ actualUsd: u_actual,
90
+ actualLatencyMs: Math.round(l_actual),
91
+ },
92
+ });
93
+ }
94
+
95
+ writeFileSync(jsonlPath, rows.map((r) => JSON.stringify(r)).join('\n') + '\n');
96
+ assert(existsSync(jsonlPath), 'fixture JSONL written');
97
+ assert(rows.length === 40, '40 pair rows synthesized');
98
+
99
+ // ──────────────────────────────────────────────────────────────────
100
+ // PHASE 2 — Verify recorder TS module compiles and exports the API.
101
+ // ──────────────────────────────────────────────────────────────────
102
+ console.log('\nPhase 2 — verify recorder TS module exports');
103
+
104
+ const recorderTs = join(SCRIPTS_DIR, '..', '..', '..', 'v3', '@claude-flow', 'cli', 'src', 'ruvector', 'router-parallel-recorder.ts');
105
+ const recorderSrc = readFileSync(recorderTs, 'utf-8');
106
+
107
+ assert(/export function recordPair\b/.test(recorderSrc), 'recordPair exported');
108
+ assert(/export function recordPairOutcome\b/.test(recorderSrc), 'recordPairOutcome exported');
109
+ assert(/export function parallelRecorderStatus\b/.test(recorderSrc), 'parallelRecorderStatus exported');
110
+ assert(/CLAUDE_FLOW_ROUTER_PARALLEL_LOG/.test(recorderSrc), 'env-gated by CLAUDE_FLOW_ROUTER_PARALLEL_LOG');
111
+ // ADR-150 rule #3 — never throw
112
+ assert(/never throws|never (throw|block)/i.test(recorderSrc), 'no-throw doc invariant present');
113
+ // Try/catch around fs operations
114
+ assert(/try \{[\s\S]*?appendFileSync[\s\S]*?\} catch/.test(recorderSrc), 'appendFileSync wrapped in try/catch');
115
+
116
+ // ──────────────────────────────────────────────────────────────────
117
+ // PHASE 3 — Run the analyzer against the synthesized JSONL.
118
+ // ──────────────────────────────────────────────────────────────────
119
+ console.log('\nPhase 3 — run analyzer against fixture');
120
+
121
+ const analyzer = join(SCRIPTS_DIR, 'router-parallel-analyze.mjs');
122
+ const r = spawnSync('node', [analyzer, '--input', jsonlPath, '--format', 'json'], {
123
+ encoding: 'utf-8',
124
+ timeout: 30_000,
125
+ });
126
+ assert(r.status === 0, `analyzer exit 0 (got ${r.status})`);
127
+
128
+ let payload;
129
+ try {
130
+ payload = JSON.parse(r.stdout);
131
+ } catch (e) {
132
+ console.error(' stdout:', r.stdout.slice(0, 400));
133
+ console.error(' stderr:', r.stderr.slice(0, 400));
134
+ assert(false, `analyzer output parseable as JSON (${e.message})`);
135
+ process.exit(2);
136
+ }
137
+
138
+ assert(payload.ok === true, 'payload.ok === true');
139
+ assert(payload.sufficient === true, 'payload.sufficient === true (n>=30)');
140
+ assert(payload.sampleSize === 40, `sampleSize === 40 (got ${payload.sampleSize})`);
141
+ assert(typeof payload.disagreement?.pct === 'number', 'disagreement.pct numeric');
142
+
143
+ // The 3 criteria must each be computed.
144
+ assert(typeof payload.criteria?.qualityImprovementPct === 'number', 'criteria.qualityImprovementPct present');
145
+ assert(typeof payload.criteria?.usdIncreasePct === 'number', 'criteria.usdIncreasePct present');
146
+ assert(typeof payload.criteria?.latencyIncreasePct === 'number', 'criteria.latencyIncreasePct present');
147
+
148
+ // Thresholds match ADR-150 review-round-1 exactly.
149
+ assert(payload.criteria?.qualityThresholdPct === 2, 'quality threshold = 2%');
150
+ assert(payload.criteria?.usdThresholdPct === 1, 'cost threshold = 1%');
151
+ assert(payload.criteria?.latencyThresholdPct === 5, 'latency threshold = 5%');
152
+
153
+ // For the promotable fixture, all three criteria should pass.
154
+ assert(payload.criteria?.qualityPasses === true, 'quality > 2% passes');
155
+ assert(payload.criteria?.costPasses === true, 'cost < 1% passes');
156
+ assert(payload.criteria?.latencyPasses === true, 'latency < 5% passes');
157
+ assert(payload.verdict?.promotable === true, 'verdict.promotable === true');
158
+
159
+ // ──────────────────────────────────────────────────────────────────
160
+ // PHASE 4 — Strict mode behavior.
161
+ // ──────────────────────────────────────────────────────────────────
162
+ console.log('\nPhase 4 — verify --strict semantics');
163
+
164
+ const strictOk = spawnSync('node', [analyzer, '--input', jsonlPath, '--strict'], { encoding: 'utf-8', timeout: 30_000 });
165
+ assert(strictOk.status === 0, '--strict on promotable fixture exits 0');
166
+
167
+ // Synthesize a non-promotable fixture (quality gain only +1%, below 2% threshold).
168
+ const badPath = join(fixture, 'non-promotable.jsonl');
169
+ const badRows = [];
170
+ for (let i = 0; i < 40; i++) {
171
+ const q = 0.80;
172
+ badRows.push({
173
+ v: 1, type: 'pair',
174
+ ts: new Date().toISOString(),
175
+ task_hash: `bad-${i.toString(16).padStart(8, '0')}`,
176
+ bandit: { pick: 'sonnet', predictedQuality: q, predictedCostUsd: 0.005, backend: 'thompson-bandit' },
177
+ ser: { pick: 'sonnet', predictedQuality: q + 0.005, predictedCostUsd: 0.005, backend: 'bandit-only' },
178
+ outcome: { actualModel: 'sonnet', actualQuality: q, actualUsd: 0.005, actualLatencyMs: 1500 },
179
+ });
180
+ }
181
+ writeFileSync(badPath, badRows.map((r) => JSON.stringify(r)).join('\n') + '\n');
182
+ const strictBad = spawnSync('node', [analyzer, '--input', badPath, '--strict'], { encoding: 'utf-8', timeout: 30_000 });
183
+ assert(strictBad.status === 1, '--strict on NON-promotable fixture exits 1 (got ' + strictBad.status + ')');
184
+
185
+ // Cleanup
186
+ if (process.env.TEST_PIPELINE_KEEP_FIXTURE !== '1') {
187
+ rmSync(fixture, { recursive: true, force: true });
188
+ } else {
189
+ console.log(`\n(fixture kept at ${fixture})`);
190
+ }
191
+
192
+ console.log(`\n${passed} passed, ${failed} failed`);
193
+ if (failed > 0) {
194
+ console.log('\nFailures:');
195
+ for (const f of failures) console.log(` - ${f}`);
196
+ process.exit(1);
197
+ }
198
+ console.log('\n✓ End-to-end pipeline test PASSED.');
199
+ }
200
+
201
+ main().catch((e) => {
202
+ console.error('test-parallel-pipeline crashed:', e.message || e);
203
+ process.exit(2);
204
+ });