codeply-cli 0.5.0 → 0.5.2

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 (387) hide show
  1. package/LICENSE +21 -0
  2. package/THIRD_PARTY_NOTICES.md +65 -63
  3. package/bin/codeply.js +1 -1
  4. package/lib/agent.mjs +133 -24
  5. package/lib/ai.js +268 -9
  6. package/lib/ai.mjs +6 -6
  7. package/lib/apply-patch.js +217 -217
  8. package/lib/arity-table.json +138 -138
  9. package/lib/arity.js +70 -70
  10. package/lib/auth.mjs +6 -6
  11. package/lib/bots.js +493 -16
  12. package/lib/chatgpt.js +282 -0
  13. package/lib/cloud-browser.mjs +143 -143
  14. package/lib/cloud.mjs +1186 -1105
  15. package/lib/commands.js +103 -103
  16. package/lib/config.js +83 -3
  17. package/lib/design-library/assemble-artifact.mjs +21 -21
  18. package/lib/design-library/cleanup.mjs +45 -45
  19. package/lib/design-library/index.json +22379 -22379
  20. package/lib/design-library/make-table-data.mjs +39 -39
  21. package/lib/diagnostics.js +192 -192
  22. package/lib/edit-engine.js +364 -364
  23. package/lib/github-agent.mjs +311 -311
  24. package/lib/history.js +75 -75
  25. package/lib/mail-watch.js +504 -0
  26. package/lib/mcp.js +258 -258
  27. package/lib/model-sync.js +114 -0
  28. package/lib/native-tools.mjs +16 -1
  29. package/lib/oauth-connectors.js +146 -24
  30. package/lib/permissions.js +109 -109
  31. package/lib/plugins.js +353 -353
  32. package/lib/publish.js +916 -0
  33. package/lib/reach.js +67 -0
  34. package/lib/server.mjs +457 -457
  35. package/lib/session-db.js +220 -220
  36. package/lib/share.js +168 -168
  37. package/lib/skills.js +2 -0
  38. package/lib/snapshot.js +146 -146
  39. package/lib/telemetry.js +102 -0
  40. package/lib/terminal-hints.js +154 -154
  41. package/lib/tools.mjs +316 -14
  42. package/lib/typecheck-worker.js +147 -147
  43. package/lib/web-tools.js +209 -158
  44. package/package.json +49 -49
  45. package/skills/SOURCE.md +58 -58
  46. package/skills/accessibility/SKILL.md +147 -147
  47. package/skills/agent-architecture-audit/SKILL.md +257 -257
  48. package/skills/agent-eval/SKILL.md +147 -147
  49. package/skills/agent-harness-construction/SKILL.md +74 -74
  50. package/skills/agent-introspection-debugging/SKILL.md +154 -154
  51. package/skills/agent-payment-x402/SKILL.md +225 -225
  52. package/skills/agent-self-evaluation/SKILL.md +182 -182
  53. package/skills/agent-self-evaluation/references/evaluation-criteria.md +71 -71
  54. package/skills/agent-self-evaluation/references/hook-integration.md +64 -64
  55. package/skills/agent-sort/SKILL.md +216 -216
  56. package/skills/agentic-engineering/SKILL.md +64 -64
  57. package/skills/agentic-os/SKILL.md +388 -388
  58. package/skills/ai-first-engineering/SKILL.md +52 -52
  59. package/skills/ai-regression-testing/SKILL.md +386 -386
  60. package/skills/android-clean-architecture/SKILL.md +340 -340
  61. package/skills/angular-developer/SKILL.md +155 -155
  62. package/skills/angular-developer/references/angular-animations.md +160 -160
  63. package/skills/angular-developer/references/angular-aria.md +410 -410
  64. package/skills/angular-developer/references/cli.md +86 -86
  65. package/skills/angular-developer/references/component-harnesses.md +59 -59
  66. package/skills/angular-developer/references/component-styling.md +91 -91
  67. package/skills/angular-developer/references/components.md +117 -117
  68. package/skills/angular-developer/references/creating-services.md +97 -97
  69. package/skills/angular-developer/references/data-resolvers.md +69 -69
  70. package/skills/angular-developer/references/define-routes.md +67 -67
  71. package/skills/angular-developer/references/defining-providers.md +72 -72
  72. package/skills/angular-developer/references/di-fundamentals.md +120 -120
  73. package/skills/angular-developer/references/e2e-testing.md +56 -56
  74. package/skills/angular-developer/references/effects.md +83 -83
  75. package/skills/angular-developer/references/hierarchical-injectors.md +43 -43
  76. package/skills/angular-developer/references/host-elements.md +80 -80
  77. package/skills/angular-developer/references/injection-context.md +63 -63
  78. package/skills/angular-developer/references/inputs.md +101 -101
  79. package/skills/angular-developer/references/linked-signal.md +59 -59
  80. package/skills/angular-developer/references/loading-strategies.md +61 -61
  81. package/skills/angular-developer/references/mcp.md +108 -108
  82. package/skills/angular-developer/references/navigate-to-routes.md +69 -69
  83. package/skills/angular-developer/references/outputs.md +86 -86
  84. package/skills/angular-developer/references/reactive-forms.md +122 -122
  85. package/skills/angular-developer/references/rendering-strategies.md +44 -44
  86. package/skills/angular-developer/references/resource.md +77 -77
  87. package/skills/angular-developer/references/route-animations.md +56 -56
  88. package/skills/angular-developer/references/route-guards.md +52 -52
  89. package/skills/angular-developer/references/router-lifecycle.md +45 -45
  90. package/skills/angular-developer/references/router-testing.md +87 -87
  91. package/skills/angular-developer/references/show-routes-with-outlets.md +68 -68
  92. package/skills/angular-developer/references/signal-forms.md +795 -795
  93. package/skills/angular-developer/references/signals-overview.md +94 -94
  94. package/skills/angular-developer/references/tailwind-css.md +69 -69
  95. package/skills/angular-developer/references/template-driven-forms.md +114 -114
  96. package/skills/angular-developer/references/testing-fundamentals.md +65 -65
  97. package/skills/api-connector-builder/SKILL.md +121 -121
  98. package/skills/api-design/SKILL.md +524 -524
  99. package/skills/architecture-decision-records/SKILL.md +180 -180
  100. package/skills/article-writing/SKILL.md +80 -80
  101. package/skills/automation-audit-ops/SKILL.md +143 -143
  102. package/skills/autonomous-agent-harness/SKILL.md +274 -274
  103. package/skills/autonomous-loops/SKILL.md +611 -611
  104. package/skills/backend-patterns/SKILL.md +562 -562
  105. package/skills/benchmark/SKILL.md +95 -95
  106. package/skills/benchmark-methodology/SKILL.md +191 -191
  107. package/skills/benchmark-optimization-loop/SKILL.md +71 -71
  108. package/skills/blender-motion-state-inspection/SKILL.md +165 -165
  109. package/skills/blueprint/SKILL.md +106 -106
  110. package/skills/brand-discovery/SKILL.md +145 -145
  111. package/skills/brand-discovery/references/10_purpose-why.md +40 -40
  112. package/skills/brand-discovery/references/20_positioning.md +44 -44
  113. package/skills/brand-discovery/references/30_audience-niche.md +52 -52
  114. package/skills/brand-discovery/references/40_personality-archetype.md +57 -57
  115. package/skills/brand-discovery/references/50_voice-tone.md +59 -59
  116. package/skills/brand-discovery/references/60_narrative-story.md +50 -50
  117. package/skills/brand-discovery/references/70_founder-tension.md +49 -49
  118. package/skills/brand-discovery/references/90_SYNTHESIS.md +133 -133
  119. package/skills/brand-voice/SKILL.md +98 -98
  120. package/skills/brand-voice/references/voice-profile-schema.md +55 -55
  121. package/skills/browser-qa/SKILL.md +105 -105
  122. package/skills/bun-runtime/SKILL.md +85 -85
  123. package/skills/canary-watch/SKILL.md +108 -108
  124. package/skills/carrier-relationship-management/SKILL.md +212 -212
  125. package/skills/cisco-ios-patterns/SKILL.md +164 -164
  126. package/skills/ck/SKILL.md +148 -148
  127. package/skills/claude-devfleet/SKILL.md +112 -112
  128. package/skills/click-path-audit/SKILL.md +245 -245
  129. package/skills/clickhouse-io/SKILL.md +445 -445
  130. package/skills/code-tour/SKILL.md +254 -254
  131. package/skills/codebase-onboarding/SKILL.md +234 -234
  132. package/skills/codehealth-mcp/SKILL.md +167 -167
  133. package/skills/coding-standards/SKILL.md +551 -551
  134. package/skills/competitive-platform-analysis/SKILL.md +214 -214
  135. package/skills/competitive-report-structure/SKILL.md +162 -162
  136. package/skills/compose-multiplatform-patterns/SKILL.md +300 -300
  137. package/skills/config-gc/SKILL.md +120 -120
  138. package/skills/configure-ecc/SKILL.md +385 -385
  139. package/skills/connections-optimizer/SKILL.md +190 -190
  140. package/skills/content-engine/SKILL.md +132 -132
  141. package/skills/content-hash-cache-pattern/SKILL.md +162 -162
  142. package/skills/context-budget/SKILL.md +136 -136
  143. package/skills/continuous-agent-loop/SKILL.md +46 -46
  144. package/skills/continuous-learning/SKILL.md +132 -132
  145. package/skills/continuous-learning-v2/SKILL.md +361 -361
  146. package/skills/contract-first/SKILL.md +287 -287
  147. package/skills/cost-aware-llm-pipeline/SKILL.md +184 -184
  148. package/skills/cost-tracking/SKILL.md +97 -97
  149. package/skills/council/SKILL.md +204 -204
  150. package/skills/cpp-coding-standards/SKILL.md +724 -724
  151. package/skills/cpp-testing/SKILL.md +325 -325
  152. package/skills/crosspost/SKILL.md +112 -112
  153. package/skills/csharp-testing/SKILL.md +322 -322
  154. package/skills/customer-billing-ops/SKILL.md +141 -141
  155. package/skills/customs-trade-compliance/SKILL.md +263 -263
  156. package/skills/dart-flutter-patterns/SKILL.md +564 -564
  157. package/skills/dashboard-builder/SKILL.md +109 -109
  158. package/skills/data-scraper-agent/SKILL.md +765 -765
  159. package/skills/data-throughput-accelerator/SKILL.md +74 -74
  160. package/skills/database-migrations/SKILL.md +430 -430
  161. package/skills/deep-research/SKILL.md +160 -160
  162. package/skills/defi-amm-security/SKILL.md +167 -167
  163. package/skills/delivery-gate/SKILL.md +126 -126
  164. package/skills/deployment-patterns/SKILL.md +428 -428
  165. package/skills/design-system/SKILL.md +83 -83
  166. package/skills/django-celery/SKILL.md +458 -458
  167. package/skills/django-patterns/SKILL.md +735 -735
  168. package/skills/django-security/SKILL.md +644 -644
  169. package/skills/django-tdd/SKILL.md +730 -730
  170. package/skills/django-verification/SKILL.md +470 -470
  171. package/skills/dmux-workflows/SKILL.md +192 -192
  172. package/skills/docker-patterns/SKILL.md +365 -365
  173. package/skills/documentation-lookup/SKILL.md +91 -91
  174. package/skills/dotnet-patterns/SKILL.md +322 -322
  175. package/skills/dynamic-workflow-mode/SKILL.md +124 -124
  176. package/skills/e2e-testing/SKILL.md +327 -327
  177. package/skills/ecc-guide/SKILL.md +190 -190
  178. package/skills/ecc-recipes/SKILL.md +149 -149
  179. package/skills/ecc-tools-cost-audit/SKILL.md +161 -161
  180. package/skills/email-ops/SKILL.md +122 -122
  181. package/skills/energy-procurement/SKILL.md +228 -228
  182. package/skills/enterprise-agent-ops/SKILL.md +51 -51
  183. package/skills/error-handling/SKILL.md +377 -377
  184. package/skills/eval-harness/SKILL.md +271 -271
  185. package/skills/evm-token-decimals/SKILL.md +131 -131
  186. package/skills/exa-search/SKILL.md +108 -108
  187. package/skills/fal-ai-media/SKILL.md +289 -289
  188. package/skills/fastapi-patterns/SKILL.md +514 -514
  189. package/skills/finance-billing-ops/SKILL.md +128 -128
  190. package/skills/flox-environments/SKILL.md +497 -497
  191. package/skills/flutter-dart-code-review/SKILL.md +436 -436
  192. package/skills/foundation-models-on-device/SKILL.md +243 -243
  193. package/skills/frontend-a11y/SKILL.md +446 -446
  194. package/skills/frontend-design-direction/SKILL.md +93 -93
  195. package/skills/frontend-patterns/SKILL.md +657 -657
  196. package/skills/frontend-slides/SKILL.md +185 -185
  197. package/skills/fsharp-testing/SKILL.md +281 -281
  198. package/skills/gan-style-harness/SKILL.md +279 -279
  199. package/skills/gateguard/SKILL.md +133 -133
  200. package/skills/generating-python-installer/SKILL.md +820 -820
  201. package/skills/git-workflow/SKILL.md +716 -716
  202. package/skills/github-ops/SKILL.md +145 -145
  203. package/skills/golang-patterns/SKILL.md +676 -676
  204. package/skills/golang-testing/SKILL.md +721 -721
  205. package/skills/google-workspace-ops/SKILL.md +96 -96
  206. package/skills/growth-log/SKILL.md +128 -128
  207. package/skills/healthcare-cdss-patterns/SKILL.md +246 -246
  208. package/skills/healthcare-emr-patterns/SKILL.md +160 -160
  209. package/skills/healthcare-eval-harness/SKILL.md +208 -208
  210. package/skills/healthcare-phi-compliance/SKILL.md +146 -146
  211. package/skills/hermes-imports/SKILL.md +89 -89
  212. package/skills/hexagonal-architecture/SKILL.md +277 -277
  213. package/skills/hipaa-compliance/SKILL.md +79 -79
  214. package/skills/homelab-network-readiness/SKILL.md +170 -170
  215. package/skills/homelab-network-setup/SKILL.md +130 -130
  216. package/skills/homelab-pihole-dns/SKILL.md +275 -275
  217. package/skills/homelab-vlan-segmentation/SKILL.md +312 -312
  218. package/skills/homelab-wireguard-vpn/SKILL.md +306 -306
  219. package/skills/hookify-rules/SKILL.md +128 -128
  220. package/skills/inherit-legacy-style/SKILL.md +157 -157
  221. package/skills/intent-driven-development/SKILL.md +360 -360
  222. package/skills/inventory-demand-planning/SKILL.md +247 -247
  223. package/skills/investor-materials/SKILL.md +97 -97
  224. package/skills/investor-outreach/SKILL.md +92 -92
  225. package/skills/ios-icon-gen/SKILL.md +158 -158
  226. package/skills/iterative-retrieval/SKILL.md +212 -212
  227. package/skills/ito-basket-compare/SKILL.md +64 -64
  228. package/skills/ito-compute/SKILL.md +129 -129
  229. package/skills/ito-data-atlas-agent/SKILL.md +64 -64
  230. package/skills/ito-market-intelligence/SKILL.md +61 -61
  231. package/skills/ito-trade-planner/SKILL.md +68 -68
  232. package/skills/java-coding-standards/SKILL.md +384 -384
  233. package/skills/jira-integration/SKILL.md +303 -303
  234. package/skills/jpa-patterns/SKILL.md +152 -152
  235. package/skills/knowledge-ops/SKILL.md +155 -155
  236. package/skills/kotlin-coroutines-flows/SKILL.md +285 -285
  237. package/skills/kotlin-exposed-patterns/SKILL.md +720 -720
  238. package/skills/kotlin-ktor-patterns/SKILL.md +690 -690
  239. package/skills/kotlin-patterns/SKILL.md +712 -712
  240. package/skills/kotlin-testing/SKILL.md +825 -825
  241. package/skills/kubernetes-patterns/SKILL.md +756 -756
  242. package/skills/laravel-patterns/SKILL.md +416 -416
  243. package/skills/laravel-plugin-discovery/SKILL.md +230 -230
  244. package/skills/laravel-security/SKILL.md +948 -948
  245. package/skills/laravel-tdd/SKILL.md +675 -675
  246. package/skills/laravel-verification/SKILL.md +180 -180
  247. package/skills/latency-critical-systems/SKILL.md +75 -75
  248. package/skills/lead-intelligence/SKILL.md +322 -322
  249. package/skills/liquid-glass-design/SKILL.md +279 -279
  250. package/skills/llm-trading-agent-security/SKILL.md +147 -147
  251. package/skills/logistics-exception-management/SKILL.md +222 -222
  252. package/skills/loop-design-check/SKILL.md +143 -143
  253. package/skills/mailtrap-email-integration/SKILL.md +77 -77
  254. package/skills/make-interfaces-feel-better/SKILL.md +152 -152
  255. package/skills/manim-video/SKILL.md +90 -90
  256. package/skills/market-research/SKILL.md +76 -76
  257. package/skills/marketing-campaign/SKILL.md +114 -114
  258. package/skills/mcp-server-patterns/SKILL.md +70 -70
  259. package/skills/messages-ops/SKILL.md +105 -105
  260. package/skills/ml-adoption-playbook/SKILL.md +57 -57
  261. package/skills/mle-workflow/SKILL.md +348 -348
  262. package/skills/motion-advanced/SKILL.md +596 -596
  263. package/skills/motion-foundations/SKILL.md +299 -299
  264. package/skills/motion-patterns/SKILL.md +434 -434
  265. package/skills/motion-ui/SKILL.md +576 -576
  266. package/skills/mysql-patterns/SKILL.md +413 -413
  267. package/skills/nanoclaw-repl/SKILL.md +34 -34
  268. package/skills/nestjs-patterns/SKILL.md +231 -231
  269. package/skills/netmiko-ssh-automation/SKILL.md +174 -174
  270. package/skills/network-bgp-diagnostics/SKILL.md +168 -168
  271. package/skills/network-config-validation/SKILL.md +211 -211
  272. package/skills/network-interface-health/SKILL.md +153 -153
  273. package/skills/nextjs-turbopack/SKILL.md +58 -58
  274. package/skills/nodejs-keccak256/SKILL.md +103 -103
  275. package/skills/nutrient-document-processing/SKILL.md +168 -168
  276. package/skills/nuxt4-patterns/SKILL.md +101 -101
  277. package/skills/openclaw-persona-forge/SKILL.md +289 -289
  278. package/skills/openclaw-persona-forge/references/avatar-style.md +124 -124
  279. package/skills/openclaw-persona-forge/references/boundary-rules.md +53 -53
  280. package/skills/openclaw-persona-forge/references/error-handling.md +53 -53
  281. package/skills/openclaw-persona-forge/references/identity-tension.md +48 -48
  282. package/skills/openclaw-persona-forge/references/naming-system.md +39 -39
  283. package/skills/openclaw-persona-forge/references/output-template.md +166 -166
  284. package/skills/opensource-pipeline/SKILL.md +256 -256
  285. package/skills/orch-add-feature/SKILL.md +45 -45
  286. package/skills/orch-build-mvp/SKILL.md +49 -49
  287. package/skills/orch-change-feature/SKILL.md +43 -43
  288. package/skills/orch-fix-defect/SKILL.md +43 -43
  289. package/skills/orch-pipeline/SKILL.md +121 -121
  290. package/skills/orch-refine-code/SKILL.md +44 -44
  291. package/skills/parallel-execution-optimizer/SKILL.md +74 -74
  292. package/skills/perl-patterns/SKILL.md +505 -505
  293. package/skills/perl-security/SKILL.md +504 -504
  294. package/skills/perl-testing/SKILL.md +476 -476
  295. package/skills/plan-canvas/SKILL.md +153 -153
  296. package/skills/plan-orchestrate/SKILL.md +263 -263
  297. package/skills/plankton-code-quality/SKILL.md +237 -237
  298. package/skills/postgres-patterns/SKILL.md +148 -148
  299. package/skills/prediction-market-oracle-research/SKILL.md +64 -64
  300. package/skills/prediction-market-risk-review/SKILL.md +61 -61
  301. package/skills/premium-web-design/SKILL.md +154 -154
  302. package/skills/prisma-patterns/SKILL.md +401 -401
  303. package/skills/product-capability/SKILL.md +142 -142
  304. package/skills/product-lens/SKILL.md +93 -93
  305. package/skills/production-audit/SKILL.md +207 -207
  306. package/skills/production-scheduling/SKILL.md +238 -238
  307. package/skills/project-flow-ops/SKILL.md +112 -112
  308. package/skills/prompt-optimizer/SKILL.md +398 -398
  309. package/skills/publish-website/SKILL.md +76 -0
  310. package/skills/python-patterns/SKILL.md +751 -751
  311. package/skills/python-testing/SKILL.md +817 -817
  312. package/skills/pytorch-patterns/SKILL.md +397 -397
  313. package/skills/quality-nonconformance/SKILL.md +260 -260
  314. package/skills/quarkus-patterns/SKILL.md +723 -723
  315. package/skills/quarkus-security/SKILL.md +468 -468
  316. package/skills/quarkus-tdd/SKILL.md +812 -812
  317. package/skills/quarkus-verification/SKILL.md +480 -480
  318. package/skills/ralphinho-rfc-pipeline/SKILL.md +68 -68
  319. package/skills/react-native-patterns/SKILL.md +326 -326
  320. package/skills/react-patterns/SKILL.md +342 -342
  321. package/skills/react-performance/SKILL.md +575 -575
  322. package/skills/react-testing/SKILL.md +424 -424
  323. package/skills/recsys-pipeline-architect/SKILL.md +115 -115
  324. package/skills/recursive-decision-ledger/SKILL.md +81 -81
  325. package/skills/redis-patterns/SKILL.md +404 -404
  326. package/skills/regex-vs-llm-structured-text/SKILL.md +221 -221
  327. package/skills/remotion-video-creation/SKILL.md +43 -43
  328. package/skills/repo-scan/SKILL.md +79 -79
  329. package/skills/research-ops/SKILL.md +113 -113
  330. package/skills/returns-reverse-logistics/SKILL.md +240 -240
  331. package/skills/rules-distill/SKILL.md +265 -265
  332. package/skills/rust-patterns/SKILL.md +500 -500
  333. package/skills/rust-testing/SKILL.md +501 -501
  334. package/skills/safety-guard/SKILL.md +76 -76
  335. package/skills/santa-method/SKILL.md +307 -307
  336. package/skills/scientific-db-pubmed-database/SKILL.md +176 -176
  337. package/skills/scientific-db-uspto-database/SKILL.md +178 -178
  338. package/skills/scientific-pkg-gget/SKILL.md +167 -167
  339. package/skills/scientific-thinking-literature-review/SKILL.md +193 -193
  340. package/skills/scientific-thinking-scholar-evaluation/SKILL.md +161 -161
  341. package/skills/search-first/SKILL.md +183 -183
  342. package/skills/security-bounty-hunter/SKILL.md +100 -100
  343. package/skills/security-review/SKILL.md +504 -504
  344. package/skills/security-scan/SKILL.md +166 -166
  345. package/skills/seo/SKILL.md +155 -155
  346. package/skills/skill-comply/SKILL.md +59 -59
  347. package/skills/skill-scout/SKILL.md +141 -141
  348. package/skills/skill-stocktake/SKILL.md +195 -195
  349. package/skills/social-graph-ranker/SKILL.md +155 -155
  350. package/skills/social-publisher/SKILL.md +130 -130
  351. package/skills/springboot-patterns/SKILL.md +315 -315
  352. package/skills/springboot-security/SKILL.md +273 -273
  353. package/skills/springboot-tdd/SKILL.md +159 -159
  354. package/skills/springboot-verification/SKILL.md +232 -232
  355. package/skills/strategic-compact/SKILL.md +142 -142
  356. package/skills/swift-actor-persistence/SKILL.md +144 -144
  357. package/skills/swift-concurrency-6-2/SKILL.md +216 -216
  358. package/skills/swift-protocol-di-testing/SKILL.md +191 -191
  359. package/skills/swiftui-patterns/SKILL.md +259 -259
  360. package/skills/taste/SKILL.md +264 -264
  361. package/skills/taste/references/genre-taxonomy.md +87 -87
  362. package/skills/tdd-workflow/SKILL.md +583 -583
  363. package/skills/team-agent-orchestration/SKILL.md +111 -111
  364. package/skills/team-builder/SKILL.md +169 -169
  365. package/skills/terminal-ops/SKILL.md +110 -110
  366. package/skills/tinystruct-patterns/SKILL.md +279 -279
  367. package/skills/tinystruct-patterns/references/architecture.md +90 -90
  368. package/skills/tinystruct-patterns/references/data-handling.md +60 -60
  369. package/skills/tinystruct-patterns/references/database.md +99 -99
  370. package/skills/tinystruct-patterns/references/routing.md +64 -64
  371. package/skills/tinystruct-patterns/references/system-usage.md +97 -97
  372. package/skills/tinystruct-patterns/references/testing.md +72 -72
  373. package/skills/token-budget-advisor/SKILL.md +134 -134
  374. package/skills/ui-demo/SKILL.md +466 -466
  375. package/skills/ui-to-vue/SKILL.md +135 -135
  376. package/skills/uncloud/SKILL.md +344 -344
  377. package/skills/unified-memory/SKILL.md +170 -170
  378. package/skills/unified-notifications-ops/SKILL.md +188 -188
  379. package/skills/verification-loop/SKILL.md +129 -129
  380. package/skills/video-editing/SKILL.md +311 -311
  381. package/skills/videodb/SKILL.md +375 -375
  382. package/skills/visa-doc-translate/SKILL.md +117 -117
  383. package/skills/vite-patterns/SKILL.md +450 -450
  384. package/skills/vue-patterns/SKILL.md +471 -471
  385. package/skills/windows-desktop-e2e/SKILL.md +888 -888
  386. package/skills/workspace-surface-audit/SKILL.md +126 -126
  387. package/skills/x-api/SKILL.md +235 -235
@@ -1,49 +1,49 @@
1
- ---
2
- name: orch-build-mvp
3
- description: Orchestrate bootstrapping a working MVP from a design or spec document — ingest the doc, plan thin vertical slices, scaffold the first end-to-end slice, then TDD-implement, review, and gated commit. Use to turn an SDD/PRD into a running starting point.
4
- metadata:
5
- origin: ECC
6
- ---
7
-
8
- # orch-build-mvp
9
-
10
- Actor · action · target: **orch · build · mvp**. Thin wrapper over the shared
11
- engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
-
13
- ## When to Use
14
-
15
- - The user has a **design / spec document** (SDD, PRD, system_design) and wants a
16
- working vertical slice bootstrapped from it.
17
- - Takes a doc path as its argument, e.g. `civicpulse/docs/SDD-v0.6.md`.
18
-
19
- ## Operation settings
20
-
21
- - **Default size floor:** large — this is the full pipeline including Scaffold.
22
- - **Phase mask:** 0 (read the spec) → 1 → 2 (heavy) → 3 (scaffold) → 4 → 5 → 6.
23
- - **First move (phase 0 → 2):** read the doc; extract scope, locked decisions,
24
- and the feature list; order it into **thin vertical slices** (one end-to-end
25
- path first, not all-models-then-all-views). Phase 3 stands up that first slice.
26
-
27
- ## How It Works
28
-
29
- 1. Run the `orch-pipeline` engine with the settings above.
30
- 2. **Reuse the existing GAN harness** instead of hand-rolling an iterate loop:
31
- - Translate the SDD into `gan-harness/spec.md` + `gan-harness/eval-rubric.md`
32
- (this stands in for what `gan-planner` would generate — you already have the spec).
33
- - Drive the build with `/gan-build "<one-line brief>" --skip-planner`
34
- (defaults: `--max-iterations 15`, `--pass-threshold 7.0`,
35
- `--eval-mode playwright`; use `--eval-mode code-only` for non-UI slices).
36
- - That command runs the `gan-generator` → `gan-evaluator` loop and writes
37
- `gan-harness/feedback/feedback-NNN.md` until the score passes or plateaus.
38
- 3. Stop at **Gate 1** (slice plan) and **Gate 2** (pre-commit). Commit the
39
- scaffold and each slice as separate `feat:` commits.
40
- 4. Add `security-reviewer` for any slice touching a security trigger.
41
-
42
- ## Example
43
-
44
- ```
45
- orch-build-mvp: civicpulse/docs/SDD-v0.6.md
46
- → read SDD → slice list (vertical) → scaffold slice 1 [GATE 1: approve]
47
- → /gan-build --skip-planner (generator → evaluator loop) scores vs spec → review
48
- → commit feat: [GATE 2: confirm] → next slice
49
- ```
1
+ ---
2
+ name: orch-build-mvp
3
+ description: Orchestrate bootstrapping a working MVP from a design or spec document — ingest the doc, plan thin vertical slices, scaffold the first end-to-end slice, then TDD-implement, review, and gated commit. Use to turn an SDD/PRD into a running starting point.
4
+ metadata:
5
+ origin: ECC
6
+ ---
7
+
8
+ # orch-build-mvp
9
+
10
+ Actor · action · target: **orch · build · mvp**. Thin wrapper over the shared
11
+ engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
+
13
+ ## When to Use
14
+
15
+ - The user has a **design / spec document** (SDD, PRD, system_design) and wants a
16
+ working vertical slice bootstrapped from it.
17
+ - Takes a doc path as its argument, e.g. `civicpulse/docs/SDD-v0.6.md`.
18
+
19
+ ## Operation settings
20
+
21
+ - **Default size floor:** large — this is the full pipeline including Scaffold.
22
+ - **Phase mask:** 0 (read the spec) → 1 → 2 (heavy) → 3 (scaffold) → 4 → 5 → 6.
23
+ - **First move (phase 0 → 2):** read the doc; extract scope, locked decisions,
24
+ and the feature list; order it into **thin vertical slices** (one end-to-end
25
+ path first, not all-models-then-all-views). Phase 3 stands up that first slice.
26
+
27
+ ## How It Works
28
+
29
+ 1. Run the `orch-pipeline` engine with the settings above.
30
+ 2. **Reuse the existing GAN harness** instead of hand-rolling an iterate loop:
31
+ - Translate the SDD into `gan-harness/spec.md` + `gan-harness/eval-rubric.md`
32
+ (this stands in for what `gan-planner` would generate — you already have the spec).
33
+ - Drive the build with `/gan-build "<one-line brief>" --skip-planner`
34
+ (defaults: `--max-iterations 15`, `--pass-threshold 7.0`,
35
+ `--eval-mode playwright`; use `--eval-mode code-only` for non-UI slices).
36
+ - That command runs the `gan-generator` → `gan-evaluator` loop and writes
37
+ `gan-harness/feedback/feedback-NNN.md` until the score passes or plateaus.
38
+ 3. Stop at **Gate 1** (slice plan) and **Gate 2** (pre-commit). Commit the
39
+ scaffold and each slice as separate `feat:` commits.
40
+ 4. Add `security-reviewer` for any slice touching a security trigger.
41
+
42
+ ## Example
43
+
44
+ ```
45
+ orch-build-mvp: civicpulse/docs/SDD-v0.6.md
46
+ → read SDD → slice list (vertical) → scaffold slice 1 [GATE 1: approve]
47
+ → /gan-build --skip-planner (generator → evaluator loop) scores vs spec → review
48
+ → commit feat: [GATE 2: confirm] → next slice
49
+ ```
@@ -1,43 +1,43 @@
1
- ---
2
- name: orch-change-feature
3
- description: Orchestrate altering an existing, working feature to new desired behavior — update its tests to the new spec, change the implementation to match, review, and gated commit. Use when behavior is not broken but should be different.
4
- metadata:
5
- origin: ECC
6
- ---
7
-
8
- # orch-change-feature
9
-
10
- Actor · action · target: **orch · change · feature**. Thin wrapper over the
11
- shared engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
-
13
- ## When to Use
14
-
15
- - An existing feature **works**, but the desired behavior is different ("change",
16
- "adjust", "make it also …", "instead of X do Y").
17
- - Distinguish from siblings:
18
- - **not** broken → not `orch-fix-defect` (no bug to reproduce).
19
- - **not** new → not `orch-add-feature` (the capability already exists).
20
-
21
- ## Operation settings
22
-
23
- - **Default size floor:** small — most tweaks are a function or two.
24
- - **Phase mask:** 0 → (1 only if the new behavior needs research) → light 2 →
25
- 4 → 5 → 6.
26
- - **First move (phase 4):** update the *existing* tests to express the new
27
- desired behavior, then change the implementation until they pass. Changing the
28
- tests first is what separates a tweak from a fix.
29
-
30
- ## How It Works
31
-
32
- 1. Run the `orch-pipeline` engine with the settings above.
33
- 2. Keep the plan light — only `standard`+ size warrants the full `planner` pass.
34
- 3. Stop at **Gate 1** (plan / changed-test approval) and **Gate 2** (pre-commit).
35
- 4. Add `security-reviewer` if the change touches a security trigger.
36
-
37
- ## Example
38
-
39
- ```
40
- orch-change-feature: make nws-poller alert at 2 warnings instead of 3
41
- → update threshold tests to new spec → change impl to green
42
- → code-review → commit [GATE 2: confirm]
43
- ```
1
+ ---
2
+ name: orch-change-feature
3
+ description: Orchestrate altering an existing, working feature to new desired behavior — update its tests to the new spec, change the implementation to match, review, and gated commit. Use when behavior is not broken but should be different.
4
+ metadata:
5
+ origin: ECC
6
+ ---
7
+
8
+ # orch-change-feature
9
+
10
+ Actor · action · target: **orch · change · feature**. Thin wrapper over the
11
+ shared engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
+
13
+ ## When to Use
14
+
15
+ - An existing feature **works**, but the desired behavior is different ("change",
16
+ "adjust", "make it also …", "instead of X do Y").
17
+ - Distinguish from siblings:
18
+ - **not** broken → not `orch-fix-defect` (no bug to reproduce).
19
+ - **not** new → not `orch-add-feature` (the capability already exists).
20
+
21
+ ## Operation settings
22
+
23
+ - **Default size floor:** small — most tweaks are a function or two.
24
+ - **Phase mask:** 0 → (1 only if the new behavior needs research) → light 2 →
25
+ 4 → 5 → 6.
26
+ - **First move (phase 4):** update the *existing* tests to express the new
27
+ desired behavior, then change the implementation until they pass. Changing the
28
+ tests first is what separates a tweak from a fix.
29
+
30
+ ## How It Works
31
+
32
+ 1. Run the `orch-pipeline` engine with the settings above.
33
+ 2. Keep the plan light — only `standard`+ size warrants the full `planner` pass.
34
+ 3. Stop at **Gate 1** (plan / changed-test approval) and **Gate 2** (pre-commit).
35
+ 4. Add `security-reviewer` if the change touches a security trigger.
36
+
37
+ ## Example
38
+
39
+ ```
40
+ orch-change-feature: make nws-poller alert at 2 warnings instead of 3
41
+ → update threshold tests to new spec → change impl to green
42
+ → code-review → commit [GATE 2: confirm]
43
+ ```
@@ -1,43 +1,43 @@
1
- ---
2
- name: orch-fix-defect
3
- description: Orchestrate fixing a bug — reproduce it as a failing regression test, fix to green, review, and gated commit — by delegating each phase to the matching ECC agent. Use when existing behavior is broken or wrong.
4
- metadata:
5
- origin: ECC
6
- ---
7
-
8
- # orch-fix-defect
9
-
10
- Actor · action · target: **orch · fix · defect**. Thin wrapper over the shared
11
- engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
-
13
- ## When to Use
14
-
15
- - Something is **broken**: wrong output, an error, a crash, a regression.
16
- - Distinguish from siblings:
17
- - behavior is correct but you want it different → `orch-change-feature`.
18
- - the capability does not exist yet → `orch-add-feature`.
19
-
20
- ## Operation settings
21
-
22
- - **Default size floor:** small (often trivial).
23
- - **Phase mask:** 0 → (light 2 only if root cause is non-obvious or standard+) →
24
- 4 → 5 → 6. Research (1) is usually skipped.
25
- - **First move (phase 4):** reproduce the bug as a **new failing** test
26
- (regression test), then fix until it goes green. Proving the bug exists first
27
- is what separates a fix from a tweak.
28
-
29
- ## How It Works
30
-
31
- 1. Run the `orch-pipeline` engine with the settings above.
32
- 2. If the root cause is unclear, scope it with `code-explorer` before the red
33
- test; escalate build breaks to `build-error-resolver` / `/build-fix`.
34
- 3. Stop at **Gate 1** (only if a plan was produced) and **Gate 2** (pre-commit).
35
- 4. Add `security-reviewer` if the defect sits in a security-sensitive path.
36
-
37
- ## Example
38
-
39
- ```
40
- orch-fix-defect: poller crashes on empty NWS response
41
- → write failing test reproducing the crash → fix to green
42
- → code-review → commit [GATE 2: confirm] (commit: fix:)
43
- ```
1
+ ---
2
+ name: orch-fix-defect
3
+ description: Orchestrate fixing a bug — reproduce it as a failing regression test, fix to green, review, and gated commit — by delegating each phase to the matching ECC agent. Use when existing behavior is broken or wrong.
4
+ metadata:
5
+ origin: ECC
6
+ ---
7
+
8
+ # orch-fix-defect
9
+
10
+ Actor · action · target: **orch · fix · defect**. Thin wrapper over the shared
11
+ engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
+
13
+ ## When to Use
14
+
15
+ - Something is **broken**: wrong output, an error, a crash, a regression.
16
+ - Distinguish from siblings:
17
+ - behavior is correct but you want it different → `orch-change-feature`.
18
+ - the capability does not exist yet → `orch-add-feature`.
19
+
20
+ ## Operation settings
21
+
22
+ - **Default size floor:** small (often trivial).
23
+ - **Phase mask:** 0 → (light 2 only if root cause is non-obvious or standard+) →
24
+ 4 → 5 → 6. Research (1) is usually skipped.
25
+ - **First move (phase 4):** reproduce the bug as a **new failing** test
26
+ (regression test), then fix until it goes green. Proving the bug exists first
27
+ is what separates a fix from a tweak.
28
+
29
+ ## How It Works
30
+
31
+ 1. Run the `orch-pipeline` engine with the settings above.
32
+ 2. If the root cause is unclear, scope it with `code-explorer` before the red
33
+ test; escalate build breaks to `build-error-resolver` / `/build-fix`.
34
+ 3. Stop at **Gate 1** (only if a plan was produced) and **Gate 2** (pre-commit).
35
+ 4. Add `security-reviewer` if the defect sits in a security-sensitive path.
36
+
37
+ ## Example
38
+
39
+ ```
40
+ orch-fix-defect: poller crashes on empty NWS response
41
+ → write failing test reproducing the crash → fix to green
42
+ → code-review → commit [GATE 2: confirm] (commit: fix:)
43
+ ```
@@ -1,121 +1,121 @@
1
- ---
2
- name: orch-pipeline
3
- description: Shared orchestration engine for the orch-* skill family. Defines the gated Research-Plan-TDD-Review-Commit pipeline, the size classifier, the agent map, and the two human gates that the orch-* operation skills delegate to. Not usually invoked directly.
4
- metadata:
5
- origin: ECC
6
- ---
7
-
8
- # Orchestrator Pipeline (shared engine)
9
-
10
- The `orch-*` skills are thin wrappers. They do not re-implement any work — they
11
- classify the request, choose which phases of *this* pipeline run, and delegate
12
- each phase to an existing ECC agent or command. This file is that pipeline.
13
-
14
- > Invoke an operation skill (`orch-add-feature`, `orch-fix-defect`, …) rather
15
- > than this engine directly. This file is the reference they point at.
16
-
17
- ## When to Use
18
-
19
- - Loaded indirectly whenever an `orch-*` operation skill runs.
20
- - Read directly only when adding a new operation to the family or tuning the
21
- shared phases, gates, or agent map.
22
-
23
- ## The operation family
24
-
25
- | Skill | Operation | Trigger | First move |
26
- |-------|-----------|---------|------------|
27
- | `orch-add-feature` | feature | capability does not exist yet | research + plan a new slice |
28
- | `orch-change-feature` | tweak | works, but desired behavior differs | amend existing behavior *and its tests* |
29
- | `orch-fix-defect` | fix | broken; behavior is wrong | reproduce as a failing test, then fix |
30
- | `orch-refine-code` | refactor | behavior stays, structure improves | restructure while keeping tests green |
31
- | `orch-build-mvp` | mvp | bootstrap from a design/spec doc | ingest doc → vertical slices |
32
-
33
- > These wrappers **compose** existing ECC commands rather than replace them:
34
- > `/feature-dev`, `/plan`, `/code-review`, `/build-fix`, `/refactor-clean`, and
35
- > `/gan-build`, plus the `tdd-workflow` skill. The orch-* family adds the shared
36
- > size classifier and the two gates
37
- > on top of them, so one umbrella covers all five operations consistently.
38
-
39
- ## Step 0 — Classify size (right-sizing)
40
-
41
- Ceremony scales to blast radius. Score the request on three signals, take the
42
- **highest** tier any signal reaches, and state the result in one line so the user
43
- can override:
44
-
45
- | Tier | Files touched | New dependency / contract | Design ambiguity | Phases that run |
46
- |------|---------------|---------------------------|------------------|-----------------|
47
- | trivial | 1, a few lines | none | none — the change is obvious | 4 → 5 → 6 |
48
- | small | 1 file / 1 function | none | clear once you read the code | (1 light) → 4 → 5 → 6 |
49
- | standard | 2–5 files | maybe a new internal module | one real choice to make | 1 → 2 → 4 → 5 → 6 |
50
- | large | many / cross-cutting | new external dep, public API, or a spec doc | multiple open questions | 1 → 2 → (3) → 4 → 5 → 6 |
51
-
52
- Phase 0 (Intake) always runs and is omitted from the mask column above. The
53
- tie-breaker: anything touching a security trigger (below) or a public API /
54
- contract is **at least** standard, regardless of file count.
55
-
56
- ## The phases
57
-
58
- Each phase delegates — it does not do the work inline.
59
-
60
- - **0. Intake** — restate the request. For `orch-build-mvp`, read the spec/design
61
- doc and extract scope, locked decisions, and a feature list.
62
- - **1. Research & Reuse** — per `rules/common/development-workflow.md`: `gh search repos` /
63
- `gh search code`, then Context7 / vendor docs, then package registries, then
64
- Exa. Prefer adopting a proven implementation over net-new code.
65
- - **2. Plan** — delegate to the `planner` agent (or `architect` /
66
- `code-architect` for structural decisions). Output a `task_list` ordered as
67
- thin vertical slices. → **GATE 1.**
68
- - **3. Scaffold** — `orch-build-mvp` only: stand up the first end-to-end slice.
69
- - **4. Implement (TDD)** — drive each task through the `tdd-guide` agent (or the `tdd-workflow` skill):
70
- red → green → refactor. Honor the operation's first-move rule.
71
- - **5. Review** — `code-reviewer` agent / `/code-review`. Add `security-reviewer`
72
- whenever the diff touches a security trigger (below).
73
- - **6. Commit** — conventional commits (`feat:` / `fix:` / `refactor:` / …), one
74
- per logical chunk. → **GATE 2.**
75
-
76
- ## The two gates
77
-
78
- This family is **gated, not autonomous**:
79
-
80
- 1. **GATE 1 — after Plan.** Present the `task_list`; do not write implementation
81
- code until the user approves.
82
- 2. **GATE 2 — before Commit.** Present the diff summary and proposed messages;
83
- do not commit until the user confirms.
84
-
85
- Everything between the gates flows without stopping.
86
-
87
- ## Agent / command map
88
-
89
- | Phase | Primary | Fallback / escalation |
90
- |-------|---------|----------------------|
91
- | Intake / understand | `code-explorer` | trace existing paths before a tweak, fix, or refactor |
92
- | Plan | `planner` | `architect`, `code-architect` for structural calls |
93
- | Implement | `tdd-guide` (or `tdd-workflow` skill) | `build-error-resolver` / `/build-fix` on build breaks |
94
- | Review | `code-reviewer` / `/code-review` | language reviewer (`python-reviewer`, `typescript-reviewer`, …) |
95
- | Security | `security-reviewer` | — |
96
- | MVP inner loop | `/gan-build "<brief>" --skip-planner` | drives `gan-generator` → `gan-evaluator`; tune `--max-iterations` / `--pass-threshold` |
97
-
98
- Match the language reviewer to the repo (see the repo's own `CLAUDE.md`).
99
-
100
- ## Security-review trigger
101
-
102
- Pull in `security-reviewer` when the diff touches any of: authentication or
103
- authorization, user-input handling, database queries, file-system paths,
104
- external API calls, cryptography, or secrets / credentials. (Per `rules/common/security.md`.)
105
-
106
- ## Handoff artifacts
107
-
108
- The pipeline carries no hidden state — the planning docs *are* the handoff:
109
-
110
- - `task_list` (from Plan) drives the Implement loop.
111
- - Larger work may also emit PRD / architecture / system_design under the repo's
112
- `docs/` per `rules/common/development-workflow.md`.
113
- - Review findings (CRITICAL / HIGH) must be resolved before Gate 2.
114
-
115
- ## Verification
116
-
117
- - size tier was stated and matched the work
118
- - Gate 1 (plan) and Gate 2 (commit) were both honored
119
- - `security-reviewer` ran iff a security trigger was touched
120
- - commits are conventional and scoped to one logical change
121
- - new / changed behavior has tests; coverage ≥ 80% per `rules/common/testing.md`
1
+ ---
2
+ name: orch-pipeline
3
+ description: Shared orchestration engine for the orch-* skill family. Defines the gated Research-Plan-TDD-Review-Commit pipeline, the size classifier, the agent map, and the two human gates that the orch-* operation skills delegate to. Not usually invoked directly.
4
+ metadata:
5
+ origin: ECC
6
+ ---
7
+
8
+ # Orchestrator Pipeline (shared engine)
9
+
10
+ The `orch-*` skills are thin wrappers. They do not re-implement any work — they
11
+ classify the request, choose which phases of *this* pipeline run, and delegate
12
+ each phase to an existing ECC agent or command. This file is that pipeline.
13
+
14
+ > Invoke an operation skill (`orch-add-feature`, `orch-fix-defect`, …) rather
15
+ > than this engine directly. This file is the reference they point at.
16
+
17
+ ## When to Use
18
+
19
+ - Loaded indirectly whenever an `orch-*` operation skill runs.
20
+ - Read directly only when adding a new operation to the family or tuning the
21
+ shared phases, gates, or agent map.
22
+
23
+ ## The operation family
24
+
25
+ | Skill | Operation | Trigger | First move |
26
+ |-------|-----------|---------|------------|
27
+ | `orch-add-feature` | feature | capability does not exist yet | research + plan a new slice |
28
+ | `orch-change-feature` | tweak | works, but desired behavior differs | amend existing behavior *and its tests* |
29
+ | `orch-fix-defect` | fix | broken; behavior is wrong | reproduce as a failing test, then fix |
30
+ | `orch-refine-code` | refactor | behavior stays, structure improves | restructure while keeping tests green |
31
+ | `orch-build-mvp` | mvp | bootstrap from a design/spec doc | ingest doc → vertical slices |
32
+
33
+ > These wrappers **compose** existing ECC commands rather than replace them:
34
+ > `/feature-dev`, `/plan`, `/code-review`, `/build-fix`, `/refactor-clean`, and
35
+ > `/gan-build`, plus the `tdd-workflow` skill. The orch-* family adds the shared
36
+ > size classifier and the two gates
37
+ > on top of them, so one umbrella covers all five operations consistently.
38
+
39
+ ## Step 0 — Classify size (right-sizing)
40
+
41
+ Ceremony scales to blast radius. Score the request on three signals, take the
42
+ **highest** tier any signal reaches, and state the result in one line so the user
43
+ can override:
44
+
45
+ | Tier | Files touched | New dependency / contract | Design ambiguity | Phases that run |
46
+ |------|---------------|---------------------------|------------------|-----------------|
47
+ | trivial | 1, a few lines | none | none — the change is obvious | 4 → 5 → 6 |
48
+ | small | 1 file / 1 function | none | clear once you read the code | (1 light) → 4 → 5 → 6 |
49
+ | standard | 2–5 files | maybe a new internal module | one real choice to make | 1 → 2 → 4 → 5 → 6 |
50
+ | large | many / cross-cutting | new external dep, public API, or a spec doc | multiple open questions | 1 → 2 → (3) → 4 → 5 → 6 |
51
+
52
+ Phase 0 (Intake) always runs and is omitted from the mask column above. The
53
+ tie-breaker: anything touching a security trigger (below) or a public API /
54
+ contract is **at least** standard, regardless of file count.
55
+
56
+ ## The phases
57
+
58
+ Each phase delegates — it does not do the work inline.
59
+
60
+ - **0. Intake** — restate the request. For `orch-build-mvp`, read the spec/design
61
+ doc and extract scope, locked decisions, and a feature list.
62
+ - **1. Research & Reuse** — per `rules/common/development-workflow.md`: `gh search repos` /
63
+ `gh search code`, then Context7 / vendor docs, then package registries, then
64
+ Exa. Prefer adopting a proven implementation over net-new code.
65
+ - **2. Plan** — delegate to the `planner` agent (or `architect` /
66
+ `code-architect` for structural decisions). Output a `task_list` ordered as
67
+ thin vertical slices. → **GATE 1.**
68
+ - **3. Scaffold** — `orch-build-mvp` only: stand up the first end-to-end slice.
69
+ - **4. Implement (TDD)** — drive each task through the `tdd-guide` agent (or the `tdd-workflow` skill):
70
+ red → green → refactor. Honor the operation's first-move rule.
71
+ - **5. Review** — `code-reviewer` agent / `/code-review`. Add `security-reviewer`
72
+ whenever the diff touches a security trigger (below).
73
+ - **6. Commit** — conventional commits (`feat:` / `fix:` / `refactor:` / …), one
74
+ per logical chunk. → **GATE 2.**
75
+
76
+ ## The two gates
77
+
78
+ This family is **gated, not autonomous**:
79
+
80
+ 1. **GATE 1 — after Plan.** Present the `task_list`; do not write implementation
81
+ code until the user approves.
82
+ 2. **GATE 2 — before Commit.** Present the diff summary and proposed messages;
83
+ do not commit until the user confirms.
84
+
85
+ Everything between the gates flows without stopping.
86
+
87
+ ## Agent / command map
88
+
89
+ | Phase | Primary | Fallback / escalation |
90
+ |-------|---------|----------------------|
91
+ | Intake / understand | `code-explorer` | trace existing paths before a tweak, fix, or refactor |
92
+ | Plan | `planner` | `architect`, `code-architect` for structural calls |
93
+ | Implement | `tdd-guide` (or `tdd-workflow` skill) | `build-error-resolver` / `/build-fix` on build breaks |
94
+ | Review | `code-reviewer` / `/code-review` | language reviewer (`python-reviewer`, `typescript-reviewer`, …) |
95
+ | Security | `security-reviewer` | — |
96
+ | MVP inner loop | `/gan-build "<brief>" --skip-planner` | drives `gan-generator` → `gan-evaluator`; tune `--max-iterations` / `--pass-threshold` |
97
+
98
+ Match the language reviewer to the repo (see the repo's own `CLAUDE.md`).
99
+
100
+ ## Security-review trigger
101
+
102
+ Pull in `security-reviewer` when the diff touches any of: authentication or
103
+ authorization, user-input handling, database queries, file-system paths,
104
+ external API calls, cryptography, or secrets / credentials. (Per `rules/common/security.md`.)
105
+
106
+ ## Handoff artifacts
107
+
108
+ The pipeline carries no hidden state — the planning docs *are* the handoff:
109
+
110
+ - `task_list` (from Plan) drives the Implement loop.
111
+ - Larger work may also emit PRD / architecture / system_design under the repo's
112
+ `docs/` per `rules/common/development-workflow.md`.
113
+ - Review findings (CRITICAL / HIGH) must be resolved before Gate 2.
114
+
115
+ ## Verification
116
+
117
+ - size tier was stated and matched the work
118
+ - Gate 1 (plan) and Gate 2 (commit) were both honored
119
+ - `security-reviewer` ran iff a security trigger was touched
120
+ - commits are conventional and scoped to one logical change
121
+ - new / changed behavior has tests; coverage ≥ 80% per `rules/common/testing.md`
@@ -1,44 +1,44 @@
1
- ---
2
- name: orch-refine-code
3
- description: Orchestrate a behavior-preserving refactor — confirm tests are green, restructure without changing behavior, keep tests green, review, and gated commit. Use when the structure should improve but behavior must not change.
4
- metadata:
5
- origin: ECC
6
- ---
7
-
8
- # orch-refine-code
9
-
10
- Actor · action · target: **orch · refine · code**. Thin wrapper over the shared
11
- engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
-
13
- ## When to Use
14
-
15
- - Same behavior, **better structure**: extract modules, remove duplication, kill
16
- dead code, reduce nesting, rename for clarity.
17
- - Distinguish from siblings: if behavior is meant to change at all, this is the
18
- wrong skill (`orch-change-feature` / `orch-fix-defect`).
19
-
20
- ## Operation settings
21
-
22
- - **Default size floor:** standard — restructures touch multiple files.
23
- - **Phase mask:** 0 → 2 (plan the restructure) → 4 (keep green) → 5 → 6. No new
24
- behavior tests are written — the existing suite is the safety net.
25
- - **First move (phase 4):** confirm the relevant tests exist and are **green
26
- before** touching code; if coverage is thin, add characterization tests first.
27
- Then restructure in small steps, re-running tests after each.
28
-
29
- ## How It Works
30
-
31
- 1. Run the `orch-pipeline` engine with the settings above.
32
- 2. For dead-code / duplication sweeps, delegate to the `refactor-cleaner` agent
33
- (it runs knip / depcheck / ts-prune and removes safely).
34
- 3. Stop at **Gate 1** (restructure plan) and **Gate 2** (pre-commit).
35
- 4. Commit as `refactor:` — the diff must be behavior-neutral.
36
-
37
- ## Example
38
-
39
- ```
40
- orch-refine-code: extract the NWS HTTP client out of poller.py
41
- → confirm tests green → plan extraction [GATE 1: approve]
42
- → move in small steps, tests green throughout → code-review
43
- → commit refactor: [GATE 2: confirm]
44
- ```
1
+ ---
2
+ name: orch-refine-code
3
+ description: Orchestrate a behavior-preserving refactor — confirm tests are green, restructure without changing behavior, keep tests green, review, and gated commit. Use when the structure should improve but behavior must not change.
4
+ metadata:
5
+ origin: ECC
6
+ ---
7
+
8
+ # orch-refine-code
9
+
10
+ Actor · action · target: **orch · refine · code**. Thin wrapper over the shared
11
+ engine in [`orch-pipeline`](../orch-pipeline/SKILL.md).
12
+
13
+ ## When to Use
14
+
15
+ - Same behavior, **better structure**: extract modules, remove duplication, kill
16
+ dead code, reduce nesting, rename for clarity.
17
+ - Distinguish from siblings: if behavior is meant to change at all, this is the
18
+ wrong skill (`orch-change-feature` / `orch-fix-defect`).
19
+
20
+ ## Operation settings
21
+
22
+ - **Default size floor:** standard — restructures touch multiple files.
23
+ - **Phase mask:** 0 → 2 (plan the restructure) → 4 (keep green) → 5 → 6. No new
24
+ behavior tests are written — the existing suite is the safety net.
25
+ - **First move (phase 4):** confirm the relevant tests exist and are **green
26
+ before** touching code; if coverage is thin, add characterization tests first.
27
+ Then restructure in small steps, re-running tests after each.
28
+
29
+ ## How It Works
30
+
31
+ 1. Run the `orch-pipeline` engine with the settings above.
32
+ 2. For dead-code / duplication sweeps, delegate to the `refactor-cleaner` agent
33
+ (it runs knip / depcheck / ts-prune and removes safely).
34
+ 3. Stop at **Gate 1** (restructure plan) and **Gate 2** (pre-commit).
35
+ 4. Commit as `refactor:` — the diff must be behavior-neutral.
36
+
37
+ ## Example
38
+
39
+ ```
40
+ orch-refine-code: extract the NWS HTTP client out of poller.py
41
+ → confirm tests green → plan extraction [GATE 1: approve]
42
+ → move in small steps, tests green throughout → code-review
43
+ → commit refactor: [GATE 2: confirm]
44
+ ```