@claude-flow/cli 3.10.45 → 3.11.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (560) 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/helpers/README.md +96 -96
  259. package/.claude/helpers/adr-compliance.sh +186 -186
  260. package/.claude/helpers/auto-commit.sh +178 -178
  261. package/.claude/helpers/auto-memory-hook.mjs +368 -368
  262. package/.claude/helpers/checkpoint-manager.sh +251 -251
  263. package/.claude/helpers/daemon-manager.sh +252 -252
  264. package/.claude/helpers/ddd-tracker.sh +144 -144
  265. package/.claude/helpers/github-safe.js +156 -156
  266. package/.claude/helpers/github-setup.sh +45 -45
  267. package/.claude/helpers/guidance-hook.sh +13 -13
  268. package/.claude/helpers/guidance-hooks.sh +102 -102
  269. package/.claude/helpers/health-monitor.sh +108 -108
  270. package/.claude/helpers/hook-handler.cjs +286 -286
  271. package/.claude/helpers/intelligence.cjs +1031 -1031
  272. package/.claude/helpers/learning-hooks.sh +329 -329
  273. package/.claude/helpers/learning-optimizer.sh +127 -127
  274. package/.claude/helpers/learning-service.mjs +1144 -1144
  275. package/.claude/helpers/memory.js +83 -83
  276. package/.claude/helpers/metrics-db.mjs +488 -488
  277. package/.claude/helpers/pattern-consolidator.sh +86 -86
  278. package/.claude/helpers/perf-worker.sh +160 -160
  279. package/.claude/helpers/post-commit +16 -16
  280. package/.claude/helpers/pre-commit +26 -26
  281. package/.claude/helpers/quick-start.sh +19 -19
  282. package/.claude/helpers/router.js +105 -105
  283. package/.claude/helpers/security-scanner.sh +127 -127
  284. package/.claude/helpers/session.js +157 -157
  285. package/.claude/helpers/setup-mcp.sh +18 -18
  286. package/.claude/helpers/standard-checkpoint-hooks.sh +189 -189
  287. package/.claude/helpers/statusline-hook.sh +21 -21
  288. package/.claude/helpers/statusline.cjs +604 -604
  289. package/.claude/helpers/statusline.js +352 -352
  290. package/.claude/helpers/swarm-comms.sh +353 -353
  291. package/.claude/helpers/swarm-hooks.sh +761 -761
  292. package/.claude/helpers/swarm-monitor.sh +210 -210
  293. package/.claude/helpers/sync-v3-metrics.sh +245 -245
  294. package/.claude/helpers/update-v3-progress.sh +165 -165
  295. package/.claude/helpers/v3-quick-status.sh +57 -57
  296. package/.claude/helpers/v3.sh +110 -110
  297. package/.claude/helpers/validate-v3-config.sh +215 -215
  298. package/.claude/helpers/worker-manager.sh +170 -170
  299. package/.claude/settings.json +182 -182
  300. package/.claude/skills/agentdb-advanced/SKILL.md +550 -550
  301. package/.claude/skills/agentdb-learning/SKILL.md +545 -545
  302. package/.claude/skills/agentdb-memory-patterns/SKILL.md +339 -339
  303. package/.claude/skills/agentdb-optimization/SKILL.md +509 -509
  304. package/.claude/skills/agentdb-vector-search/SKILL.md +339 -339
  305. package/.claude/skills/browser/SKILL.md +204 -204
  306. package/.claude/skills/dual-mode/README.md +71 -71
  307. package/.claude/skills/dual-mode/dual-collect.md +103 -103
  308. package/.claude/skills/dual-mode/dual-coordinate.md +85 -85
  309. package/.claude/skills/dual-mode/dual-spawn.md +81 -81
  310. package/.claude/skills/flow-nexus-neural/SKILL.md +727 -727
  311. package/.claude/skills/flow-nexus-platform/SKILL.md +1154 -1154
  312. package/.claude/skills/flow-nexus-swarm/SKILL.md +604 -604
  313. package/.claude/skills/github-code-review/SKILL.md +1125 -1125
  314. package/.claude/skills/github-multi-repo/SKILL.md +862 -862
  315. package/.claude/skills/github-project-management/SKILL.md +1262 -1262
  316. package/.claude/skills/github-release-management/SKILL.md +1064 -1064
  317. package/.claude/skills/github-workflow-automation/SKILL.md +1047 -1047
  318. package/.claude/skills/hooks-automation/SKILL.md +1201 -1201
  319. package/.claude/skills/pair-programming/SKILL.md +1202 -1202
  320. package/.claude/skills/reasoningbank-agentdb/SKILL.md +446 -446
  321. package/.claude/skills/reasoningbank-intelligence/SKILL.md +201 -201
  322. package/.claude/skills/skill-builder/SKILL.md +910 -910
  323. package/.claude/skills/sparc-methodology/SKILL.md +1106 -1106
  324. package/.claude/skills/stream-chain/SKILL.md +560 -560
  325. package/.claude/skills/swarm-advanced/SKILL.md +970 -970
  326. package/.claude/skills/swarm-orchestration/SKILL.md +179 -179
  327. package/.claude/skills/v3-cli-modernization/SKILL.md +871 -871
  328. package/.claude/skills/v3-core-implementation/SKILL.md +796 -796
  329. package/.claude/skills/v3-ddd-architecture/SKILL.md +441 -441
  330. package/.claude/skills/v3-integration-deep/SKILL.md +240 -240
  331. package/.claude/skills/v3-mcp-optimization/SKILL.md +776 -776
  332. package/.claude/skills/v3-memory-unification/SKILL.md +173 -173
  333. package/.claude/skills/v3-performance-optimization/SKILL.md +389 -389
  334. package/.claude/skills/v3-security-overhaul/SKILL.md +81 -81
  335. package/.claude/skills/v3-swarm-coordination/SKILL.md +339 -339
  336. package/.claude/skills/verification-quality/SKILL.md +691 -691
  337. package/README.md +412 -412
  338. package/bin/cli.js +314 -314
  339. package/bin/mcp-server.js +224 -224
  340. package/bin/preinstall.cjs +2 -2
  341. package/dist/src/benchmarks/capability-tasks.d.ts +126 -0
  342. package/dist/src/benchmarks/capability-tasks.d.ts.map +1 -0
  343. package/dist/src/benchmarks/capability-tasks.js +147 -0
  344. package/dist/src/benchmarks/capability-tasks.js.map +1 -0
  345. package/dist/src/benchmarks/gaia-agent-contrastive.d.ts +53 -0
  346. package/dist/src/benchmarks/gaia-agent-contrastive.d.ts.map +1 -0
  347. package/dist/src/benchmarks/gaia-agent-contrastive.js +177 -0
  348. package/dist/src/benchmarks/gaia-agent-contrastive.js.map +1 -0
  349. package/dist/src/benchmarks/gaia-agent-gemini.d.ts +51 -0
  350. package/dist/src/benchmarks/gaia-agent-gemini.d.ts.map +1 -0
  351. package/dist/src/benchmarks/gaia-agent-gemini.js +352 -0
  352. package/dist/src/benchmarks/gaia-agent-gemini.js.map +1 -0
  353. package/dist/src/benchmarks/gaia-claude-p.d.ts +105 -0
  354. package/dist/src/benchmarks/gaia-claude-p.d.ts.map +1 -0
  355. package/dist/src/benchmarks/gaia-claude-p.js +269 -0
  356. package/dist/src/benchmarks/gaia-claude-p.js.map +1 -0
  357. package/dist/src/benchmarks/gaia-claude-p.smoke.d.ts +20 -0
  358. package/dist/src/benchmarks/gaia-claude-p.smoke.d.ts.map +1 -0
  359. package/dist/src/benchmarks/gaia-claude-p.smoke.js +168 -0
  360. package/dist/src/benchmarks/gaia-claude-p.smoke.js.map +1 -0
  361. package/dist/src/benchmarks/gaia-codeagent-runner.py +556 -0
  362. package/dist/src/benchmarks/gaia-codeagent.d.ts +129 -0
  363. package/dist/src/benchmarks/gaia-codeagent.d.ts.map +1 -0
  364. package/dist/src/benchmarks/gaia-codeagent.js +590 -0
  365. package/dist/src/benchmarks/gaia-codeagent.js.map +1 -0
  366. package/dist/src/benchmarks/gaia-codeagent.smoke.d.ts +33 -0
  367. package/dist/src/benchmarks/gaia-codeagent.smoke.d.ts.map +1 -0
  368. package/dist/src/benchmarks/gaia-codeagent.smoke.js +241 -0
  369. package/dist/src/benchmarks/gaia-codeagent.smoke.js.map +1 -0
  370. package/dist/src/benchmarks/gaia-critic.js +24 -24
  371. package/dist/src/benchmarks/gaia-dag-pilot.d.ts +22 -0
  372. package/dist/src/benchmarks/gaia-dag-pilot.d.ts.map +1 -0
  373. package/dist/src/benchmarks/gaia-dag-pilot.js +158 -0
  374. package/dist/src/benchmarks/gaia-dag-pilot.js.map +1 -0
  375. package/dist/src/benchmarks/gaia-dag.d.ts +110 -0
  376. package/dist/src/benchmarks/gaia-dag.d.ts.map +1 -0
  377. package/dist/src/benchmarks/gaia-dag.js +658 -0
  378. package/dist/src/benchmarks/gaia-dag.js.map +1 -0
  379. package/dist/src/benchmarks/gaia-dag.smoke.d.ts +21 -0
  380. package/dist/src/benchmarks/gaia-dag.smoke.d.ts.map +1 -0
  381. package/dist/src/benchmarks/gaia-dag.smoke.js +290 -0
  382. package/dist/src/benchmarks/gaia-dag.smoke.js.map +1 -0
  383. package/dist/src/benchmarks/gaia-ensemble-pilot.d.ts +17 -0
  384. package/dist/src/benchmarks/gaia-ensemble-pilot.d.ts.map +1 -0
  385. package/dist/src/benchmarks/gaia-ensemble-pilot.js +106 -0
  386. package/dist/src/benchmarks/gaia-ensemble-pilot.js.map +1 -0
  387. package/dist/src/benchmarks/gaia-ensemble.d.ts +103 -0
  388. package/dist/src/benchmarks/gaia-ensemble.d.ts.map +1 -0
  389. package/dist/src/benchmarks/gaia-ensemble.js +462 -0
  390. package/dist/src/benchmarks/gaia-ensemble.js.map +1 -0
  391. package/dist/src/benchmarks/gaia-gemini-pilot.d.ts +17 -0
  392. package/dist/src/benchmarks/gaia-gemini-pilot.d.ts.map +1 -0
  393. package/dist/src/benchmarks/gaia-gemini-pilot.js +168 -0
  394. package/dist/src/benchmarks/gaia-gemini-pilot.js.map +1 -0
  395. package/dist/src/benchmarks/gaia-mode-router.d.ts +67 -0
  396. package/dist/src/benchmarks/gaia-mode-router.d.ts.map +1 -0
  397. package/dist/src/benchmarks/gaia-mode-router.js +126 -0
  398. package/dist/src/benchmarks/gaia-mode-router.js.map +1 -0
  399. package/dist/src/benchmarks/gaia-mode-router.smoke.d.ts +15 -0
  400. package/dist/src/benchmarks/gaia-mode-router.smoke.d.ts.map +1 -0
  401. package/dist/src/benchmarks/gaia-mode-router.smoke.js +131 -0
  402. package/dist/src/benchmarks/gaia-mode-router.smoke.js.map +1 -0
  403. package/dist/src/benchmarks/gaia-tools/grounded_query.smoke.d.ts +24 -0
  404. package/dist/src/benchmarks/gaia-tools/grounded_query.smoke.d.ts.map +1 -0
  405. package/dist/src/benchmarks/gaia-tools/grounded_query.smoke.js +276 -0
  406. package/dist/src/benchmarks/gaia-tools/grounded_query.smoke.js.map +1 -0
  407. package/dist/src/benchmarks/gaia-tools/image_describe.d.ts +45 -0
  408. package/dist/src/benchmarks/gaia-tools/image_describe.d.ts.map +1 -0
  409. package/dist/src/benchmarks/gaia-tools/image_describe.js +222 -0
  410. package/dist/src/benchmarks/gaia-tools/image_describe.js.map +1 -0
  411. package/dist/src/benchmarks/gaia-tools/image_describe.smoke.d.ts +24 -0
  412. package/dist/src/benchmarks/gaia-tools/image_describe.smoke.d.ts.map +1 -0
  413. package/dist/src/benchmarks/gaia-tools/image_describe.smoke.js +169 -0
  414. package/dist/src/benchmarks/gaia-tools/image_describe.smoke.js.map +1 -0
  415. package/dist/src/benchmarks/gaia-tools/pdf_read.d.ts +27 -0
  416. package/dist/src/benchmarks/gaia-tools/pdf_read.d.ts.map +1 -0
  417. package/dist/src/benchmarks/gaia-tools/pdf_read.js +226 -0
  418. package/dist/src/benchmarks/gaia-tools/pdf_read.js.map +1 -0
  419. package/dist/src/benchmarks/gaia-tools/python_exec.d.ts +31 -0
  420. package/dist/src/benchmarks/gaia-tools/python_exec.d.ts.map +1 -0
  421. package/dist/src/benchmarks/gaia-tools/python_exec.js +94 -0
  422. package/dist/src/benchmarks/gaia-tools/python_exec.js.map +1 -0
  423. package/dist/src/benchmarks/gaia-tools/python_exec.smoke.d.ts +13 -0
  424. package/dist/src/benchmarks/gaia-tools/python_exec.smoke.d.ts.map +1 -0
  425. package/dist/src/benchmarks/gaia-tools/python_exec.smoke.js +98 -0
  426. package/dist/src/benchmarks/gaia-tools/python_exec.smoke.js.map +1 -0
  427. package/dist/src/benchmarks/gaia-tools/visit_webpage.d.ts +52 -0
  428. package/dist/src/benchmarks/gaia-tools/visit_webpage.d.ts.map +1 -0
  429. package/dist/src/benchmarks/gaia-tools/visit_webpage.js +304 -0
  430. package/dist/src/benchmarks/gaia-tools/visit_webpage.js.map +1 -0
  431. package/dist/src/benchmarks/gaia-tools/web_browse.d.ts +61 -0
  432. package/dist/src/benchmarks/gaia-tools/web_browse.d.ts.map +1 -0
  433. package/dist/src/benchmarks/gaia-tools/web_browse.js +217 -0
  434. package/dist/src/benchmarks/gaia-tools/web_browse.js.map +1 -0
  435. package/dist/src/benchmarks/gaia-tools/web_browse.smoke.d.ts +26 -0
  436. package/dist/src/benchmarks/gaia-tools/web_browse.smoke.d.ts.map +1 -0
  437. package/dist/src/benchmarks/gaia-tools/web_browse.smoke.js +152 -0
  438. package/dist/src/benchmarks/gaia-tools/web_browse.smoke.js.map +1 -0
  439. package/dist/src/benchmarks/gaia-tools/web_search.smoke.d.ts +24 -0
  440. package/dist/src/benchmarks/gaia-tools/web_search.smoke.d.ts.map +1 -0
  441. package/dist/src/benchmarks/gaia-tools/web_search.smoke.js +311 -0
  442. package/dist/src/benchmarks/gaia-tools/web_search.smoke.js.map +1 -0
  443. package/dist/src/benchmarks/gaia-visit-webpage.smoke.d.ts +25 -0
  444. package/dist/src/benchmarks/gaia-visit-webpage.smoke.d.ts.map +1 -0
  445. package/dist/src/benchmarks/gaia-visit-webpage.smoke.js +304 -0
  446. package/dist/src/benchmarks/gaia-visit-webpage.smoke.js.map +1 -0
  447. package/dist/src/commands/completions.js +409 -409
  448. package/dist/src/commands/daemon.js +44 -44
  449. package/dist/src/commands/embeddings.js +26 -26
  450. package/dist/src/commands/hive-mind.js +97 -97
  451. package/dist/src/commands/hooks.js +9 -9
  452. package/dist/src/commands/neural.d.ts.map +1 -1
  453. package/dist/src/commands/neural.js +2351 -1
  454. package/dist/src/commands/neural.js.map +1 -1
  455. package/dist/src/commands/performance-capability.d.ts +30 -0
  456. package/dist/src/commands/performance-capability.d.ts.map +1 -0
  457. package/dist/src/commands/performance-capability.js +349 -0
  458. package/dist/src/commands/performance-capability.js.map +1 -0
  459. package/dist/src/commands/ruvector/backup.js +23 -23
  460. package/dist/src/commands/ruvector/benchmark.js +31 -31
  461. package/dist/src/commands/ruvector/import.js +14 -14
  462. package/dist/src/commands/ruvector/init.js +115 -115
  463. package/dist/src/commands/ruvector/migrate.js +99 -99
  464. package/dist/src/commands/ruvector/optimize.js +51 -51
  465. package/dist/src/commands/ruvector/setup.js +624 -624
  466. package/dist/src/commands/ruvector/status.js +38 -38
  467. package/dist/src/commands/swarm.d.ts.map +1 -1
  468. package/dist/src/commands/swarm.js +5 -1
  469. package/dist/src/commands/swarm.js.map +1 -1
  470. package/dist/src/init/claudemd-generator.js +231 -231
  471. package/dist/src/init/executor.d.ts.map +1 -1
  472. package/dist/src/init/executor.js +487 -455
  473. package/dist/src/init/executor.js.map +1 -1
  474. package/dist/src/init/helpers-generator.js +727 -727
  475. package/dist/src/init/statusline-generator.js +717 -717
  476. package/dist/src/mcp-tools/agent-execute-core.d.ts +23 -1
  477. package/dist/src/mcp-tools/agent-execute-core.d.ts.map +1 -1
  478. package/dist/src/mcp-tools/agent-execute-core.js +136 -9
  479. package/dist/src/mcp-tools/agent-execute-core.js.map +1 -1
  480. package/dist/src/mcp-tools/agent-tools.d.ts.map +1 -1
  481. package/dist/src/mcp-tools/agent-tools.js +48 -4
  482. package/dist/src/mcp-tools/agent-tools.js.map +1 -1
  483. package/dist/src/mcp-tools/agentdb-tools.js +15 -15
  484. package/dist/src/mcp-tools/hooks-tools.d.ts.map +1 -1
  485. package/dist/src/mcp-tools/hooks-tools.js +196 -0
  486. package/dist/src/mcp-tools/hooks-tools.js.map +1 -1
  487. package/dist/src/memory/graph-edge-writer.js +18 -18
  488. package/dist/src/memory/memory-bridge.js +84 -84
  489. package/dist/src/memory/memory-initializer.js +393 -393
  490. package/dist/src/memory/rabitq-index.js +5 -5
  491. package/dist/src/runtime/headless.js +28 -28
  492. package/dist/src/ruvector/enhanced-model-router.d.ts +5 -0
  493. package/dist/src/ruvector/enhanced-model-router.d.ts.map +1 -1
  494. package/dist/src/ruvector/enhanced-model-router.js +18 -1
  495. package/dist/src/ruvector/enhanced-model-router.js.map +1 -1
  496. package/dist/src/ruvector/model-prices.d.ts +50 -0
  497. package/dist/src/ruvector/model-prices.d.ts.map +1 -0
  498. package/dist/src/ruvector/model-prices.js +72 -0
  499. package/dist/src/ruvector/model-prices.js.map +1 -0
  500. package/dist/src/ruvector/model-router.d.ts +147 -8
  501. package/dist/src/ruvector/model-router.d.ts.map +1 -1
  502. package/dist/src/ruvector/model-router.js +418 -18
  503. package/dist/src/ruvector/model-router.js.map +1 -1
  504. package/dist/src/ruvector/neural-router.d.ts +172 -39
  505. package/dist/src/ruvector/neural-router.d.ts.map +1 -1
  506. package/dist/src/ruvector/neural-router.js +780 -91
  507. package/dist/src/ruvector/neural-router.js.map +1 -1
  508. package/dist/src/ruvector/router-calibrator.d.ts +65 -0
  509. package/dist/src/ruvector/router-calibrator.d.ts.map +1 -0
  510. package/dist/src/ruvector/router-calibrator.js +120 -0
  511. package/dist/src/ruvector/router-calibrator.js.map +1 -0
  512. package/dist/src/ruvector/router-trajectory.d.ts +137 -45
  513. package/dist/src/ruvector/router-trajectory.d.ts.map +1 -1
  514. package/dist/src/ruvector/router-trajectory.js +226 -63
  515. package/dist/src/ruvector/router-trajectory.js.map +1 -1
  516. package/dist/src/ruvector/task-embedder.d.ts +56 -0
  517. package/dist/src/ruvector/task-embedder.d.ts.map +1 -0
  518. package/dist/src/ruvector/task-embedder.js +237 -0
  519. package/dist/src/ruvector/task-embedder.js.map +1 -0
  520. package/dist/src/services/container-worker-pool.d.ts +8 -1
  521. package/dist/src/services/container-worker-pool.d.ts.map +1 -1
  522. package/dist/src/services/container-worker-pool.js +9 -3
  523. package/dist/src/services/container-worker-pool.js.map +1 -1
  524. package/dist/src/services/headless-worker-executor.js +84 -84
  525. package/dist/src/transfer/deploy-seraphine.js +23 -23
  526. package/dist/tsconfig.tsbuildinfo +1 -1
  527. package/package.json +136 -134
  528. package/scripts/beir-bootstrap-significance.mjs +162 -162
  529. package/scripts/bench-rvagent.mjs +312 -312
  530. package/scripts/benchmark-codemods.mjs +127 -127
  531. package/scripts/benchmark-cross-repo.mjs +206 -206
  532. package/scripts/benchmark-pretrained-retrieval.mjs +305 -305
  533. package/scripts/benchmark-self-learning.mjs +241 -241
  534. package/scripts/benchmark-trajectory-mrr.mjs +191 -191
  535. package/scripts/deploy-ipfs-node.sh +153 -153
  536. package/scripts/grid-search-retrieval.mjs +216 -216
  537. package/scripts/postinstall.cjs +153 -153
  538. package/scripts/pretrain-from-github.mjs +358 -358
  539. package/scripts/publish-registry.ts +345 -345
  540. package/scripts/publish.sh +57 -57
  541. package/scripts/run-beir-bge.mjs +324 -324
  542. package/scripts/run-beir-hybrid.mjs +261 -261
  543. package/scripts/run-beir-lucene-bm25.mjs +154 -154
  544. package/scripts/run-beir-nfcorpus.mjs +231 -231
  545. package/scripts/run-beir-rrf-ablation.mjs +221 -221
  546. package/scripts/setup-ipfs-registry.md +366 -366
  547. package/dist/src/ruvector/flash-attention.d.ts +0 -195
  548. package/dist/src/ruvector/flash-attention.d.ts.map +0 -1
  549. package/dist/src/ruvector/flash-attention.js +0 -643
  550. package/dist/src/ruvector/flash-attention.js.map +0 -1
  551. package/dist/src/ruvector/moe-router.d.ts +0 -206
  552. package/dist/src/ruvector/moe-router.d.ts.map +0 -1
  553. package/dist/src/ruvector/moe-router.js +0 -626
  554. package/dist/src/ruvector/moe-router.js.map +0 -1
  555. package/dist/src/services/event-stream.d.ts +0 -25
  556. package/dist/src/services/event-stream.js +0 -27
  557. package/dist/src/services/loop-worker-runner.d.ts +0 -16
  558. package/dist/src/services/loop-worker-runner.js +0 -34
  559. package/dist/src/services/runtime-capabilities.d.ts +0 -22
  560. package/dist/src/services/runtime-capabilities.js +0 -45
@@ -0,0 +1,98 @@
1
+ /**
2
+ * Smoke tests for python_exec — ADR-133-PR4
3
+ *
4
+ * Run with:
5
+ * cd v3/@claude-flow/cli
6
+ * npx tsx src/benchmarks/gaia-tools/python_exec.smoke.ts
7
+ *
8
+ * Expects python3 to be installed and on PATH.
9
+ *
10
+ * Refs: ADR-133, #2156
11
+ */
12
+ import { createPythonExecTool } from './python_exec.js';
13
+ async function runTest(name, fn) {
14
+ try {
15
+ await fn();
16
+ return { name, passed: true, message: 'OK' };
17
+ }
18
+ catch (e) {
19
+ return { name, passed: false, message: String(e) };
20
+ }
21
+ }
22
+ function assert(condition, msg) {
23
+ if (!condition)
24
+ throw new Error(`Assertion failed: ${msg}`);
25
+ }
26
+ // ---------------------------------------------------------------------------
27
+ // Tests
28
+ // ---------------------------------------------------------------------------
29
+ const tool = createPythonExecTool();
30
+ const tests = [
31
+ runTest('Test 1: basic arithmetic — print(2+2)', async () => {
32
+ const out = await tool.execute({ code: 'print(2+2)' });
33
+ assert(out.includes('4'), `Expected stdout to contain "4", got:\n${out}`);
34
+ assert(out.includes('exit_code: 0'), `Expected exit_code 0, got:\n${out}`);
35
+ }),
36
+ runTest('Test 2: large computation — sum(range(1000000))', async () => {
37
+ const out = await tool.execute({ code: 'print(sum(range(1000000)))' });
38
+ assert(out.includes('499999500000'), `Expected "499999500000", got:\n${out}`);
39
+ assert(out.includes('exit_code: 0'), `Expected exit_code 0, got:\n${out}`);
40
+ }),
41
+ runTest('Test 3: stdlib import — math.sqrt(144)', async () => {
42
+ const out = await tool.execute({
43
+ code: 'import math; print(math.sqrt(144))',
44
+ });
45
+ assert(out.includes('12.0'), `Expected "12.0", got:\n${out}`);
46
+ assert(out.includes('exit_code: 0'), `Expected exit_code 0, got:\n${out}`);
47
+ }),
48
+ runTest('Test 4: timeout — infinite loop with 2s limit', async () => {
49
+ const start = Date.now();
50
+ const out = await tool.execute({
51
+ code: 'while True: pass',
52
+ timeout_seconds: 2,
53
+ });
54
+ const elapsed = Date.now() - start;
55
+ assert(elapsed < 6_000, `Timeout should have fired within ~2s, took ${elapsed}ms`);
56
+ const hasTimedOut = out.includes('timed out') ||
57
+ out.includes('exit_code: 124') ||
58
+ // SIGKILL on some platforms gives exit code -9 or 137
59
+ out.includes('exit_code: -9') ||
60
+ out.includes('exit_code: 137');
61
+ assert(hasTimedOut, `Expected timeout signal, got:\n${out}`);
62
+ }),
63
+ runTest('Test 5: syntax error — exit_code != 0, stderr contains SyntaxError', async () => {
64
+ const out = await tool.execute({ code: 'x = (' });
65
+ assert(out.includes('SyntaxError'), `Expected SyntaxError in output, got:\n${out}`);
66
+ const hasNonZeroExit = !out.includes('exit_code: 0') || out.includes('stderr:');
67
+ assert(hasNonZeroExit, `Expected non-zero exit code or stderr for syntax error, got:\n${out}`);
68
+ }),
69
+ ];
70
+ // ---------------------------------------------------------------------------
71
+ // Runner
72
+ // ---------------------------------------------------------------------------
73
+ async function main() {
74
+ console.log('python_exec smoke tests — ADR-133-PR4\n');
75
+ const results = await Promise.all(tests);
76
+ let passed = 0;
77
+ let failed = 0;
78
+ for (const r of results) {
79
+ const icon = r.passed ? 'PASS' : 'FAIL';
80
+ console.log(`[${icon}] ${r.name}`);
81
+ if (!r.passed) {
82
+ console.log(` ${r.message}`);
83
+ }
84
+ if (r.passed)
85
+ passed++;
86
+ else
87
+ failed++;
88
+ }
89
+ console.log(`\nResults: ${passed}/${results.length} passed`);
90
+ if (failed > 0) {
91
+ process.exit(1);
92
+ }
93
+ }
94
+ main().catch((e) => {
95
+ console.error('Fatal:', e);
96
+ process.exit(1);
97
+ });
98
+ //# sourceMappingURL=python_exec.smoke.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"python_exec.smoke.js","sourceRoot":"","sources":["../../../../src/benchmarks/gaia-tools/python_exec.smoke.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;GAUG;AAEH,OAAO,EAAE,oBAAoB,EAAE,MAAM,kBAAkB,CAAC;AAYxD,KAAK,UAAU,OAAO,CACpB,IAAY,EACZ,EAAuB;IAEvB,IAAI,CAAC;QACH,MAAM,EAAE,EAAE,CAAC;QACX,OAAO,EAAE,IAAI,EAAE,MAAM,EAAE,IAAI,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC;IAC/C,CAAC;IAAC,OAAO,CAAU,EAAE,CAAC;QACpB,OAAO,EAAE,IAAI,EAAE,MAAM,EAAE,KAAK,EAAE,OAAO,EAAE,MAAM,CAAC,CAAC,CAAC,EAAE,CAAC;IACrD,CAAC;AACH,CAAC;AAED,SAAS,MAAM,CAAC,SAAkB,EAAE,GAAW;IAC7C,IAAI,CAAC,SAAS;QAAE,MAAM,IAAI,KAAK,CAAC,qBAAqB,GAAG,EAAE,CAAC,CAAC;AAC9D,CAAC;AAED,8EAA8E;AAC9E,QAAQ;AACR,8EAA8E;AAE9E,MAAM,IAAI,GAAG,oBAAoB,EAAE,CAAC;AAEpC,MAAM,KAAK,GAAG;IACZ,OAAO,CAAC,uCAAuC,EAAE,KAAK,IAAI,EAAE;QAC1D,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC,EAAE,IAAI,EAAE,YAAY,EAAE,CAAC,CAAC;QACvD,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,GAAG,CAAC,EAAE,yCAAyC,GAAG,EAAE,CAAC,CAAC;QAC1E,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,EAAE,+BAA+B,GAAG,EAAE,CAAC,CAAC;IAC7E,CAAC,CAAC;IAEF,OAAO,CAAC,iDAAiD,EAAE,KAAK,IAAI,EAAE;QACpE,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC,EAAE,IAAI,EAAE,4BAA4B,EAAE,CAAC,CAAC;QACvE,MAAM,CACJ,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,EAC5B,kCAAkC,GAAG,EAAE,CACxC,CAAC;QACF,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,EAAE,+BAA+B,GAAG,EAAE,CAAC,CAAC;IAC7E,CAAC,CAAC;IAEF,OAAO,CAAC,wCAAwC,EAAE,KAAK,IAAI,EAAE;QAC3D,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC;YAC7B,IAAI,EAAE,oCAAoC;SAC3C,CAAC,CAAC;QACH,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,MAAM,CAAC,EAAE,0BAA0B,GAAG,EAAE,CAAC,CAAC;QAC9D,MAAM,CAAC,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,EAAE,+BAA+B,GAAG,EAAE,CAAC,CAAC;IAC7E,CAAC,CAAC;IAEF,OAAO,CAAC,+CAA+C,EAAE,KAAK,IAAI,EAAE;QAClE,MAAM,KAAK,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QACzB,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC;YAC7B,IAAI,EAAE,kBAAkB;YACxB,eAAe,EAAE,CAAC;SACnB,CAAC,CAAC;QACH,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,GAAG,KAAK,CAAC;QACnC,MAAM,CACJ,OAAO,GAAG,KAAK,EACf,8CAA8C,OAAO,IAAI,CAC1D,CAAC;QACF,MAAM,WAAW,GACf,GAAG,CAAC,QAAQ,CAAC,WAAW,CAAC;YACzB,GAAG,CAAC,QAAQ,CAAC,gBAAgB,CAAC;YAC9B,sDAAsD;YACtD,GAAG,CAAC,QAAQ,CAAC,eAAe,CAAC;YAC7B,GAAG,CAAC,QAAQ,CAAC,gBAAgB,CAAC,CAAC;QACjC,MAAM,CAAC,WAAW,EAAE,kCAAkC,GAAG,EAAE,CAAC,CAAC;IAC/D,CAAC,CAAC;IAEF,OAAO,CAAC,oEAAoE,EAAE,KAAK,IAAI,EAAE;QACvF,MAAM,GAAG,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC,EAAE,IAAI,EAAE,OAAO,EAAE,CAAC,CAAC;QAClD,MAAM,CACJ,GAAG,CAAC,QAAQ,CAAC,aAAa,CAAC,EAC3B,yCAAyC,GAAG,EAAE,CAC/C,CAAC;QACF,MAAM,cAAc,GAClB,CAAC,GAAG,CAAC,QAAQ,CAAC,cAAc,CAAC,IAAI,GAAG,CAAC,QAAQ,CAAC,SAAS,CAAC,CAAC;QAC3D,MAAM,CACJ,cAAc,EACd,iEAAiE,GAAG,EAAE,CACvE,CAAC;IACJ,CAAC,CAAC;CACH,CAAC;AAEF,8EAA8E;AAC9E,SAAS;AACT,8EAA8E;AAE9E,KAAK,UAAU,IAAI;IACjB,OAAO,CAAC,GAAG,CAAC,yCAAyC,CAAC,CAAC;IAEvD,MAAM,OAAO,GAAG,MAAM,OAAO,CAAC,GAAG,CAAC,KAAK,CAAC,CAAC;IACzC,IAAI,MAAM,GAAG,CAAC,CAAC;IACf,IAAI,MAAM,GAAG,CAAC,CAAC;IAEf,KAAK,MAAM,CAAC,IAAI,OAAO,EAAE,CAAC;QACxB,MAAM,IAAI,GAAG,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,CAAC;QACxC,OAAO,CAAC,GAAG,CAAC,IAAI,IAAI,KAAK,CAAC,CAAC,IAAI,EAAE,CAAC,CAAC;QACnC,IAAI,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC;YACd,OAAO,CAAC,GAAG,CAAC,UAAU,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC;QACrC,CAAC;QACD,IAAI,CAAC,CAAC,MAAM;YAAE,MAAM,EAAE,CAAC;;YAClB,MAAM,EAAE,CAAC;IAChB,CAAC;IAED,OAAO,CAAC,GAAG,CAAC,cAAc,MAAM,IAAI,OAAO,CAAC,MAAM,SAAS,CAAC,CAAC;IAE7D,IAAI,MAAM,GAAG,CAAC,EAAE,CAAC;QACf,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IAClB,CAAC;AACH,CAAC;AAED,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,EAAE;IACjB,OAAO,CAAC,KAAK,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC;IAC3B,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;AAClB,CAAC,CAAC,CAAC"}
@@ -0,0 +1,52 @@
1
+ /**
2
+ * GAIA Tool: visit_webpage — iter-57 (upgraded from iter-54 base)
3
+ *
4
+ * Fetches the FULL readable text content of a webpage URL — not just snippets.
5
+ * This closes the HAL parity gap identified in iter-51 surrender analysis
6
+ * (~25-35% of L1 questions need full page reading that snippet-only tools miss).
7
+ *
8
+ * Upgrade over iter-54 base (iter-57 changes):
9
+ * - max_chars input parameter (default 50k, was hardcoded 8k)
10
+ * - Test hooks interface (visitWebpageTestHooks) for unit tests without live HTTP
11
+ * - URL validation with helpful error messages
12
+ * - Link extraction (up to 30 links) — enables follow-up navigation
13
+ * - Better HTML extraction: content-block heuristics (main/article selection)
14
+ * - AIDefence PII gate (optional dep, graceful skip if not installed)
15
+ * - Structured VisitWebpageResult type + diagnostic stderr logging
16
+ * - Python bs4 primary path preserved (already in benchmark env)
17
+ *
18
+ * Extraction pipeline:
19
+ * 1. fetch() + Python bs4.get_text() (primary — better quality, handles encoding)
20
+ * 2. Regex-based fallback (no external dep) if bs4 unavailable or output too short
21
+ * 3. Content passed through AIDefence PII gate before returning to agent
22
+ *
23
+ * Refs: ADR-138, ADR-133, iter-54, iter-57, #2156
24
+ */
25
+ import { GaiaTool, ToolDefinition } from './types.js';
26
+ export interface VisitWebpageResult {
27
+ title: string;
28
+ content: string;
29
+ links: Array<{
30
+ href: string;
31
+ text: string;
32
+ }>;
33
+ sources_consulted: string[];
34
+ chars_returned: number;
35
+ truncated: boolean;
36
+ }
37
+ export interface VisitWebpageTestHooks {
38
+ /** Override the HTTP fetch so tests never make live calls. */
39
+ fetchHtml?: (url: string) => Promise<{
40
+ html: string;
41
+ finalUrl: string;
42
+ }>;
43
+ }
44
+ export declare const visitWebpageTestHooks: VisitWebpageTestHooks;
45
+ export declare function visitWebpage(url: string, maxChars?: number): Promise<VisitWebpageResult>;
46
+ export declare class VisitWebpageTool implements GaiaTool {
47
+ readonly name = "visit_webpage";
48
+ readonly definition: ToolDefinition;
49
+ execute(input: Record<string, unknown>): Promise<string>;
50
+ }
51
+ export declare function createVisitWebpageTool(): VisitWebpageTool;
52
+ //# sourceMappingURL=visit_webpage.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"visit_webpage.d.ts","sourceRoot":"","sources":["../../../../src/benchmarks/gaia-tools/visit_webpage.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AAGH,OAAO,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,YAAY,CAAC;AAiBtD,MAAM,WAAW,kBAAkB;IACjC,KAAK,EAAE,MAAM,CAAC;IACd,OAAO,EAAE,MAAM,CAAC;IAChB,KAAK,EAAE,KAAK,CAAC;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,IAAI,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;IAC7C,iBAAiB,EAAE,MAAM,EAAE,CAAC;IAC5B,cAAc,EAAE,MAAM,CAAC;IACvB,SAAS,EAAE,OAAO,CAAC;CACpB;AAMD,MAAM,WAAW,qBAAqB;IACpC,8DAA8D;IAC9D,SAAS,CAAC,EAAE,CAAC,GAAG,EAAE,MAAM,KAAK,OAAO,CAAC;QAAE,IAAI,EAAE,MAAM,CAAC;QAAC,QAAQ,EAAE,MAAM,CAAA;KAAE,CAAC,CAAC;CAC1E;AAED,eAAO,MAAM,qBAAqB,EAAE,qBAA0B,CAAC;AAkL/D,wBAAsB,YAAY,CAChC,GAAG,EAAE,MAAM,EACX,QAAQ,GAAE,MAA0B,GACnC,OAAO,CAAC,kBAAkB,CAAC,CA+B7B;AA6BD,qBAAa,gBAAiB,YAAW,QAAQ;IAC/C,QAAQ,CAAC,IAAI,mBAAmB;IAEhC,QAAQ,CAAC,UAAU,EAAE,cAAc,CAwBjC;IAEI,OAAO,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,OAAO,CAAC,MAAM,CAAC;CAyB/D;AAMD,wBAAgB,sBAAsB,IAAI,gBAAgB,CAEzD"}
@@ -0,0 +1,304 @@
1
+ /**
2
+ * GAIA Tool: visit_webpage — iter-57 (upgraded from iter-54 base)
3
+ *
4
+ * Fetches the FULL readable text content of a webpage URL — not just snippets.
5
+ * This closes the HAL parity gap identified in iter-51 surrender analysis
6
+ * (~25-35% of L1 questions need full page reading that snippet-only tools miss).
7
+ *
8
+ * Upgrade over iter-54 base (iter-57 changes):
9
+ * - max_chars input parameter (default 50k, was hardcoded 8k)
10
+ * - Test hooks interface (visitWebpageTestHooks) for unit tests without live HTTP
11
+ * - URL validation with helpful error messages
12
+ * - Link extraction (up to 30 links) — enables follow-up navigation
13
+ * - Better HTML extraction: content-block heuristics (main/article selection)
14
+ * - AIDefence PII gate (optional dep, graceful skip if not installed)
15
+ * - Structured VisitWebpageResult type + diagnostic stderr logging
16
+ * - Python bs4 primary path preserved (already in benchmark env)
17
+ *
18
+ * Extraction pipeline:
19
+ * 1. fetch() + Python bs4.get_text() (primary — better quality, handles encoding)
20
+ * 2. Regex-based fallback (no external dep) if bs4 unavailable or output too short
21
+ * 3. Content passed through AIDefence PII gate before returning to agent
22
+ *
23
+ * Refs: ADR-138, ADR-133, iter-54, iter-57, #2156
24
+ */
25
+ import { execFileSync } from 'node:child_process';
26
+ // ---------------------------------------------------------------------------
27
+ // Constants
28
+ // ---------------------------------------------------------------------------
29
+ const DEFAULT_MAX_CHARS = 50_000;
30
+ const MIN_CONTENT_CHARS = 50;
31
+ const REQUEST_TIMEOUT_MS = 30_000;
32
+ const MAX_LINKS = 30;
33
+ const UA = 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/124.0.0.0 Safari/537.36';
34
+ export const visitWebpageTestHooks = {};
35
+ // ---------------------------------------------------------------------------
36
+ // URL validation
37
+ // ---------------------------------------------------------------------------
38
+ function validateUrl(url) {
39
+ if (!url || typeof url !== 'string') {
40
+ throw new Error('visit_webpage: `url` must be a non-empty string.');
41
+ }
42
+ if (!url.startsWith('http://') && !url.startsWith('https://')) {
43
+ throw new Error(`visit_webpage: URL must start with http:// or https://. Got: "${url}"`);
44
+ }
45
+ try {
46
+ new URL(url);
47
+ }
48
+ catch {
49
+ throw new Error(`visit_webpage: invalid URL: "${url}"`);
50
+ }
51
+ }
52
+ // ---------------------------------------------------------------------------
53
+ // Page fetcher (uses Node built-in fetch — Node 18+)
54
+ // ---------------------------------------------------------------------------
55
+ async function fetchPage(url) {
56
+ const controller = new AbortController();
57
+ const timer = setTimeout(() => controller.abort(), REQUEST_TIMEOUT_MS);
58
+ let res;
59
+ try {
60
+ res = await fetch(url, {
61
+ headers: {
62
+ 'User-Agent': UA,
63
+ Accept: 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
64
+ 'Accept-Language': 'en-US,en;q=0.9',
65
+ },
66
+ signal: controller.signal,
67
+ redirect: 'follow',
68
+ });
69
+ }
70
+ finally {
71
+ clearTimeout(timer);
72
+ }
73
+ if (!res.ok) {
74
+ throw new Error(`visit_webpage: HTTP ${res.status} for ${url}`);
75
+ }
76
+ const ct = res.headers.get('content-type') ?? '';
77
+ if (ct.includes('application/pdf')) {
78
+ throw new Error(`visit_webpage: URL returns a PDF. Use file_read with the downloaded path instead. URL: ${url}`);
79
+ }
80
+ const html = await res.text();
81
+ const finalUrl = res.url || url;
82
+ return { html, finalUrl };
83
+ }
84
+ // ---------------------------------------------------------------------------
85
+ // HTML extraction helpers
86
+ // ---------------------------------------------------------------------------
87
+ /**
88
+ * Extract title from raw HTML.
89
+ */
90
+ function extractTitle(html) {
91
+ const match = /<title[^>]*>([\s\S]*?)<\/title>/i.exec(html);
92
+ if (!match)
93
+ return '';
94
+ return match[1]
95
+ .replace(/<[^>]+>/g, '')
96
+ .replace(/&amp;/g, '&').replace(/&lt;/g, '<').replace(/&gt;/g, '>')
97
+ .replace(/&quot;/g, '"').replace(/&nbsp;/g, ' ')
98
+ .trim();
99
+ }
100
+ /**
101
+ * Extract anchor links from HTML (up to MAX_LINKS).
102
+ */
103
+ function extractLinks(html, baseUrl) {
104
+ const links = [];
105
+ const anchorRe = /<a[^>]+href="([^"#][^"]*)"[^>]*>([\s\S]*?)<\/a>/gi;
106
+ let match;
107
+ let base;
108
+ try {
109
+ base = new URL(baseUrl);
110
+ }
111
+ catch {
112
+ return links;
113
+ }
114
+ while ((match = anchorRe.exec(html)) !== null && links.length < MAX_LINKS) {
115
+ const rawHref = match[1].trim();
116
+ const rawText = match[2].replace(/<[^>]+>/g, ' ').replace(/\s+/g, ' ').trim().slice(0, 120);
117
+ if (!rawText || rawHref.startsWith('javascript:') || rawHref.startsWith('mailto:'))
118
+ continue;
119
+ let href;
120
+ try {
121
+ href = new URL(rawHref, base.origin + base.pathname).href;
122
+ }
123
+ catch {
124
+ continue;
125
+ }
126
+ links.push({ href, text: rawText });
127
+ }
128
+ return links;
129
+ }
130
+ /**
131
+ * Primary extraction: Python bs4.get_text().
132
+ * bs4 is available in the benchmark env (already installed for gaia-codeagent).
133
+ * HTML is passed via stdin to avoid shell-escaping issues.
134
+ */
135
+ function extractTextViaPython(html) {
136
+ const script = [
137
+ 'import sys, re',
138
+ 'from bs4 import BeautifulSoup',
139
+ 'html = sys.stdin.read()',
140
+ "soup = BeautifulSoup(html, 'html.parser')",
141
+ "for tag in soup(['script', 'style', 'nav', 'footer', 'header', 'aside']):",
142
+ ' tag.decompose()',
143
+ "text = soup.get_text(separator='\\n', strip=True)",
144
+ "text = re.sub(r'\\n{3,}', '\\n\\n', text)",
145
+ 'print(text)',
146
+ ].join('\n');
147
+ try {
148
+ return execFileSync('python3', ['-'], {
149
+ input: script + '\n' + html,
150
+ encoding: 'utf-8',
151
+ timeout: 15_000,
152
+ env: { ...process.env, PYTHONIOENCODING: 'utf-8' },
153
+ maxBuffer: 10 * 1024 * 1024,
154
+ }).trim();
155
+ }
156
+ catch {
157
+ return '';
158
+ }
159
+ }
160
+ /**
161
+ * Fallback: regex-based HTML stripping (no external dep, less accurate).
162
+ */
163
+ function extractTextViaRegex(html) {
164
+ return html
165
+ .replace(/<script[\s\S]*?<\/script>/gi, '')
166
+ .replace(/<style[\s\S]*?<\/style>/gi, '')
167
+ .replace(/<nav[\s\S]*?<\/nav>/gi, '')
168
+ .replace(/<footer[\s\S]*?<\/footer>/gi, '')
169
+ .replace(/<header[\s\S]*?<\/header>/gi, '')
170
+ .replace(/<\/?(p|div|section|article|main|li|h[1-6]|br)[^>]*>/gi, '\n')
171
+ .replace(/<[^>]+>/g, ' ')
172
+ .replace(/&amp;/g, '&').replace(/&lt;/g, '<').replace(/&gt;/g, '>')
173
+ .replace(/&quot;/g, '"').replace(/&nbsp;/g, ' ').replace(/&#x27;/g, "'")
174
+ .replace(/&#(\d+);/g, (_, n) => String.fromCharCode(Number(n)))
175
+ .replace(/ +/g, ' ')
176
+ .replace(/\n{4,}/g, '\n\n\n')
177
+ .replace(/^\s+|\s+$/gm, '')
178
+ .trim();
179
+ }
180
+ // ---------------------------------------------------------------------------
181
+ // AIDefence PII gate (optional — graceful skip if not installed)
182
+ // ---------------------------------------------------------------------------
183
+ async function applyPiiGate(content) {
184
+ try {
185
+ const aidefence = await import('@claude-flow/aidefence').catch(() => null);
186
+ if (!aidefence)
187
+ return content;
188
+ const { hasPii } = aidefence;
189
+ if (typeof hasPii !== 'function')
190
+ return content;
191
+ const detected = await hasPii(content);
192
+ if (detected) {
193
+ process.stderr.write('[visit_webpage] AIDefence: PII detected in page content (not blocked)\n');
194
+ }
195
+ }
196
+ catch {
197
+ // AIDefence unavailable — continue without gate
198
+ }
199
+ return content;
200
+ }
201
+ // ---------------------------------------------------------------------------
202
+ // Core visit logic (exported for unit testing via visitWebpageTestHooks)
203
+ // ---------------------------------------------------------------------------
204
+ export async function visitWebpage(url, maxChars = DEFAULT_MAX_CHARS) {
205
+ const sourcesConsulted = [url];
206
+ const { html, finalUrl } = visitWebpageTestHooks.fetchHtml
207
+ ? await visitWebpageTestHooks.fetchHtml(url)
208
+ : await fetchPage(url);
209
+ if (finalUrl !== url)
210
+ sourcesConsulted.push(finalUrl);
211
+ const title = extractTitle(html);
212
+ const links = extractLinks(html, finalUrl);
213
+ // Primary: Python bs4; fallback: regex
214
+ let content = extractTextViaPython(html);
215
+ if (!content || content.length < MIN_CONTENT_CHARS) {
216
+ content = extractTextViaRegex(html);
217
+ }
218
+ content = await applyPiiGate(content);
219
+ const truncated = content.length > maxChars;
220
+ if (truncated)
221
+ content = content.slice(0, maxChars);
222
+ return {
223
+ title,
224
+ content,
225
+ links: links.slice(0, MAX_LINKS),
226
+ sources_consulted: sourcesConsulted,
227
+ chars_returned: content.length,
228
+ truncated,
229
+ };
230
+ }
231
+ // ---------------------------------------------------------------------------
232
+ // Format output for Claude
233
+ // ---------------------------------------------------------------------------
234
+ function formatResult(result, url) {
235
+ const lines = [];
236
+ lines.push(`[visit_webpage: ${url}]`);
237
+ if (result.title)
238
+ lines.push(`Title: ${result.title}`);
239
+ if (result.sources_consulted.length > 1) {
240
+ lines.push(`Redirected to: ${result.sources_consulted[result.sources_consulted.length - 1]}`);
241
+ }
242
+ if (result.truncated)
243
+ lines.push(`[Content truncated at ${result.chars_returned} chars]`);
244
+ lines.push('');
245
+ lines.push(result.content || '[No readable content extracted]');
246
+ if (result.links.length > 0) {
247
+ lines.push('');
248
+ lines.push(`Links (${result.links.length}):`);
249
+ for (const link of result.links)
250
+ lines.push(` ${link.text} → ${link.href}`);
251
+ }
252
+ return lines.join('\n');
253
+ }
254
+ // ---------------------------------------------------------------------------
255
+ // GaiaTool implementation
256
+ // ---------------------------------------------------------------------------
257
+ export class VisitWebpageTool {
258
+ name = 'visit_webpage';
259
+ definition = {
260
+ name: 'visit_webpage',
261
+ description: 'Fetch the FULL text content of a webpage URL. Unlike web_search (snippets only) ' +
262
+ 'or grounded_query (synthesised summary), visit_webpage returns the complete readable ' +
263
+ "text — equivalent to HAL's visit_webpage tool. Use this when you need to read a " +
264
+ 'specific Wikipedia article, documentation page, government site, or any web page ' +
265
+ 'in full to extract detailed facts, tables, or text not available from snippets. ' +
266
+ 'Returns: page title, full text content (up to max_chars), and up to 30 extracted links. ' +
267
+ `Default max_chars: ${DEFAULT_MAX_CHARS}. Maximum: 100,000.`,
268
+ input_schema: {
269
+ type: 'object',
270
+ properties: {
271
+ url: {
272
+ type: 'string',
273
+ description: 'The full URL (http:// or https://) of the webpage to fetch.',
274
+ },
275
+ max_chars: {
276
+ type: 'number',
277
+ description: `Maximum characters of content to return (default: ${DEFAULT_MAX_CHARS}, max: 100000).`,
278
+ },
279
+ },
280
+ required: ['url'],
281
+ },
282
+ };
283
+ async execute(input) {
284
+ const url = String(input['url'] ?? '').trim();
285
+ validateUrl(url);
286
+ const maxChars = Math.min(Math.max(500, Number(input['max_chars'] ?? DEFAULT_MAX_CHARS)), 100_000);
287
+ process.stderr.write(`[visit_webpage] url=${url} max_chars=${maxChars}\n`);
288
+ const result = await visitWebpage(url, maxChars);
289
+ process.stderr.write(`[visit_webpage] title=${JSON.stringify(result.title)} ` +
290
+ `chars=${result.chars_returned} truncated=${result.truncated} ` +
291
+ `links=${result.links.length}\n`);
292
+ if (!result.content || result.content.length < MIN_CONTENT_CHARS) {
293
+ return `[visit_webpage: page at ${url} returned no readable text content]`;
294
+ }
295
+ return formatResult(result, url);
296
+ }
297
+ }
298
+ // ---------------------------------------------------------------------------
299
+ // Convenience factory
300
+ // ---------------------------------------------------------------------------
301
+ export function createVisitWebpageTool() {
302
+ return new VisitWebpageTool();
303
+ }
304
+ //# sourceMappingURL=visit_webpage.js.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"visit_webpage.js","sourceRoot":"","sources":["../../../../src/benchmarks/gaia-tools/visit_webpage.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;GAuBG;AAEH,OAAO,EAAE,YAAY,EAAE,MAAM,oBAAoB,CAAC;AAGlD,8EAA8E;AAC9E,YAAY;AACZ,8EAA8E;AAE9E,MAAM,iBAAiB,GAAG,MAAM,CAAC;AACjC,MAAM,iBAAiB,GAAG,EAAE,CAAC;AAC7B,MAAM,kBAAkB,GAAG,MAAM,CAAC;AAClC,MAAM,SAAS,GAAG,EAAE,CAAC;AACrB,MAAM,EAAE,GACN,uGAAuG,CAAC;AAwB1G,MAAM,CAAC,MAAM,qBAAqB,GAA0B,EAAE,CAAC;AAE/D,8EAA8E;AAC9E,iBAAiB;AACjB,8EAA8E;AAE9E,SAAS,WAAW,CAAC,GAAW;IAC9B,IAAI,CAAC,GAAG,IAAI,OAAO,GAAG,KAAK,QAAQ,EAAE,CAAC;QACpC,MAAM,IAAI,KAAK,CAAC,kDAAkD,CAAC,CAAC;IACtE,CAAC;IACD,IAAI,CAAC,GAAG,CAAC,UAAU,CAAC,SAAS,CAAC,IAAI,CAAC,GAAG,CAAC,UAAU,CAAC,UAAU,CAAC,EAAE,CAAC;QAC9D,MAAM,IAAI,KAAK,CACb,iEAAiE,GAAG,GAAG,CACxE,CAAC;IACJ,CAAC;IACD,IAAI,CAAC;QACH,IAAI,GAAG,CAAC,GAAG,CAAC,CAAC;IACf,CAAC;IAAC,MAAM,CAAC;QACP,MAAM,IAAI,KAAK,CAAC,gCAAgC,GAAG,GAAG,CAAC,CAAC;IAC1D,CAAC;AACH,CAAC;AAED,8EAA8E;AAC9E,qDAAqD;AACrD,8EAA8E;AAE9E,KAAK,UAAU,SAAS,CAAC,GAAW;IAClC,MAAM,UAAU,GAAG,IAAI,eAAe,EAAE,CAAC;IACzC,MAAM,KAAK,GAAG,UAAU,CAAC,GAAG,EAAE,CAAC,UAAU,CAAC,KAAK,EAAE,EAAE,kBAAkB,CAAC,CAAC;IAEvE,IAAI,GAAa,CAAC;IAClB,IAAI,CAAC;QACH,GAAG,GAAG,MAAM,KAAK,CAAC,GAAG,EAAE;YACrB,OAAO,EAAE;gBACP,YAAY,EAAE,EAAE;gBAChB,MAAM,EAAE,iEAAiE;gBACzE,iBAAiB,EAAE,gBAAgB;aACpC;YACD,MAAM,EAAE,UAAU,CAAC,MAAM;YACzB,QAAQ,EAAE,QAAQ;SACnB,CAAC,CAAC;IACL,CAAC;YAAS,CAAC;QACT,YAAY,CAAC,KAAK,CAAC,CAAC;IACtB,CAAC;IAED,IAAI,CAAC,GAAG,CAAC,EAAE,EAAE,CAAC;QACZ,MAAM,IAAI,KAAK,CAAC,uBAAuB,GAAG,CAAC,MAAM,QAAQ,GAAG,EAAE,CAAC,CAAC;IAClE,CAAC;IAED,MAAM,EAAE,GAAG,GAAG,CAAC,OAAO,CAAC,GAAG,CAAC,cAAc,CAAC,IAAI,EAAE,CAAC;IACjD,IAAI,EAAE,CAAC,QAAQ,CAAC,iBAAiB,CAAC,EAAE,CAAC;QACnC,MAAM,IAAI,KAAK,CACb,0FAA0F,GAAG,EAAE,CAChG,CAAC;IACJ,CAAC;IAED,MAAM,IAAI,GAAG,MAAM,GAAG,CAAC,IAAI,EAAE,CAAC;IAC9B,MAAM,QAAQ,GAAG,GAAG,CAAC,GAAG,IAAI,GAAG,CAAC;IAChC,OAAO,EAAE,IAAI,EAAE,QAAQ,EAAE,CAAC;AAC5B,CAAC;AAED,8EAA8E;AAC9E,0BAA0B;AAC1B,8EAA8E;AAE9E;;GAEG;AACH,SAAS,YAAY,CAAC,IAAY;IAChC,MAAM,KAAK,GAAG,kCAAkC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAC5D,IAAI,CAAC,KAAK;QAAE,OAAO,EAAE,CAAC;IACtB,OAAO,KAAK,CAAC,CAAC,CAAC;SACZ,OAAO,CAAC,UAAU,EAAE,EAAE,CAAC;SACvB,OAAO,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC;SAClE,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC;SAC/C,IAAI,EAAE,CAAC;AACZ,CAAC;AAED;;GAEG;AACH,SAAS,YAAY,CAAC,IAAY,EAAE,OAAe;IACjD,MAAM,KAAK,GAA0C,EAAE,CAAC;IACxD,MAAM,QAAQ,GAAG,mDAAmD,CAAC;IACrE,IAAI,KAA6B,CAAC;IAClC,IAAI,IAAS,CAAC;IACd,IAAI,CAAC;QAAC,IAAI,GAAG,IAAI,GAAG,CAAC,OAAO,CAAC,CAAC;IAAC,CAAC;IAAC,MAAM,CAAC;QAAC,OAAO,KAAK,CAAC;IAAC,CAAC;IAExD,OAAO,CAAC,KAAK,GAAG,QAAQ,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC,KAAK,IAAI,IAAI,KAAK,CAAC,MAAM,GAAG,SAAS,EAAE,CAAC;QAC1E,MAAM,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,CAAC;QAChC,MAAM,OAAO,GAAG,KAAK,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,UAAU,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC,CAAC;QAE5F,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,UAAU,CAAC,aAAa,CAAC,IAAI,OAAO,CAAC,UAAU,CAAC,SAAS,CAAC;YAAE,SAAS;QAE7F,IAAI,IAAY,CAAC;QACjB,IAAI,CAAC;YAAC,IAAI,GAAG,IAAI,GAAG,CAAC,OAAO,EAAE,IAAI,CAAC,MAAM,GAAG,IAAI,CAAC,QAAQ,CAAC,CAAC,IAAI,CAAC;QAAC,CAAC;QAAC,MAAM,CAAC;YAAC,SAAS;QAAC,CAAC;QAEtF,KAAK,CAAC,IAAI,CAAC,EAAE,IAAI,EAAE,IAAI,EAAE,OAAO,EAAE,CAAC,CAAC;IACtC,CAAC;IACD,OAAO,KAAK,CAAC;AACf,CAAC;AAED;;;;GAIG;AACH,SAAS,oBAAoB,CAAC,IAAY;IACxC,MAAM,MAAM,GAAG;QACb,gBAAgB;QAChB,+BAA+B;QAC/B,yBAAyB;QACzB,2CAA2C;QAC3C,2EAA2E;QAC3E,qBAAqB;QACrB,mDAAmD;QACnD,2CAA2C;QAC3C,aAAa;KACd,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAEb,IAAI,CAAC;QACH,OAAO,YAAY,CAAC,SAAS,EAAE,CAAC,GAAG,CAAC,EAAE;YACpC,KAAK,EAAE,MAAM,GAAG,IAAI,GAAG,IAAI;YAC3B,QAAQ,EAAE,OAAO;YACjB,OAAO,EAAE,MAAM;YACf,GAAG,EAAE,EAAE,GAAG,OAAO,CAAC,GAAG,EAAE,gBAAgB,EAAE,OAAO,EAAE;YAClD,SAAS,EAAE,EAAE,GAAG,IAAI,GAAG,IAAI;SAC5B,CAAC,CAAC,IAAI,EAAE,CAAC;IACZ,CAAC;IAAC,MAAM,CAAC;QACP,OAAO,EAAE,CAAC;IACZ,CAAC;AACH,CAAC;AAED;;GAEG;AACH,SAAS,mBAAmB,CAAC,IAAY;IACvC,OAAO,IAAI;SACR,OAAO,CAAC,6BAA6B,EAAE,EAAE,CAAC;SAC1C,OAAO,CAAC,2BAA2B,EAAE,EAAE,CAAC;SACxC,OAAO,CAAC,uBAAuB,EAAE,EAAE,CAAC;SACpC,OAAO,CAAC,6BAA6B,EAAE,EAAE,CAAC;SAC1C,OAAO,CAAC,6BAA6B,EAAE,EAAE,CAAC;SAC1C,OAAO,CAAC,uDAAuD,EAAE,IAAI,CAAC;SACtE,OAAO,CAAC,UAAU,EAAE,GAAG,CAAC;SACxB,OAAO,CAAC,QAAQ,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,OAAO,EAAE,GAAG,CAAC;SAClE,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC,CAAC,OAAO,CAAC,SAAS,EAAE,GAAG,CAAC;SACvE,OAAO,CAAC,WAAW,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,MAAM,CAAC,YAAY,CAAC,MAAM,CAAC,CAAC,CAAC,CAAC,CAAC;SAC9D,OAAO,CAAC,KAAK,EAAE,GAAG,CAAC;SACnB,OAAO,CAAC,SAAS,EAAE,QAAQ,CAAC;SAC5B,OAAO,CAAC,aAAa,EAAE,EAAE,CAAC;SAC1B,IAAI,EAAE,CAAC;AACZ,CAAC;AAED,8EAA8E;AAC9E,iEAAiE;AACjE,8EAA8E;AAE9E,KAAK,UAAU,YAAY,CAAC,OAAe;IACzC,IAAI,CAAC;QACH,MAAM,SAAS,GAAG,MAAM,MAAM,CAAC,wBAAwB,CAAC,CAAC,KAAK,CAAC,GAAG,EAAE,CAAC,IAAI,CAAC,CAAC;QAC3E,IAAI,CAAC,SAAS;YAAE,OAAO,OAAO,CAAC;QAC/B,MAAM,EAAE,MAAM,EAAE,GAAG,SAA4D,CAAC;QAChF,IAAI,OAAO,MAAM,KAAK,UAAU;YAAE,OAAO,OAAO,CAAC;QACjD,MAAM,QAAQ,GAAG,MAAM,MAAM,CAAC,OAAO,CAAC,CAAC;QACvC,IAAI,QAAQ,EAAE,CAAC;YACb,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,yEAAyE,CAAC,CAAC;QAClG,CAAC;IACH,CAAC;IAAC,MAAM,CAAC;QACP,gDAAgD;IAClD,CAAC;IACD,OAAO,OAAO,CAAC;AACjB,CAAC;AAED,8EAA8E;AAC9E,yEAAyE;AACzE,8EAA8E;AAE9E,MAAM,CAAC,KAAK,UAAU,YAAY,CAChC,GAAW,EACX,WAAmB,iBAAiB;IAEpC,MAAM,gBAAgB,GAAa,CAAC,GAAG,CAAC,CAAC;IAEzC,MAAM,EAAE,IAAI,EAAE,QAAQ,EAAE,GAAG,qBAAqB,CAAC,SAAS;QACxD,CAAC,CAAC,MAAM,qBAAqB,CAAC,SAAS,CAAC,GAAG,CAAC;QAC5C,CAAC,CAAC,MAAM,SAAS,CAAC,GAAG,CAAC,CAAC;IAEzB,IAAI,QAAQ,KAAK,GAAG;QAAE,gBAAgB,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IAEtD,MAAM,KAAK,GAAG,YAAY,CAAC,IAAI,CAAC,CAAC;IACjC,MAAM,KAAK,GAAG,YAAY,CAAC,IAAI,EAAE,QAAQ,CAAC,CAAC;IAE3C,uCAAuC;IACvC,IAAI,OAAO,GAAG,oBAAoB,CAAC,IAAI,CAAC,CAAC;IACzC,IAAI,CAAC,OAAO,IAAI,OAAO,CAAC,MAAM,GAAG,iBAAiB,EAAE,CAAC;QACnD,OAAO,GAAG,mBAAmB,CAAC,IAAI,CAAC,CAAC;IACtC,CAAC;IAED,OAAO,GAAG,MAAM,YAAY,CAAC,OAAO,CAAC,CAAC;IAEtC,MAAM,SAAS,GAAG,OAAO,CAAC,MAAM,GAAG,QAAQ,CAAC;IAC5C,IAAI,SAAS;QAAE,OAAO,GAAG,OAAO,CAAC,KAAK,CAAC,CAAC,EAAE,QAAQ,CAAC,CAAC;IAEpD,OAAO;QACL,KAAK;QACL,OAAO;QACP,KAAK,EAAE,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC;QAChC,iBAAiB,EAAE,gBAAgB;QACnC,cAAc,EAAE,OAAO,CAAC,MAAM;QAC9B,SAAS;KACV,CAAC;AACJ,CAAC;AAED,8EAA8E;AAC9E,2BAA2B;AAC3B,8EAA8E;AAE9E,SAAS,YAAY,CAAC,MAA0B,EAAE,GAAW;IAC3D,MAAM,KAAK,GAAa,EAAE,CAAC;IAC3B,KAAK,CAAC,IAAI,CAAC,mBAAmB,GAAG,GAAG,CAAC,CAAC;IACtC,IAAI,MAAM,CAAC,KAAK;QAAE,KAAK,CAAC,IAAI,CAAC,UAAU,MAAM,CAAC,KAAK,EAAE,CAAC,CAAC;IACvD,IAAI,MAAM,CAAC,iBAAiB,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QACxC,KAAK,CAAC,IAAI,CAAC,kBAAkB,MAAM,CAAC,iBAAiB,CAAC,MAAM,CAAC,iBAAiB,CAAC,MAAM,GAAG,CAAC,CAAC,EAAE,CAAC,CAAC;IAChG,CAAC;IACD,IAAI,MAAM,CAAC,SAAS;QAAE,KAAK,CAAC,IAAI,CAAC,yBAAyB,MAAM,CAAC,cAAc,SAAS,CAAC,CAAC;IAC1F,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;IACf,KAAK,CAAC,IAAI,CAAC,MAAM,CAAC,OAAO,IAAI,iCAAiC,CAAC,CAAC;IAEhE,IAAI,MAAM,CAAC,KAAK,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;QAC5B,KAAK,CAAC,IAAI,CAAC,EAAE,CAAC,CAAC;QACf,KAAK,CAAC,IAAI,CAAC,UAAU,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,CAAC,CAAC;QAC9C,KAAK,MAAM,IAAI,IAAI,MAAM,CAAC,KAAK;YAAE,KAAK,CAAC,IAAI,CAAC,KAAK,IAAI,CAAC,IAAI,MAAM,IAAI,CAAC,IAAI,EAAE,CAAC,CAAC;IAC/E,CAAC;IACD,OAAO,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;AAC1B,CAAC;AAED,8EAA8E;AAC9E,0BAA0B;AAC1B,8EAA8E;AAE9E,MAAM,OAAO,gBAAgB;IAClB,IAAI,GAAG,eAAe,CAAC;IAEvB,UAAU,GAAmB;QACpC,IAAI,EAAE,eAAe;QACrB,WAAW,EACT,kFAAkF;YAClF,uFAAuF;YACvF,kFAAkF;YAClF,mFAAmF;YACnF,kFAAkF;YAClF,0FAA0F;YAC1F,sBAAsB,iBAAiB,qBAAqB;QAC9D,YAAY,EAAE;YACZ,IAAI,EAAE,QAAQ;YACd,UAAU,EAAE;gBACV,GAAG,EAAE;oBACH,IAAI,EAAE,QAAQ;oBACd,WAAW,EAAE,6DAA6D;iBAC3E;gBACD,SAAS,EAAE;oBACT,IAAI,EAAE,QAAQ;oBACd,WAAW,EAAE,qDAAqD,iBAAiB,iBAAiB;iBACrG;aACF;YACD,QAAQ,EAAE,CAAC,KAAK,CAAC;SAClB;KACF,CAAC;IAEF,KAAK,CAAC,OAAO,CAAC,KAA8B;QAC1C,MAAM,GAAG,GAAG,MAAM,CAAC,KAAK,CAAC,KAAK,CAAC,IAAI,EAAE,CAAC,CAAC,IAAI,EAAE,CAAC;QAC9C,WAAW,CAAC,GAAG,CAAC,CAAC;QAEjB,MAAM,QAAQ,GAAG,IAAI,CAAC,GAAG,CACvB,IAAI,CAAC,GAAG,CAAC,GAAG,EAAE,MAAM,CAAC,KAAK,CAAC,WAAW,CAAC,IAAI,iBAAiB,CAAC,CAAC,EAC9D,OAAO,CACR,CAAC;QAEF,OAAO,CAAC,MAAM,CAAC,KAAK,CAAC,uBAAuB,GAAG,cAAc,QAAQ,IAAI,CAAC,CAAC;QAE3E,MAAM,MAAM,GAAG,MAAM,YAAY,CAAC,GAAG,EAAE,QAAQ,CAAC,CAAC;QAEjD,OAAO,CAAC,MAAM,CAAC,KAAK,CAClB,yBAAyB,IAAI,CAAC,SAAS,CAAC,MAAM,CAAC,KAAK,CAAC,GAAG;YACtD,SAAS,MAAM,CAAC,cAAc,cAAc,MAAM,CAAC,SAAS,GAAG;YAC/D,SAAS,MAAM,CAAC,KAAK,CAAC,MAAM,IAAI,CACnC,CAAC;QAEF,IAAI,CAAC,MAAM,CAAC,OAAO,IAAI,MAAM,CAAC,OAAO,CAAC,MAAM,GAAG,iBAAiB,EAAE,CAAC;YACjE,OAAO,2BAA2B,GAAG,qCAAqC,CAAC;QAC7E,CAAC;QAED,OAAO,YAAY,CAAC,MAAM,EAAE,GAAG,CAAC,CAAC;IACnC,CAAC;CACF;AAED,8EAA8E;AAC9E,sBAAsB;AACtB,8EAA8E;AAE9E,MAAM,UAAU,sBAAsB;IACpC,OAAO,IAAI,gBAAgB,EAAE,CAAC;AAChC,CAAC"}
@@ -0,0 +1,61 @@
1
+ /**
2
+ * GAIA Tool: web_browse — ADR-133-PR5
3
+ *
4
+ * Opens a URL in a headless Chromium browser via Playwright and extracts
5
+ * page content (text, HTML, or screenshot). Covers the ~10-15pp of GAIA
6
+ * Level-1 questions that require navigating dynamic JS pages, video pages
7
+ * (YouTube, Vimeo), and paywalled/login-required content.
8
+ *
9
+ * ============================================================
10
+ * PLAYWRIGHT DEPENDENCY — OPT-IN INSTALL
11
+ * ============================================================
12
+ * Playwright is NOT a hard runtime dep of @claude-flow/cli. It is loaded
13
+ * lazily via a dynamic import so the package installs cleanly without it.
14
+ *
15
+ * To use web_browse, run once:
16
+ * npm install playwright
17
+ * npx playwright install chromium
18
+ *
19
+ * If Playwright is not installed, execute() returns a structured error
20
+ * message that Claude can relay to the user rather than crashing.
21
+ *
22
+ * Install size: ~80 MB for the Playwright package + Chromium binary.
23
+ * Add as a devDependency in benchmark-specific contexts to avoid bloating
24
+ * the production bundle.
25
+ *
26
+ * ============================================================
27
+ * RESOURCE CAPS
28
+ * ============================================================
29
+ * - Text/HTML extraction capped at 8 000 characters (prevents context-window
30
+ * overflow; roughly 2 000 tokens).
31
+ * - Default timeout: 30 seconds.
32
+ * - Screenshots returned as base64-encoded PNG strings.
33
+ * - Browser instance is always closed in a `finally` block.
34
+ *
35
+ * Refs: ADR-133, #2156
36
+ */
37
+ import { GaiaTool, ToolDefinition } from './types.js';
38
+ export type BrowseExtract = 'text' | 'html' | 'screenshot';
39
+ export interface WebBrowseInput {
40
+ url: string;
41
+ wait_for_selector?: string;
42
+ extract?: BrowseExtract;
43
+ timeout_seconds?: number;
44
+ }
45
+ export interface WebBrowseResult {
46
+ content: string;
47
+ final_url: string;
48
+ status: number;
49
+ truncated?: boolean;
50
+ }
51
+ export declare class WebBrowseTool implements GaiaTool {
52
+ readonly name = "web_browse";
53
+ readonly definition: ToolDefinition;
54
+ execute(input: Record<string, unknown>): Promise<string>;
55
+ }
56
+ export interface WebBrowseToolOptions {
57
+ /** Override default timeout in milliseconds. */
58
+ defaultTimeoutMs?: number;
59
+ }
60
+ export declare function createWebBrowseTool(_opts?: WebBrowseToolOptions): WebBrowseTool;
61
+ //# sourceMappingURL=web_browse.d.ts.map
@@ -0,0 +1 @@
1
+ {"version":3,"file":"web_browse.d.ts","sourceRoot":"","sources":["../../../../src/benchmarks/gaia-tools/web_browse.ts"],"names":[],"mappings":"AAAA;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;GAmCG;AAEH,OAAO,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,YAAY,CAAC;AAatD,MAAM,MAAM,aAAa,GAAG,MAAM,GAAG,MAAM,GAAG,YAAY,CAAC;AAE3D,MAAM,WAAW,cAAc;IAC7B,GAAG,EAAE,MAAM,CAAC;IACZ,iBAAiB,CAAC,EAAE,MAAM,CAAC;IAC3B,OAAO,CAAC,EAAE,aAAa,CAAC;IACxB,eAAe,CAAC,EAAE,MAAM,CAAC;CAC1B;AAED,MAAM,WAAW,eAAe;IAC9B,OAAO,EAAE,MAAM,CAAC;IAChB,SAAS,EAAE,MAAM,CAAC;IAClB,MAAM,EAAE,MAAM,CAAC;IACf,SAAS,CAAC,EAAE,OAAO,CAAC;CACrB;AA6JD,qBAAa,aAAc,YAAW,QAAQ;IAC5C,QAAQ,CAAC,IAAI,gBAAgB;IAE7B,QAAQ,CAAC,UAAU,EAAE,cAAc,CAoCjC;IAEI,OAAO,CAAC,KAAK,EAAE,MAAM,CAAC,MAAM,EAAE,OAAO,CAAC,GAAG,OAAO,CAAC,MAAM,CAAC;CAoB/D;AAMD,MAAM,WAAW,oBAAoB;IACnC,gDAAgD;IAChD,gBAAgB,CAAC,EAAE,MAAM,CAAC;CAC3B;AAED,wBAAgB,mBAAmB,CAAC,KAAK,CAAC,EAAE,oBAAoB,GAAG,aAAa,CAE/E"}