@rpamis/comet 0.4.2 → 0.4.4

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 (391) hide show
  1. package/LICENSE +21 -21
  2. package/assets/manifest.json +1 -1
  3. package/assets/skills/comet/scripts/comet-archive.mjs +109 -96
  4. package/assets/skills/comet/scripts/comet-check.mjs +117 -104
  5. package/assets/skills/comet/scripts/comet-entry-runtime.mjs +59 -59
  6. package/assets/skills/comet/scripts/comet-guard.mjs +173 -160
  7. package/assets/skills/comet/scripts/comet-handoff.mjs +110 -97
  8. package/assets/skills/comet/scripts/comet-hook-guard.mjs +110 -97
  9. package/assets/skills/comet/scripts/comet-hook-router.mjs +269 -159
  10. package/assets/skills/comet/scripts/comet-intent.mjs +31 -31
  11. package/assets/skills/comet/scripts/comet-resume-probe.mjs +104 -91
  12. package/assets/skills/comet/scripts/comet-runtime.mjs +272 -259
  13. package/assets/skills/comet/scripts/comet-state.mjs +204 -191
  14. package/assets/skills/comet/scripts/comet-yaml-validate.mjs +25 -25
  15. package/assets/skills/comet-archive/SKILL.md +4 -4
  16. package/assets/skills/comet-build/SKILL.md +4 -6
  17. package/assets/skills/comet-classic/SKILL.md +3 -1
  18. package/assets/skills/comet-classic/reference/scripts.md +17 -1
  19. package/assets/skills/comet-design/SKILL.md +0 -2
  20. package/assets/skills/comet-hotfix/SKILL.md +2 -3
  21. package/assets/skills/comet-native/SKILL.md +11 -8
  22. package/assets/skills/comet-native/reference/artifacts.md +7 -7
  23. package/assets/skills/comet-native/reference/clarification.md +4 -4
  24. package/assets/skills/comet-native/reference/commands.md +25 -13
  25. package/assets/skills/comet-native/reference/recovery.md +3 -3
  26. package/assets/skills/comet-native/reference/workspace.md +10 -4
  27. package/assets/skills/comet-native/scripts/comet-native-archive.mjs +230 -149
  28. package/assets/skills/comet-native/scripts/comet-native-doctor.mjs +239 -161
  29. package/assets/skills/comet-native/scripts/comet-native-hook-guard.mjs +232 -122
  30. package/assets/skills/comet-native/scripts/comet-native-init.mjs +184 -107
  31. package/assets/skills/comet-native/scripts/comet-native-new.mjs +219 -142
  32. package/assets/skills/comet-native/scripts/comet-native-next.mjs +239 -161
  33. package/assets/skills/comet-native/scripts/comet-native-root.mjs +205 -128
  34. package/assets/skills/comet-native/scripts/comet-native-runtime.mjs +276 -195
  35. package/assets/skills/comet-native/scripts/comet-native-select.mjs +204 -127
  36. package/assets/skills/comet-native/scripts/comet-native-show.mjs +202 -125
  37. package/assets/skills/comet-native/scripts/comet-native-spec.mjs +203 -125
  38. package/assets/skills/comet-native/scripts/comet-native-status.mjs +200 -123
  39. package/assets/skills/comet-open/SKILL.md +0 -2
  40. package/assets/skills/comet-tweak/SKILL.md +2 -1
  41. package/assets/skills/comet-verify/SKILL.md +0 -2
  42. package/assets/skills-zh/comet-archive/SKILL.md +4 -4
  43. package/assets/skills-zh/comet-build/SKILL.md +4 -6
  44. package/assets/skills-zh/comet-classic/SKILL.md +3 -1
  45. package/assets/skills-zh/comet-classic/reference/scripts.md +11 -1
  46. package/assets/skills-zh/comet-design/SKILL.md +0 -2
  47. package/assets/skills-zh/comet-hotfix/SKILL.md +2 -3
  48. package/assets/skills-zh/comet-native/SKILL.md +11 -8
  49. package/assets/skills-zh/comet-native/reference/artifacts.md +7 -7
  50. package/assets/skills-zh/comet-native/reference/clarification.md +4 -4
  51. package/assets/skills-zh/comet-native/reference/commands.md +26 -10
  52. package/assets/skills-zh/comet-native/reference/recovery.md +3 -3
  53. package/assets/skills-zh/comet-native/reference/workspace.md +10 -4
  54. package/assets/skills-zh/comet-open/SKILL.md +0 -2
  55. package/assets/skills-zh/comet-tweak/SKILL.md +2 -1
  56. package/assets/skills-zh/comet-verify/SKILL.md +0 -2
  57. package/bin/comet-daemon-router.js +57 -13
  58. package/bin/fast-runtime-router.js +1 -1
  59. package/dist/app/commands/daemon-server.d.ts.map +1 -1
  60. package/dist/app/commands/daemon-server.js +14 -2
  61. package/dist/app/commands/daemon-server.js.map +1 -1
  62. package/dist/app/commands/project-knowledge.d.ts.map +1 -1
  63. package/dist/app/commands/project-knowledge.js +0 -1
  64. package/dist/app/commands/project-knowledge.js.map +1 -1
  65. package/dist/domains/comet-classic/classic-archive-entry.d.ts +1 -1
  66. package/dist/domains/comet-classic/classic-archive-entry.d.ts.map +1 -1
  67. package/dist/domains/comet-classic/classic-archive-entry.js +5 -2
  68. package/dist/domains/comet-classic/classic-archive-entry.js.map +1 -1
  69. package/dist/domains/comet-classic/classic-branch-binding.d.ts.map +1 -1
  70. package/dist/domains/comet-classic/classic-branch-binding.js +5 -4
  71. package/dist/domains/comet-classic/classic-branch-binding.js.map +1 -1
  72. package/dist/domains/comet-classic/classic-check-entry.d.ts +1 -1
  73. package/dist/domains/comet-classic/classic-check-entry.d.ts.map +1 -1
  74. package/dist/domains/comet-classic/classic-check-entry.js +5 -2
  75. package/dist/domains/comet-classic/classic-check-entry.js.map +1 -1
  76. package/dist/domains/comet-classic/classic-check-snapshot.d.ts +2 -0
  77. package/dist/domains/comet-classic/classic-check-snapshot.d.ts.map +1 -1
  78. package/dist/domains/comet-classic/classic-check-snapshot.js +90 -27
  79. package/dist/domains/comet-classic/classic-check-snapshot.js.map +1 -1
  80. package/dist/domains/comet-classic/classic-guard-entry.d.ts +1 -1
  81. package/dist/domains/comet-classic/classic-guard-entry.d.ts.map +1 -1
  82. package/dist/domains/comet-classic/classic-guard-entry.js +5 -2
  83. package/dist/domains/comet-classic/classic-guard-entry.js.map +1 -1
  84. package/dist/domains/comet-classic/classic-handoff-entry.d.ts +1 -1
  85. package/dist/domains/comet-classic/classic-handoff-entry.d.ts.map +1 -1
  86. package/dist/domains/comet-classic/classic-handoff-entry.js +5 -2
  87. package/dist/domains/comet-classic/classic-handoff-entry.js.map +1 -1
  88. package/dist/domains/comet-classic/classic-resume-probe.d.ts.map +1 -1
  89. package/dist/domains/comet-classic/classic-resume-probe.js +3 -2
  90. package/dist/domains/comet-classic/classic-resume-probe.js.map +1 -1
  91. package/dist/domains/comet-classic/classic-script-entry.d.ts +3 -1
  92. package/dist/domains/comet-classic/classic-script-entry.d.ts.map +1 -1
  93. package/dist/domains/comet-classic/classic-script-entry.js +34 -3
  94. package/dist/domains/comet-classic/classic-script-entry.js.map +1 -1
  95. package/dist/domains/comet-classic/classic-state-command.d.ts.map +1 -1
  96. package/dist/domains/comet-classic/classic-state-command.js +3 -2
  97. package/dist/domains/comet-classic/classic-state-command.js.map +1 -1
  98. package/dist/domains/comet-classic/classic-state-entry.d.ts +1 -1
  99. package/dist/domains/comet-classic/classic-state-entry.d.ts.map +1 -1
  100. package/dist/domains/comet-classic/classic-state-entry.js +5 -2
  101. package/dist/domains/comet-classic/classic-state-entry.js.map +1 -1
  102. package/dist/domains/comet-classic/classic-store.d.ts.map +1 -1
  103. package/dist/domains/comet-classic/classic-store.js +67 -1
  104. package/dist/domains/comet-classic/classic-store.js.map +1 -1
  105. package/dist/domains/comet-entry/resume-probe.d.ts.map +1 -1
  106. package/dist/domains/comet-entry/resume-probe.js +3 -2
  107. package/dist/domains/comet-entry/resume-probe.js.map +1 -1
  108. package/dist/domains/comet-native/native-acceptance.d.ts +5 -1
  109. package/dist/domains/comet-native/native-acceptance.d.ts.map +1 -1
  110. package/dist/domains/comet-native/native-acceptance.js +13 -2
  111. package/dist/domains/comet-native/native-acceptance.js.map +1 -1
  112. package/dist/domains/comet-native/native-archive-command.d.ts.map +1 -1
  113. package/dist/domains/comet-native/native-archive-command.js +167 -37
  114. package/dist/domains/comet-native/native-archive-command.js.map +1 -1
  115. package/dist/domains/comet-native/native-artifact-language.d.ts +3 -1
  116. package/dist/domains/comet-native/native-artifact-language.d.ts.map +1 -1
  117. package/dist/domains/comet-native/native-artifact-language.js +4 -11
  118. package/dist/domains/comet-native/native-artifact-language.js.map +1 -1
  119. package/dist/domains/comet-native/native-artifacts.d.ts +8 -1
  120. package/dist/domains/comet-native/native-artifacts.d.ts.map +1 -1
  121. package/dist/domains/comet-native/native-artifacts.js +14 -2
  122. package/dist/domains/comet-native/native-artifacts.js.map +1 -1
  123. package/dist/domains/comet-native/native-builder-acceptance-review.d.ts +16 -0
  124. package/dist/domains/comet-native/native-builder-acceptance-review.d.ts.map +1 -0
  125. package/dist/domains/comet-native/native-builder-acceptance-review.js +65 -0
  126. package/dist/domains/comet-native/native-builder-acceptance-review.js.map +1 -0
  127. package/dist/domains/comet-native/native-cli-help.d.ts.map +1 -1
  128. package/dist/domains/comet-native/native-cli-help.js +8 -4
  129. package/dist/domains/comet-native/native-cli-help.js.map +1 -1
  130. package/dist/domains/comet-native/native-cli-shared.d.ts +4 -2
  131. package/dist/domains/comet-native/native-cli-shared.d.ts.map +1 -1
  132. package/dist/domains/comet-native/native-cli-shared.js +37 -9
  133. package/dist/domains/comet-native/native-cli-shared.js.map +1 -1
  134. package/dist/domains/comet-native/native-cli.d.ts +4 -1
  135. package/dist/domains/comet-native/native-cli.d.ts.map +1 -1
  136. package/dist/domains/comet-native/native-cli.js +9 -5
  137. package/dist/domains/comet-native/native-cli.js.map +1 -1
  138. package/dist/domains/comet-native/native-doctor-command.d.ts.map +1 -1
  139. package/dist/domains/comet-native/native-doctor-command.js +144 -3
  140. package/dist/domains/comet-native/native-doctor-command.js.map +1 -1
  141. package/dist/domains/comet-native/native-git-provenance.d.ts.map +1 -1
  142. package/dist/domains/comet-native/native-git-provenance.js +3 -2
  143. package/dist/domains/comet-native/native-git-provenance.js.map +1 -1
  144. package/dist/domains/comet-native/native-hook-guard.d.ts.map +1 -1
  145. package/dist/domains/comet-native/native-hook-guard.js +56 -26
  146. package/dist/domains/comet-native/native-hook-guard.js.map +1 -1
  147. package/dist/domains/comet-native/native-local-execution.d.ts +1 -0
  148. package/dist/domains/comet-native/native-local-execution.d.ts.map +1 -1
  149. package/dist/domains/comet-native/native-local-execution.js +7 -0
  150. package/dist/domains/comet-native/native-local-execution.js.map +1 -1
  151. package/dist/domains/comet-native/native-lock.d.ts.map +1 -1
  152. package/dist/domains/comet-native/native-lock.js +12 -0
  153. package/dist/domains/comet-native/native-lock.js.map +1 -1
  154. package/dist/domains/comet-native/native-loop-runtime.d.ts +2 -0
  155. package/dist/domains/comet-native/native-loop-runtime.d.ts.map +1 -1
  156. package/dist/domains/comet-native/native-loop-runtime.js +6 -0
  157. package/dist/domains/comet-native/native-loop-runtime.js.map +1 -1
  158. package/dist/domains/comet-native/native-next-command.d.ts.map +1 -1
  159. package/dist/domains/comet-native/native-next-command.js +57 -28
  160. package/dist/domains/comet-native/native-next-command.js.map +1 -1
  161. package/dist/domains/comet-native/native-output-language.d.ts.map +1 -1
  162. package/dist/domains/comet-native/native-output-language.js +7 -0
  163. package/dist/domains/comet-native/native-output-language.js.map +1 -1
  164. package/dist/domains/comet-native/native-portable-acceptance.d.ts +2 -0
  165. package/dist/domains/comet-native/native-portable-acceptance.d.ts.map +1 -1
  166. package/dist/domains/comet-native/native-portable-acceptance.js +11 -1
  167. package/dist/domains/comet-native/native-portable-acceptance.js.map +1 -1
  168. package/dist/domains/comet-native/native-portable-archive.d.ts +1 -0
  169. package/dist/domains/comet-native/native-portable-archive.d.ts.map +1 -1
  170. package/dist/domains/comet-native/native-portable-archive.js +3 -0
  171. package/dist/domains/comet-native/native-portable-archive.js.map +1 -1
  172. package/dist/domains/comet-native/native-portable-checks.d.ts +15 -4
  173. package/dist/domains/comet-native/native-portable-checks.d.ts.map +1 -1
  174. package/dist/domains/comet-native/native-portable-checks.js +167 -66
  175. package/dist/domains/comet-native/native-portable-checks.js.map +1 -1
  176. package/dist/domains/comet-native/native-portable-continuation.d.ts +2 -1
  177. package/dist/domains/comet-native/native-portable-continuation.d.ts.map +1 -1
  178. package/dist/domains/comet-native/native-portable-continuation.js +121 -39
  179. package/dist/domains/comet-native/native-portable-continuation.js.map +1 -1
  180. package/dist/domains/comet-native/native-portable-requirements.d.ts +1 -0
  181. package/dist/domains/comet-native/native-portable-requirements.d.ts.map +1 -1
  182. package/dist/domains/comet-native/native-portable-requirements.js +61 -18
  183. package/dist/domains/comet-native/native-portable-requirements.js.map +1 -1
  184. package/dist/domains/comet-native/native-portable-shape-coordination.d.ts.map +1 -1
  185. package/dist/domains/comet-native/native-portable-shape-coordination.js +18 -5
  186. package/dist/domains/comet-native/native-portable-shape-coordination.js.map +1 -1
  187. package/dist/domains/comet-native/native-portable-state.d.ts.map +1 -1
  188. package/dist/domains/comet-native/native-portable-state.js +10 -3
  189. package/dist/domains/comet-native/native-portable-state.js.map +1 -1
  190. package/dist/domains/comet-native/native-portable-status.d.ts.map +1 -1
  191. package/dist/domains/comet-native/native-portable-status.js +10 -4
  192. package/dist/domains/comet-native/native-portable-status.js.map +1 -1
  193. package/dist/domains/comet-native/native-portable-types.d.ts +2 -1
  194. package/dist/domains/comet-native/native-portable-types.d.ts.map +1 -1
  195. package/dist/domains/comet-native/native-portable-types.js.map +1 -1
  196. package/dist/domains/comet-native/native-portable-verification.d.ts +3 -0
  197. package/dist/domains/comet-native/native-portable-verification.d.ts.map +1 -1
  198. package/dist/domains/comet-native/native-portable-verification.js +28 -35
  199. package/dist/domains/comet-native/native-portable-verification.js.map +1 -1
  200. package/dist/domains/comet-native/native-portable-verifier-coordination.d.ts +2 -0
  201. package/dist/domains/comet-native/native-portable-verifier-coordination.d.ts.map +1 -1
  202. package/dist/domains/comet-native/native-portable-verifier-coordination.js +9 -0
  203. package/dist/domains/comet-native/native-portable-verifier-coordination.js.map +1 -1
  204. package/dist/domains/comet-native/native-runner-input-artifacts.d.ts +9 -0
  205. package/dist/domains/comet-native/native-runner-input-artifacts.d.ts.map +1 -0
  206. package/dist/domains/comet-native/native-runner-input-artifacts.js +102 -0
  207. package/dist/domains/comet-native/native-runner-input-artifacts.js.map +1 -0
  208. package/dist/domains/comet-native/native-runner-input.d.ts +90 -64
  209. package/dist/domains/comet-native/native-runner-input.d.ts.map +1 -1
  210. package/dist/domains/comet-native/native-runner-input.js +107 -22
  211. package/dist/domains/comet-native/native-runner-input.js.map +1 -1
  212. package/dist/domains/comet-native/native-script-entry.d.ts.map +1 -1
  213. package/dist/domains/comet-native/native-script-entry.js +9 -3
  214. package/dist/domains/comet-native/native-script-entry.js.map +1 -1
  215. package/dist/domains/comet-native/native-snapshot.d.ts.map +1 -1
  216. package/dist/domains/comet-native/native-snapshot.js +7 -2
  217. package/dist/domains/comet-native/native-snapshot.js.map +1 -1
  218. package/dist/domains/comet-native/native-verification-receipt-runtime.d.ts.map +1 -1
  219. package/dist/domains/comet-native/native-verification-receipt-runtime.js +5 -4
  220. package/dist/domains/comet-native/native-verification-receipt-runtime.js.map +1 -1
  221. package/dist/domains/comet-native/native-workspace-finish.d.ts +26 -0
  222. package/dist/domains/comet-native/native-workspace-finish.d.ts.map +1 -1
  223. package/dist/domains/comet-native/native-workspace-finish.js +520 -15
  224. package/dist/domains/comet-native/native-workspace-finish.js.map +1 -1
  225. package/dist/domains/dashboard/web/assets/{abnfDiagram-VCTEODGH-qlcciI25.js → abnfDiagram-VCTEODGH-CFEU4PQU.js} +1 -1
  226. package/dist/domains/dashboard/web/assets/{arc-KVm4k9kp.js → arc-BeYYJor9.js} +1 -1
  227. package/dist/domains/dashboard/web/assets/architecture-7GRP2DOG-YphM8Wtg.js +1 -0
  228. package/dist/domains/dashboard/web/assets/{architectureDiagram-5GKGNRK7-C366RdBK.js → architectureDiagram-5GKGNRK7-CqTy4fyH.js} +1 -1
  229. package/dist/domains/dashboard/web/assets/{blockDiagram-I7D4REHJ-DAuTc2OO.js → blockDiagram-I7D4REHJ-SR_5S4SA.js} +1 -1
  230. package/dist/domains/dashboard/web/assets/{c4Diagram-7LVT6UL2-CFB0RZmL.js → c4Diagram-7LVT6UL2-DzphG2XS.js} +1 -1
  231. package/dist/domains/dashboard/web/assets/channel-BkwcGbYR.js +1 -0
  232. package/dist/domains/dashboard/web/assets/{chunk-4HAMMTFA-DTV6lyGQ.js → chunk-4HAMMTFA-BpthUtIs.js} +1 -1
  233. package/dist/domains/dashboard/web/assets/{chunk-75Z2AOVW-DgJm4cEX.js → chunk-75Z2AOVW-CxAdhPhG.js} +1 -1
  234. package/dist/domains/dashboard/web/assets/{chunk-DU6HZSFF-NAtcgGZr.js → chunk-DU6HZSFF-5peVNfxp.js} +1 -1
  235. package/dist/domains/dashboard/web/assets/{chunk-F27PBJKO-BcFXHZMJ.js → chunk-F27PBJKO-CwG4xi2V.js} +1 -1
  236. package/dist/domains/dashboard/web/assets/{chunk-GMAD6QVW-wNemar9b.js → chunk-GMAD6QVW-BuFFY783.js} +1 -1
  237. package/dist/domains/dashboard/web/assets/{chunk-GVQU2GXP-BiVCDO9d.js → chunk-GVQU2GXP-OyUFmjE-.js} +1 -1
  238. package/dist/domains/dashboard/web/assets/{chunk-IMKFNOWR-D9GDj4b6.js → chunk-IMKFNOWR-DmTme45C.js} +1 -1
  239. package/dist/domains/dashboard/web/assets/{chunk-L3NEJ4N5-DkPl5vac.js → chunk-L3NEJ4N5-D6BHbKty.js} +1 -1
  240. package/dist/domains/dashboard/web/assets/{chunk-OSK3NFVY-RbbxGqtJ.js → chunk-OSK3NFVY-CCz1v1Nv.js} +1 -1
  241. package/dist/domains/dashboard/web/assets/{chunk-P2QGCYS3-B9_X57eT.js → chunk-P2QGCYS3-C1_7kDYS.js} +1 -1
  242. package/dist/domains/dashboard/web/assets/{chunk-POPQ4Y6H-DE4bOH0D.js → chunk-POPQ4Y6H-buFKlAdG.js} +1 -1
  243. package/dist/domains/dashboard/web/assets/{chunk-PWAF6VOD-Bo-W1iA1.js → chunk-PWAF6VOD-B_K6Jrsv.js} +1 -1
  244. package/dist/domains/dashboard/web/assets/{chunk-SHT3W25Y-BhLxs4rZ.js → chunk-SHT3W25Y-CRRLsXmS.js} +1 -1
  245. package/dist/domains/dashboard/web/assets/{chunk-SVP7TREG-BCXfn8uF.js → chunk-SVP7TREG-Q5zCQTul.js} +1 -1
  246. package/dist/domains/dashboard/web/assets/{chunk-TICWLB2K-DqSBgE5L.js → chunk-TICWLB2K-Zc5XaO5Z.js} +1 -1
  247. package/dist/domains/dashboard/web/assets/{chunk-XXDRQBXY-Dbkt1xEZ.js → chunk-XXDRQBXY-BAIB2h_r.js} +1 -1
  248. package/dist/domains/dashboard/web/assets/classDiagram-ZZMXUADV-DzpQhyL3.js +1 -0
  249. package/dist/domains/dashboard/web/assets/classDiagram-v2-VYDZK3BY-DzpQhyL3.js +1 -0
  250. package/dist/domains/dashboard/web/assets/{cose-bilkent-JH36ORCC-DrwNpMyX.js → cose-bilkent-JH36ORCC-C7jkFYhm.js} +1 -1
  251. package/dist/domains/dashboard/web/assets/{cynefin-OW5HDTMX-BbDlivxs.js → cynefin-OW5HDTMX-4yS-l7kx.js} +1 -1
  252. package/dist/domains/dashboard/web/assets/{cynefinDiagram-5FMLGOSQ-DmLmRN42.js → cynefinDiagram-5FMLGOSQ-DAkfQ8jW.js} +1 -1
  253. package/dist/domains/dashboard/web/assets/{dagre-GXQ25YYZ-Cgm5DNgI.js → dagre-GXQ25YYZ-D3b16ruz.js} +1 -1
  254. package/dist/domains/dashboard/web/assets/{diagram-S7CK7UJ4-Bnht1X2m.js → diagram-S7CK7UJ4-BXZHtdNz.js} +1 -1
  255. package/dist/domains/dashboard/web/assets/{diagram-UQ7AKVKN-LELTI4PR.js → diagram-UQ7AKVKN-B4nIzofw.js} +1 -1
  256. package/dist/domains/dashboard/web/assets/{diagram-VSXAHHWV-Do0fZuNG.js → diagram-VSXAHHWV-B6d7KNj1.js} +1 -1
  257. package/dist/domains/dashboard/web/assets/{diagram-VX7I27RA-B0rZTUwt.js → diagram-VX7I27RA-CHr0Rg8a.js} +1 -1
  258. package/dist/domains/dashboard/web/assets/{diagram-Z3DM3KII-wn8xCc_j.js → diagram-Z3DM3KII-BO2IxR9N.js} +1 -1
  259. package/dist/domains/dashboard/web/assets/{dist-CYwDKy2m.js → dist-0wyW4qbm.js} +1 -1
  260. package/dist/domains/dashboard/web/assets/{ebnfDiagram-PWID7BFC-CMDOpP2T.js → ebnfDiagram-PWID7BFC-CHRK8Hyg.js} +1 -1
  261. package/dist/domains/dashboard/web/assets/{erDiagram-RLTQ6QDP-C2d43bdq.js → erDiagram-RLTQ6QDP-_HFHzZlc.js} +1 -1
  262. package/dist/domains/dashboard/web/assets/{es-BzaOcXwu.js → es-kjndkyYX.js} +1 -1
  263. package/dist/domains/dashboard/web/assets/eventmodeling-NTZA5JFV-CG-cVTEg.js +1 -0
  264. package/dist/domains/dashboard/web/assets/flowDiagram-HODETNUW-C3RaJz-z.js +1 -0
  265. package/dist/domains/dashboard/web/assets/{ganttDiagram-EL5Y4UJY-hTH7TzhY.js → ganttDiagram-EL5Y4UJY-CxhSw1PI.js} +1 -1
  266. package/dist/domains/dashboard/web/assets/{gitGraph-4MIJSDKK-C3CCywb-.js → gitGraph-4MIJSDKK-Cbw-3Dnk.js} +1 -1
  267. package/dist/domains/dashboard/web/assets/{gitGraphDiagram-WWUBYQGX-DW6jDckj.js → gitGraphDiagram-WWUBYQGX-Cx93hwtL.js} +1 -1
  268. package/dist/domains/dashboard/web/assets/{index-CTZPz_mx.js → index-CQb8j5OE.js} +3 -3
  269. package/dist/domains/dashboard/web/assets/{index-CWNqBYyI.css → index-Gaxnq6Jd.css} +1 -1
  270. package/dist/domains/dashboard/web/assets/{info-A6RAGUB7-Bg3eKis8.js → info-A6RAGUB7--d4EMiIK.js} +1 -1
  271. package/dist/domains/dashboard/web/assets/{infoDiagram-27XIBGKW-7zk6ytVJ.js → infoDiagram-27XIBGKW-DMkRRNGB.js} +1 -1
  272. package/dist/domains/dashboard/web/assets/{ishikawaDiagram-5VMMS53U-CGDCTmCQ.js → ishikawaDiagram-5VMMS53U-CjKh_jHT.js} +1 -1
  273. package/dist/domains/dashboard/web/assets/{journeyDiagram-3NMN7TZE-C8_L_e8a.js → journeyDiagram-3NMN7TZE-CG735M_M.js} +1 -1
  274. package/dist/domains/dashboard/web/assets/{kanban-definition-UXKFOSKX-D5x6qTgM.js → kanban-definition-UXKFOSKX-CgXFmhFM.js} +1 -1
  275. package/dist/domains/dashboard/web/assets/{line-DC93oja4.js → line-Bhf8l3gG.js} +1 -1
  276. package/dist/domains/dashboard/web/assets/{linear-81TNubkB.js → linear-YWT9mnmu.js} +1 -1
  277. package/dist/domains/dashboard/web/assets/{mermaid-parser.core-D60d8uy9.js → mermaid-parser.core-COzQ1kq0.js} +3 -3
  278. package/dist/domains/dashboard/web/assets/{mermaid.core-l4Wgjz2Q.js → mermaid.core-Cc1lMpx-.js} +4 -4
  279. package/dist/domains/dashboard/web/assets/{mindmap-definition-YA3MSWOX-Dfqu1201.js → mindmap-definition-YA3MSWOX-B_VOgUSO.js} +1 -1
  280. package/dist/domains/dashboard/web/assets/{packet-AYTQ26CC-CsFZBhTs.js → packet-AYTQ26CC-CXbJLNC9.js} +1 -1
  281. package/dist/domains/dashboard/web/assets/{pegDiagram-XKGWAZYB-myFMaA7n.js → pegDiagram-XKGWAZYB-BuS0hioo.js} +1 -1
  282. package/dist/domains/dashboard/web/assets/{pie-WAS4IAKB-FMuCTNSk.js → pie-WAS4IAKB-BEaGwdWZ.js} +1 -1
  283. package/dist/domains/dashboard/web/assets/{pieDiagram-E7YTZNPT-CSR4B4Z1.js → pieDiagram-E7YTZNPT-DDzYv-_9.js} +1 -1
  284. package/dist/domains/dashboard/web/assets/purify.es-Bvo9QlJ8.js +3 -0
  285. package/dist/domains/dashboard/web/assets/{quadrantDiagram-AXDQQJYC-BiE4kEeP.js → quadrantDiagram-AXDQQJYC-DExT--Ut.js} +1 -1
  286. package/dist/domains/dashboard/web/assets/{radar-RG4KPBEZ-BZ-n7JtG.js → radar-RG4KPBEZ-C1JtzGr8.js} +1 -1
  287. package/dist/domains/dashboard/web/assets/{railroad-74A4TZTK-CWVtRfHu.js → railroad-74A4TZTK-CMpkXATR.js} +1 -1
  288. package/dist/domains/dashboard/web/assets/railroad-abnf-HS5TGJTU-BM35gYWg.js +1 -0
  289. package/dist/domains/dashboard/web/assets/railroad-ebnf-LZEXJU2U-CN60GGp0.js +1 -0
  290. package/dist/domains/dashboard/web/assets/railroad-peg-WCYAUIDC-DRUSW4g7.js +1 -0
  291. package/dist/domains/dashboard/web/assets/{railroadDiagram-O6MQD6OU-CBj7v1de.js → railroadDiagram-O6MQD6OU-BCE65sLR.js} +1 -1
  292. package/dist/domains/dashboard/web/assets/{requirementDiagram-BXWQKSXE-DSDgb54n.js → requirementDiagram-BXWQKSXE-CmyX3vGG.js} +1 -1
  293. package/dist/domains/dashboard/web/assets/{sankeyDiagram-P5KCCOFB-1-oXLUAP.js → sankeyDiagram-P5KCCOFB-BFKrX2dv.js} +1 -1
  294. package/dist/domains/dashboard/web/assets/{sequenceDiagram-WJ2MYXX4-DAF4nGs_.js → sequenceDiagram-WJ2MYXX4-DqodS9W-.js} +1 -1
  295. package/dist/domains/dashboard/web/assets/{src-CTWLVUqE.js → src-BVUP3QzK.js} +1 -1
  296. package/dist/domains/dashboard/web/assets/{stateDiagram-D77RDMKH-DzENBaJT.js → stateDiagram-D77RDMKH-nHai_EZh.js} +1 -1
  297. package/dist/domains/dashboard/web/assets/stateDiagram-v2-MP3YSRHH-BL6vE1m9.js +1 -0
  298. package/dist/domains/dashboard/web/assets/{swimlanes-42K2YHIH-CnxFUZJc.js → swimlanes-42K2YHIH-BODBARYN.js} +1 -1
  299. package/dist/domains/dashboard/web/assets/swimlanesDiagram-VR7AAH4N-XHK_nuTj.js +8 -0
  300. package/dist/domains/dashboard/web/assets/{timeline-definition-24CTP7MA-KtvPzpZM.js → timeline-definition-24CTP7MA-RVnG8rgK.js} +1 -1
  301. package/dist/domains/dashboard/web/assets/{treeView-Q6P3EWNA-Bdu3P9iY.js → treeView-Q6P3EWNA-B6lrg3KQ.js} +1 -1
  302. package/dist/domains/dashboard/web/assets/{treemap-WGGIJYW6-DFnn23JR.js → treemap-WGGIJYW6-CJfMKVf4.js} +1 -1
  303. package/dist/domains/dashboard/web/assets/{vennDiagram-4TSXK5OY-BjpviP26.js → vennDiagram-4TSXK5OY-CZMWJiN7.js} +1 -1
  304. package/dist/domains/dashboard/web/assets/{wardley-WFR3VGLG-CEo4a-j4.js → wardley-WFR3VGLG-BoUsbNlx.js} +1 -1
  305. package/dist/domains/dashboard/web/assets/{wardleyDiagram-VM6X3IG4-Bf54n6im.js → wardleyDiagram-VM6X3IG4-BwBCykVb.js} +1 -1
  306. package/dist/domains/dashboard/web/assets/{xychartDiagram-S5SC5T6Z-CU-hI0Cm.js → xychartDiagram-S5SC5T6Z-vutnbx57.js} +1 -1
  307. package/dist/domains/dashboard/web/index.html +2 -2
  308. package/dist/domains/project-knowledge/corpus.d.ts +5 -0
  309. package/dist/domains/project-knowledge/corpus.d.ts.map +1 -1
  310. package/dist/domains/project-knowledge/corpus.js +34 -9
  311. package/dist/domains/project-knowledge/corpus.js.map +1 -1
  312. package/dist/domains/project-knowledge/index-store.d.ts +5 -2
  313. package/dist/domains/project-knowledge/index-store.d.ts.map +1 -1
  314. package/dist/domains/project-knowledge/index-store.js +95 -52
  315. package/dist/domains/project-knowledge/index-store.js.map +1 -1
  316. package/dist/domains/project-knowledge/local-provider.d.ts.map +1 -1
  317. package/dist/domains/project-knowledge/local-provider.js +14 -3
  318. package/dist/domains/project-knowledge/local-provider.js.map +1 -1
  319. package/dist/domains/project-knowledge/local-store.d.ts +2 -2
  320. package/dist/domains/project-knowledge/local-store.d.ts.map +1 -1
  321. package/dist/domains/project-knowledge/local-store.js +2 -2
  322. package/dist/domains/project-knowledge/local-store.js.map +1 -1
  323. package/dist/domains/project-knowledge/plugin.d.ts.map +1 -1
  324. package/dist/domains/project-knowledge/plugin.js +5 -4
  325. package/dist/domains/project-knowledge/plugin.js.map +1 -1
  326. package/dist/domains/project-knowledge/provider-factory.d.ts.map +1 -1
  327. package/dist/domains/project-knowledge/provider-factory.js +7 -5
  328. package/dist/domains/project-knowledge/provider-factory.js.map +1 -1
  329. package/dist/domains/project-knowledge/types.d.ts +1 -0
  330. package/dist/domains/project-knowledge/types.d.ts.map +1 -1
  331. package/dist/domains/skill/platform-install.d.ts.map +1 -1
  332. package/dist/domains/skill/platform-install.js +31 -3
  333. package/dist/domains/skill/platform-install.js.map +1 -1
  334. package/dist/domains/workflow-contract/hook-target-scope.d.ts.map +1 -1
  335. package/dist/domains/workflow-contract/hook-target-scope.js +2 -1
  336. package/dist/domains/workflow-contract/hook-target-scope.js.map +1 -1
  337. package/dist/platform/paths/git-worktree-cache.d.ts +3 -0
  338. package/dist/platform/paths/git-worktree-cache.d.ts.map +1 -0
  339. package/dist/platform/paths/git-worktree-cache.js +109 -0
  340. package/dist/platform/paths/git-worktree-cache.js.map +1 -0
  341. package/dist/platform/paths/git-worktree.d.ts +6 -0
  342. package/dist/platform/paths/git-worktree.d.ts.map +1 -1
  343. package/dist/platform/paths/git-worktree.js +191 -30
  344. package/dist/platform/paths/git-worktree.js.map +1 -1
  345. package/dist/platform/process/comet-daemon.d.ts +3 -0
  346. package/dist/platform/process/comet-daemon.d.ts.map +1 -1
  347. package/dist/platform/process/comet-daemon.js +9 -6
  348. package/dist/platform/process/comet-daemon.js.map +1 -1
  349. package/dist/platform/process/external-command.d.ts +3 -0
  350. package/dist/platform/process/external-command.d.ts.map +1 -1
  351. package/dist/platform/process/external-command.js +69 -2
  352. package/dist/platform/process/external-command.js.map +1 -1
  353. package/dist/platform/process/git.d.ts.map +1 -1
  354. package/dist/platform/process/git.js +3 -4
  355. package/dist/platform/process/git.js.map +1 -1
  356. package/dist/platform/process/hook-adapter.d.ts +2 -0
  357. package/dist/platform/process/hook-adapter.d.ts.map +1 -1
  358. package/dist/platform/process/hook-adapter.js +32 -3
  359. package/dist/platform/process/hook-adapter.js.map +1 -1
  360. package/dist/platform/process/hook-read-cache.d.ts +2 -0
  361. package/dist/platform/process/hook-read-cache.d.ts.map +1 -1
  362. package/dist/platform/process/hook-read-cache.js +25 -13
  363. package/dist/platform/process/hook-read-cache.js.map +1 -1
  364. package/dist/platform/process/process-identity.d.ts.map +1 -1
  365. package/dist/platform/process/process-identity.js +18 -51
  366. package/dist/platform/process/process-identity.js.map +1 -1
  367. package/dist/platform/process/runtime-metrics.d.ts +10 -1
  368. package/dist/platform/process/runtime-metrics.d.ts.map +1 -1
  369. package/dist/platform/process/runtime-metrics.js +37 -2
  370. package/dist/platform/process/runtime-metrics.js.map +1 -1
  371. package/dist/platform/process/windows-process-broker.d.ts +13 -0
  372. package/dist/platform/process/windows-process-broker.d.ts.map +1 -0
  373. package/dist/platform/process/windows-process-broker.js +150 -0
  374. package/dist/platform/process/windows-process-broker.js.map +1 -0
  375. package/dist/platform/process/windows-process-identity.d.ts +6 -0
  376. package/dist/platform/process/windows-process-identity.d.ts.map +1 -0
  377. package/dist/platform/process/windows-process-identity.js +190 -0
  378. package/dist/platform/process/windows-process-identity.js.map +1 -0
  379. package/package.json +1 -1
  380. package/dist/domains/dashboard/web/assets/architecture-7GRP2DOG-QeljAKq2.js +0 -1
  381. package/dist/domains/dashboard/web/assets/channel-_AqwP0KA.js +0 -1
  382. package/dist/domains/dashboard/web/assets/classDiagram-ZZMXUADV-CRo4aB_7.js +0 -1
  383. package/dist/domains/dashboard/web/assets/classDiagram-v2-VYDZK3BY-CRo4aB_7.js +0 -1
  384. package/dist/domains/dashboard/web/assets/eventmodeling-NTZA5JFV-KYANo3DQ.js +0 -1
  385. package/dist/domains/dashboard/web/assets/flowDiagram-HODETNUW-CSROlLuj.js +0 -1
  386. package/dist/domains/dashboard/web/assets/purify.es-ChwZkWde.js +0 -3
  387. package/dist/domains/dashboard/web/assets/railroad-abnf-HS5TGJTU-CkHwwI2i.js +0 -1
  388. package/dist/domains/dashboard/web/assets/railroad-ebnf-LZEXJU2U-8xzEHL6v.js +0 -1
  389. package/dist/domains/dashboard/web/assets/railroad-peg-WCYAUIDC-nSHBw0tH.js +0 -1
  390. package/dist/domains/dashboard/web/assets/stateDiagram-v2-MP3YSRHH-CNw52Fkn.js +0 -1
  391. package/dist/domains/dashboard/web/assets/swimlanesDiagram-VR7AAH4N-6TRvbqqI.js +0 -8
@@ -28,8 +28,6 @@ comet state select <change-name>
28
28
  comet state check <name> archive --json
29
29
  ```
30
30
 
31
- Combine multiple read-only comet commands (for example `state get`, `state next`, `state artifacts`) into a single shell invocation to reduce process startup overhead.
32
-
33
31
  When the previous phase's guard already returned this phase's state, continue from that state and `agent.continuation` without repeating select/check; run the entry checks above only when resuming, after workspace changes, or after external state changes. Continue from returned layout, configuration, nextAction, and the delivery summary. After context loss, read details according to context-recovery.md. If authorization is still valid and the delivery target is unchanged, continue only unfinished actions without asking again. Handle the specific cause on failure.
34
32
 
35
33
  If select/check returns `BLOCKED` — or a branch-binding `ERROR` — because `bound_branch` differs from the current branch, pause under `comet-classic/reference/decision-point.md`. Offer a single choice: return to the bound branch and rerun entry checks, or, after the user explicitly confirms that the current branch should take over this change, run `comet state rebind <change-name>` and rerun entry checks. Do not switch or rebind branches yourself.
@@ -129,7 +127,7 @@ Archive moves files and merges specs; it does not commit. Afterwards, expect the
129
127
  - Main spec contains the merged delta changes.
130
128
  - The Design Doc/plan contains archive metadata.
131
129
 
132
- Confirm that delivery still records valid authorization, then write the compatibility field and run the final archive guard (both commands in one shell invocation to reduce process startup overhead):
130
+ Confirm that delivery still records valid authorization, then write the compatibility field and run the final archive guard (run the second command only after the first succeeds):
133
131
 
134
132
  ```bash
135
133
  comet state set <change-name> branch_status handled && comet guard <change-name> archive
@@ -141,10 +139,12 @@ Read `git status --short` after archive and reconcile it against the ownership r
141
139
 
142
140
  Use explicit pathspecs for the inspected paths, then inspect the staged diff. Do not stage the whole repository or include pre-existing user edits:
143
141
 
142
+ Prepare the message under the project's active format, language, and body requirements without asking again when conventions are clear. Replace the placeholder below with the actual message; do not treat an English example as project policy. Project Git hooks run normally. If a hook rejects the commit, retain staged contents, correct the message, and continue the same commit. If the call was interrupted or its result is uncertain, inspect actual Git history before retrying; do not retry blindly or bypass hooks.
143
+
144
144
  ```bash
145
145
  git add -- <individually-verified-archive-paths...>
146
146
  git diff --cached --stat
147
- git commit -m "chore: archive <change-name>"
147
+ git commit -m "<archive message following project conventions>"
148
148
  ```
149
149
 
150
150
  Stop if commit fails or the staged diff contains unrelated paths. Do not continue branch handling.
@@ -123,11 +123,11 @@ When implementation reveals gaps in the initial spec, handle them according to t
123
123
 
124
124
  For implementation details within the confirmed scope that do not change public behavior or acceptance requirements, update the plan and rationale without reopening Open/Design. The following categories apply only to actual specification or scope changes:
125
125
 
126
- | Size | Trigger | Action |
127
- | ------ | ------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
128
- | Small | Missing acceptance scenarios or edge cases | Edit delta spec + design.md directly and add tasks.md work. |
126
+ | Size | Trigger | Action |
127
+ | ------ | ------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
128
+ | Small | Missing acceptance scenarios or edge cases | Edit delta spec + design.md directly and add tasks.md work. |
129
129
  | Medium | Interface changes, new components, or data-flow changes | **Pause under `comet-classic/reference/decision-point.md` and wait for explicit confirmation.** Offer: continue with the design update, or adjust the requested change. Do not reopen Design without this confirmation. After approval, load Superpowers `brainstorming` through the Skill tool to update the Design Doc + delta spec. |
130
- | Large | An entirely new feature requirement | **Pause, present split choices, and wait for explicit confirmation**. Then create an independent change through `/comet-open`. |
130
+ | Large | An entirely new feature requirement | **Pause, present split choices, and wait for explicit confirmation**. Then create an independent change through `/comet-open`. |
131
131
 
132
132
  **Recheck scope:** Before adding tasks, compare the original goal, public behavior, acceptance requirements, and risks the user has accepted. Update tasks and rationale directly when adding omitted work within that scope or adjusting task granularity. Task counts or growth percentages alone do not require a pause. Only actual scope expansion, redesign, or a new independently shippable feature requires a decision under `comet-classic/reference/decision-point.md`: continue, adjust, or split.
133
133
 
@@ -190,8 +190,6 @@ Build command detection covers three sources: a build script in the invocation d
190
190
 
191
191
  **Recording evidence:** finish the implementation and make its checks pass, then record the build evidence and run `comet guard <change-name> build --apply`. Build records build evidence only: external (non-`--local`) verify checks must be recorded **after** guard build --apply succeeds — the phase transition expires externally recorded verify evidence (`crossed a phase boundary`), so expensive non-idempotent external checks would run for nothing; `--local` evidence is exempt and can be recorded in one shell call. Runtime reuses recorded evidence automatically when inputs did not change, so a rerun after a documentation edit does not execute the command again. When guard reports invalid evidence, it names the exact reason and changed files — rerun only the listed scope and follow the message for record-check or phase-boundary situations instead of redoing everything. Commit after guard passes; tick tasks with `comet state task-complete`.
192
192
 
193
- Combine multiple read-only comet commands (such as `state get`, `state next`, `status`) into one shell call executed in sequence to reduce per-command process startup overhead.
194
-
195
193
  Before exiting, advance phase with the guard, independently of `auto_transition`:
196
194
 
197
195
  ```bash
@@ -28,12 +28,14 @@ A new full change must have its workspace selected before Open creates artifacts
28
28
 
29
29
  ## 2. Load the Current Phase
30
30
 
31
- While working on the same set of tasks, reuse state returned by the current commands and context already read. Refresh only as needed after a state write, workspace switch, session recovery, or external change; reuse valid state included in the successful result. Select checks according to the risk of the current changes; Verify still owns one final integrated review of the completed implementation.
31
+ While working on the same set of tasks, reuse state returned by the current commands and context already read. When a state update returns the latest state and `agent.continuation`, continue directly. Query again only when those fields are missing, the workspace changes, session recovery lacks context, or external state changes. Select checks according to the risk of the current changes; Verify still owns one final integrated review of the completed implementation.
32
32
 
33
33
  After binding the workspace and obtaining phase, read and follow [Task Context and Artifact Language](reference/scripts.md#task-context-and-artifact-language), and run `comet task`. Use the configured artifact language. Start from Context Manifest summaries and expand only information still missing for the current step. Report outcomes truthfully after actually using an item, and save a checkpoint at task end.
34
34
 
35
35
  Memory learning submits only reusable user information; never save task summaries, progress, command output, or test results as Personal Memory. Complete the learning check in the task context reference before task completion and record `submitted`, `no-observation`, or `not-run`.
36
36
 
37
+ Project experience and personal preferences are stored separately: before task completion, write project-validated and reusable experience to Project Memory with `comet knowledge remember`; only user preferences and stable collaboration habits go to Personal Memory. See [Task Context and Artifact Language](reference/scripts.md#task-context-and-artifact-language) for the commands and filtering rules.
38
+
37
39
  | Runtime route or phase | Entry and responsibility |
38
40
  | ----------------------------- | --------------------------------------------------------------------------------------------------------------- |
39
41
  | new full / open | `/comet-open`: clarify scope, create required artifacts, and obtain the user's review and confirmation |
@@ -40,6 +40,8 @@ Run workspace prepare during Open. If the workspace is unknown, changes, or its
40
40
 
41
41
  Entry check --json returns layout, configuration, nextAction, task summaries, coordination, and delivery. Do not separately query fields already returned. If session recovery lacks context, start with --recover for a summary; add --details only for complete tasks, checkpoints, or evidence. See context-recovery.md.
42
42
 
43
+ Query only information still missing for the current action. Putting multiple `comet` commands in one shell call reduces tool round trips; each CLI still starts separately. Reduce redundant calls by reusing fields already returned. For dependent commands, run the next only after the previous succeeds and check every exit code; a successful last command must not hide an earlier failure.
44
+
43
45
  checkpoint input requires schemaVersion:1 and taskIds/revision/stage/sessionId/evidence/unresolved/reviewRounds; reads return `{checkpoint, stale}`. Runtime validates the data and generates Markdown. See context-recovery.md for JSON. task-complete automatically synchronizes legacy-plan tasks with established comet-task ID mappings. Use sync-plan for a separate update; planSync mapping-required requires completing mappings, not reimplementing tasks.
44
46
 
45
47
  delivery input includes action (local|push|pr), targetBranch, and optional remote, commit, and prUrl; see comet-archive for examples. Normal entry and delivery reads do not access the network. Only `state delivery <change-name> --verify` checks remote and PR state read-only, returning `{delivery, verification}`. A successful record write does not prove delivery success. Stop for unavailable commands, rejected operations, or inconsistent records; do not bypass checks by editing internal state.
@@ -93,7 +95,13 @@ comet check run <change-name> verify --local --incremental -- <program> [args...
93
95
  {
94
96
  "version": 2,
95
97
  "commands": [
96
- { "argv": ["pnpm", "build"], "cwd": ".", "files": ["src/**", "package.json", "tsconfig.json"], "outputs": ["dist/**"], "git": "all" },
98
+ {
99
+ "argv": ["pnpm", "build"],
100
+ "cwd": ".",
101
+ "files": ["src/**", "package.json", "tsconfig.json"],
102
+ "outputs": ["dist/**"],
103
+ "git": "all"
104
+ },
97
105
  { "argv": ["vitest", "run"], "cwd": ".", "files": ["src/**", "test/**"] }
98
106
  ]
99
107
  }
@@ -131,6 +139,14 @@ If `<active_policies>` includes `<verification command="...">`, add those comman
131
139
 
132
140
  Use `comet memory remember ... --scope global|project` when the user explicitly asks to retain a preference or project convention long-term. Use `comet memory observe --text "<preference>" --workflow <classic|native> --change <change-id> --candidate-key <change-id>-<behavior-slug>` only for stable collaboration habits reusable across tasks without an explicit request (all three options are required; build `--candidate-key` from the change ID plus a short behavior slug). Neither may store task summaries, progress, command output, or test results.
133
141
 
142
+ Project Memory and Personal Memory are separate. Before every task ends, write project-validated and reusable experience to Project Memory:
143
+
144
+ ```text
145
+ comet knowledge remember <project-root> --title "<concise title>" --text "<symptom, approach, verification>" --type <fact|decision|pattern|procedure|constraint|failure-resolution> --json
146
+ ```
147
+
148
+ The same title updates the existing entry instead of creating a duplicate. Skip the write when nothing is reusable. The project memory index is injected with task context; expand an entry with the same task arguments plus `--expand-context "project-memory:<slug>"`. Never save task summaries, one-off command output, or unverified guesses in Project Memory.
149
+
134
150
  Before every task ends, complete one learning check. If there is a user correction, preference, or collaboration habit with a clear reuse condition, call `comet memory observe` and pass `--learning-check submitted` to the completion command. If you checked and found no qualifying observation, pass `--learning-check no-observation`; if no check was performed, pass `--learning-check not-run`. The first observation creates only a `trial` candidate, while a second independent successful observation from another change may promote it. Use the observation JSON `learning.result` and `status.learning.lastCheck` to diagnose the result.
135
151
 
136
152
  After actually using an item and learning its result, take `applications[].applicationId` from JSON (or `application_id` from Hook text) and run `comet task <project-root> --task "<original user request>" --application "<application-id>" --outcome used-successfully|ignored|overridden|corrected|contributed-to-failure --json`. Report outcomes truthfully; never mark unused items as successfully used. At task end, still run `comet task` with `--complete --workflow <workflow> --change <change-id> --learning-check submitted|no-observation|not-run` to record the checkpoint.
@@ -25,8 +25,6 @@ comet state select <change-name>
25
25
  comet state check <name> design --json
26
26
  ```
27
27
 
28
- Combine multiple read-only comet commands (for example `state get`, `state next`, `state artifacts`) into a single shell invocation to reduce process startup overhead.
29
-
30
28
  When the previous phase's guard already returned this phase's state, continue from that state and `agent.continuation` without repeating select/check; run the entry checks above only when resuming, after workspace changes, or after external state changes. After a successful check, use the returned layout, configuration, nextAction, and coordination progress summary. Avoid individual field queries or another root show. Normal entry requires only the entry check; follow context-recovery.md when resuming without context or retrieving details. Address the reported cause if validation fails.
31
29
 
32
30
  **Recovery**: Check existing artifacts and confirmation records, then complete only unfinished steps. Both normal entry and recovery preserve a registered, still-valid design and inspect `data.designReadiness`, `data.issues`, and `data.nextAction`. Restore missing files, correct the change associated with a file, or refresh an outdated handoff without clearing `design_doc`. Once the user has confirmed the design, execute the returned complete-design action; it preserves completed work. If Runtime is already in Build, the command only returns that phase's entry information.
@@ -35,6 +35,7 @@ After entering the hotfix workspace and reading current `phase`, run `comet task
35
35
 
36
36
  - Add only returned `text` to the current context. Context Manifest (`manifest` / `<context_manifest>`) contains only summaries, application reasons, and stable IDs. Add `--expand-context "<id>"` when source text, provenance, or validation details are needed. When path, operation, or phase changes, select applicable entries again using the same `--session`.
37
37
  - Use `comet memory remember ... --scope global|project` when the user explicitly asks for long-term memory. Use `comet memory observe` only for implicit, reusable, stable collaboration patterns. Do not save task summaries, progress, command output, or test results.
38
+ - Before task completion, write project-validated and reusable experience to Project Memory with `comet knowledge remember <project-root> --title "<concise title>" --text "<symptom, approach, verification>" --type <fact|decision|pattern|procedure|constraint|failure-resolution> --json`. The same title updates the existing entry; skip the write when nothing is reusable, and never save task summaries, one-off command output, or unverified guesses in Project Memory.
38
39
  - After actually using an entry and determining its outcome, take `applications[].applicationId` (`application_id` in Hook text) and run `comet task <project-root> --task "<original-user-request>" --application "<application-id>" --outcome used-successfully|ignored|overridden|corrected|contributed-to-failure --json` to record the result.
39
40
  - Before task completion, complete one learning check: when a user correction, preference, or collaboration habit has a clear reuse condition, call `comet memory observe` and pass `--learning-check submitted` to the completion command; pass `--learning-check no-observation` when you checked and found none, or `--learning-check not-run` when no check occurred. Use observation JSON `learning.result` and `status.learning.lastCheck` to distinguish candidate, promotion, deduplication, and skip; do not submit task summaries or test results.
40
41
  - At task completion, still run `comet task` with `--complete --workflow <workflow> --change <change-id> --learning-check submitted|no-observation|not-run`. Without Hooks, this Skill uses the same interface. `comet memory context` is a compatibility entry only. Plugin failures do not block the repair.
@@ -65,8 +66,6 @@ comet state select <name>
65
66
  comet state check <name> open
66
67
  ```
67
68
 
68
- Combine multiple read-only comet commands (for example `state get`, `state next`, `state artifacts`) into a single shell invocation to reduce process startup overhead.
69
-
70
69
  If select/check returns `BLOCKED` — or a branch-binding `ERROR` — because `bound_branch` differs from the current branch, pause under `comet-classic/reference/decision-point.md`. Offer a single choice: return to the bound branch and rerun entry checks, or, after the user explicitly confirms that the current branch should take over this change, run `comet state rebind <change-name>` and rerun entry checks. Do not switch or rebind branches yourself.
71
70
 
72
71
  Then create the reduced artifacts:
@@ -154,7 +153,7 @@ Reuse `/comet-verify`, whose size assessment chooses light or full verification.
154
153
 
155
154
  A small hotfix without delta spec usually meets light conditions (≤ 3 tasks and changed files below the scale threshold). Follow comet-verify's light-verification checklist. Default `review_mode: off` does not dispatch automatic code review. If the user wants review, they can set `comet state set <name> review_mode standard` or `thorough` before verification. If the hotfix creates delta spec, follow comet-verify's scale rules into full verification.
156
155
 
157
- After verification passes, record `.comet.yaml` `verify_result: pass` under `/comet-verify` rules. Do not omit that state before archive. Passing verification still leads to `/comet-archive` for final confirmation; never run archive automatically without it.
156
+ After verification and its report are complete, `/comet-verify` runs `comet guard <change-name> verify --apply`. Reuse its successful archive state and `agent.continuation`; do not manually set `verify_result: pass` or repeat Guard. If Guard fails, address its reasons without claiming acceptance passed. `/comet-archive` still requires final confirmation before archiving; never run archive automatically.
158
157
 
159
158
  ### 5. Archive
160
159
 
@@ -9,6 +9,7 @@ Native saves complete requirements, progress, and acceptance results in the proj
9
9
 
10
10
  ## Required rules
11
11
 
12
+ - Continue confirmed implementation, document repairs, evidence reuse, and recovery. Pause for new decisions, authorization, or external information. Use current continuation commands, templates, and packages; correct inputs within the same task.
12
13
  - Treat `.comet/config.yaml`, the current change, `comet-state.yaml`, and formal artifacts on disk as authoritative; chat memory is supplementary. Among formal workflow files, the Agent edits only the brief, complete target Specs, association `delta.yaml`, and `children.yaml`. Runtime owns state, check results, reports, locks, and transactions.
13
14
  - Advance through the public `comet native` CLI on PATH; do not ask the user to run commands manually. If the command is unavailable, report an incomplete installation and stop. Consult `comet native <command> --help` for arguments.
14
15
  - Create changes with the CLI; use returned paths and follow denial commands or targets before retrying. Custom and non-Comet writes stay neutral.
@@ -20,10 +21,12 @@ Native saves complete requirements, progress, and acceptance results in the proj
20
21
 
21
22
  1. If the name is known, run `comet native select <change-name> --json`. When an active change exists, enter the returned `workspace.projectRoot`; select returns the same discovery and state information as status, so a separate status call is not needed. Otherwise run `comet native status --json`. Let Runtime locate the workspace; ask the user only when multiple workspaces match equally well.
22
23
  2. If there is no matching active change, select isolation and create it using [workspace selection](reference/workspace.md#create-a-change), then enter `preparation.projectRoot`. If preparation fails, preserve any branches and directories already created and address the reported cause.
23
- 3. After entering the workspace and obtaining `phase`, retrieve context once using [memory integration](reference/commands.md#memory-integration). Expand details only when needed, record actual use outcomes, and call `comet task --complete` at the end as specified there.
24
+ 3. After entering the workspace and obtaining `phase`, retrieve context once using [memory integration](reference/commands.md#memory-integration). Expand details only when needed, record actual use outcomes, handle Project Memory and Personal Memory separately at task completion, and call `comet task --complete` as specified there.
24
25
 
25
26
  Never save task summaries, progress, command output, or test results as Personal Memory; complete the learning check.
26
27
 
28
+ Project experience and personal preferences are stored separately: before task completion, write project-validated and reusable experience to Project Memory with `comet knowledge remember`; only user preferences and stable collaboration habits go to Personal Memory. See [memory integration](reference/commands.md#memory-integration) for the commands and completion conditions.
29
+
27
30
  ## Read only what the action needs
28
31
 
29
32
  Read the section for the current action. Follow links within it only when their stated conditions apply; do not load the entire command reference or all references at once.
@@ -38,11 +41,11 @@ Read the section for the current action. Follow links within it only when their
38
41
 
39
42
  ## Shape
40
43
 
41
- Investigate facts that can be established without the user. Ask only about decisions that change user-visible outcomes and cannot be inferred reliably. For simple requests, list unresolved questions and dependencies; maintain a decision tree only when several decisions affect one another. Before asking under `native.clarification_mode`, save this round's unresolved questions in the brief. Immediately copy confirmed conclusions into Decisions, the brief, and complete target Specs. Keep unanswered parts `[blocking]`.
44
+ Investigate facts that can be established without the user. Ask only about decisions that change user-visible outcomes and cannot be inferred reliably. For simple requests, list unresolved questions and dependencies; maintain a decision tree only when several decisions affect one another. Before asking under `native.clarification_mode`, save this round's unresolved questions in the brief. Immediately copy confirmed conclusions into the relevant brief sections and complete target Specs. Keep unanswered parts `[blocking]`.
42
45
 
43
46
  Complete when requirements sources are fully processed within the coverage boundary and classified by purpose, all outcome-affecting decisions and assumptions are resolved, no `[blocking]` remains, the user explicitly confirms the outcome, scope, key decisions, all acceptance items, and non-goals, and Runtime has entered Build.
44
47
 
45
- At a new or reconfirmed Shape, Runtime checks all eight brief sections, a complete target Spec or an explicit no-product-behavior-change reason, and formal paths. Failures name the artifact and repair action; edit it and rerun continuation. Later-phase existing changes keep their progress until Shape.
48
+ New or reconfirmed briefs need outcome, scope, non-goals, and acceptance examples. Add constraints, decisions, open questions, or special verification requirements when applicable. Runtime also checks formal paths and a complete target Spec or concrete no-product-behavior-change reason. Repair reported artifacts and rerun continuation. Existing later-phase changes retain progress until Shape.
46
49
 
47
50
  ## Build ↔ Verify Loop
48
51
 
@@ -54,7 +57,7 @@ After the Builder submits a candidate, Runtime runs required checks and a new re
54
57
 
55
58
  Before the first implementation, read the current brief, complete target Specs, and every acceptance item. Edit project code and tests within confirmed scope. During repair, prioritize the Verifier's failed or blocked items and failed checks, then recheck other confirmed behavior before submission. `previous_unresolved_ids` identifies the repair focus; the next formal verification still covers every acceptance item.
56
59
 
57
- Build, Verify, and Archive recheck bound documents. Drift or stale reports return a repair action and preserve work; Runtime still checks without a Hook. Ordinary Markdown is unaffected. Ordinary documentation writes (Markdown/text and LICENSE-style files at the repository root or under `docs/`, `doc/`, `documentation/`, `.github/`, outside the Native artifact root) are neutral during Shape, Verify, and Archive: they do not return the change to Build and do not invalidate the current candidate. Set `native.document_writes: revert` in `.comet/config.yaml` to restore the strict behavior.
60
+ Build, Verify, and Archive recheck formal bindings. New Shapes bind Markdown content: blank lines and soft wraps preserve confirmation; content, structure, code, link, or acceptance changes require reconfirmation. After an edit, Hooks check actual content before the next implementation write; Runtime also checks before advancement. Existing Shapes retain their binding. Invalid documents or reports return repair actions and preserve work. Ordinary documents preserve candidates by default; use `native.document_writes: revert` for strict behavior. See [formal artifacts](reference/artifacts.md#formal-artifacts) for paths.
58
61
 
59
62
  User Hook output can use `hook.allow_paths` in `.comet/config.yaml`; see [User Hook writes](reference/commands.md#user-hook-writes).
60
63
 
@@ -72,17 +75,17 @@ Complete when the implementation and relevant checks are ready for verification,
72
75
 
73
76
  ## Verify
74
77
 
75
- Immediately launch an independent read-only Verifier under the Verify protocol. The dispatch record only registers the attempt: until the platform tool accepts the launch and the Verifier reports `verifier-started`, do not treat "dispatched" as "running" or report it that way to the user; if the launch call itself fails or is rejected, handle it through the exception flow immediately. The Verifier checks that recorded results match the current candidate, workspace, and inputs, adds only missing or invalidated checks, and independently assesses every acceptance item. The Builder passes only the implementation location, acceptance IDs and references, check-record locations, known limitations, and relevant file locations. Read log bodies on demand.
78
+ Launch a new read-only Verifier immediately under the Verify protocol with the unchanged task package. Report it as running only after the platform accepts startup and the Verifier reports `verifier-started`; handle launch failures immediately. Use inline acceptance text or page through every scopeId. The Verifier assesses all acceptance items independently, reuses Runtime checks matching implementation, workspace, and inputs, and adds missing or invalidated evidence. Read files and logs on demand.
76
79
 
77
- While awaiting the result, use `status` to distinguish "registered but not started" from "Verifier confirmed startup"; when startup is unconfirmed and the subagent is unresponsive, verify the dispatch actually succeeded before waiting on or failing the attempt. A wait-tool timeout means keep waiting for the same Verifier. Record an execution error only when the platform confirms execution failure, an execution timeout, a lost task, or completion without a usable result. Once Runtime accepts the complete result, follow the latest state. When user acceptance is required, run `--accept-result` only after explicit acceptance. Results with automated checks but no independent verification also require explicit acceptance.
80
+ After a wait-tool timeout, keep waiting for the same Verifier. If its receipt is missing and it is unresponsive, check launch success. Report errors only for confirmed failure, execution timeout, task loss, or completion without a usable result. Follow current state after Runtime accepts results. Run `--accept-result` only after explicit user acceptance, including automated-check-only results. For isolated workspaces, accept results and choose delivery in one reply with `--finish`; ask separately if no finish was chosen.
78
81
 
79
82
  Complete when Runtime accepts each verdict and supplies the next action. Continue repairs on Build, or address the specified waiting or blocking condition. Ending a phase does not mean the task is finished.
80
83
 
81
84
  ## Archive
82
85
 
83
- When `continuation` permits Archive, first read [Archive completion](reference/workspace.md#archive-completion). Use the accepted verification result and execute the complete returned `archive --dry-run` command. Address only blockers in that response. Run its single returned `archive --confirmed` command only after `ready: true`.
86
+ When `continuation` permits Archive, read [Archive completion](reference/workspace.md#archive-completion), reuse the accepted result, and follow continuation. After an explicit finish choice, execute its complete `--confirmed --finish` command; Runtime preflights and rechecks the transaction. If dry-run is returned, preview and run its unique confirmed command after `ready: true`. Address the returned blockers.
84
87
 
85
- Commit only this change's implementation and formal artifacts; preserve unrelated edits. Inspect `workspaceFinishResult`, preserving the workspace and following `recoveryArgs` if blocked.
88
+ Commit only this change's implementation and formal artifacts; preserve unrelated edits. Inspect `workspaceFinishResult`, preserving the workspace and following `recoveryArgs` if blocked. Fix commit messages and local blockers under existing authorization and continue; ask only for new authorization or external information.
86
89
 
87
90
  Complete when state is `done`, authorized workspace finishing is `completed` or `kept`, and task completion has been recorded through memory integration. Continue handling any other result.
88
91
 
@@ -31,25 +31,25 @@ Use `comet-state.yaml` to restore workflow state across devices. Runtime updates
31
31
 
32
32
  Shape, Build, Verify, and Archive recheck the binding between formal files and confirmed requirements. A target specification must be at `specs/<capability>/spec.md`; other files under `specs/` are not formal Specs, and the Hook rejects them with the expected path. Empty documents, heading- or fence-only content, and template-only placeholders such as `TODO`, `<TODO>`, or `{{reason}}` do not satisfy a complete specification. A no-product-behavior exemption must give a concrete reason; comments and placeholder reasons do not count. If Archive dry-run reports only that `verification.md` is missing, stale, or invalid, `continuation` directly supplies `comet native doctor <change> --repair`; rerun dry-run after the repair.
33
33
 
34
+ During Shape, Verify, and Archive, ordinary documentation writes preserve the current phase and candidate by default: Markdown/text and LICENSE-style files at the repository root or under `docs/`, `doc/`, `documentation/`, and `.github/`, outside the Native artifact root. Set `native.document_writes: revert` in `.comet/config.yaml` for strict behavior. Formal requirements files always follow the binding rules above.
35
+
34
36
  `.comet/config.yaml` selects the workflow and artifact directory. Synchronize it when a nondefault artifact directory must work across devices; other `.comet/*` data stays local.
35
37
 
36
38
  Runtime's `artifacts` object is the sole location authority for the current workspace: `briefPath`, `childrenPath`, `specsDir`, and `statePath` are under the configured `<artifact-root>/comet/changes/<change-name>/`; local `runtimeDir` remains under `.comet/runtime/native/`. Do not substitute a same-named file from `.comet/comet/` or another directory when a formal artifact is missing. If the Hook reports a path error, retry the original edit at the exact absolute path it supplies, preserving existing content and reading before merging when needed.
37
39
 
38
40
  ### Brief
39
41
 
40
- `brief.md` records Native requirements clarification. Use these level-one headings, each with nonempty content:
42
+ `brief.md` records Native requirements clarification. Use these four core level-one headings, each with nonempty content:
41
43
 
42
44
  ```text
43
45
  # Outcome
44
46
  # Scope
45
47
  # Non-goals
46
48
  # Acceptance examples
47
- # Constraints and invariants
48
- # Decisions
49
- # Open questions
50
- # Verification expectations
51
49
  ```
52
50
 
51
+ Add `Constraints and invariants`, `Decisions`, `Open questions`, or `Verification expectations` only for actual constraints, key decisions, unresolved questions, or special check requirements. Fill any existing section with meaningful content; simple tasks need no placeholder "none" sections. Record user decisions in relevant core sections, using Decisions when several decisions need tracking.
52
+
53
53
  Use these markers in Open questions only for genuinely unresolved user questions:
54
54
 
55
55
  ```text
@@ -57,9 +57,9 @@ Use these markers in Open questions only for genuinely unresolved user questions
57
57
  - [blocking] Q1: <Batch question>
58
58
  ```
59
59
 
60
- After the user confirms each decision, immediately record it in Decisions and the complete target Specs, then remove its blocker. Runtime saves the final Shape summary and waits through `await-user`; do not add another confirmation question to the brief. Formal artifacts record conclusions and reasons, not the model's hidden reasoning. Use [source-document full coverage](#source-document-full-coverage) to record how requirements sources map to Specs and acceptance items.
60
+ After the user confirms each decision, immediately record it in the relevant brief sections and complete target Specs, then remove its blocker. Runtime saves the final Shape summary and waits through `await-user`; do not add another confirmation question to the brief. Formal artifacts record conclusions and reasons, not the model's hidden reasoning. Use [source-document full coverage](#source-document-full-coverage) to record how requirements sources map to Specs and acceptance items.
61
61
 
62
- Acceptance criteria must be specific, verifiable, and nonduplicative. Runtime generates items only from top-level brief acceptance examples and complete Spec scenarios explicitly marked `Scenario:`. Explanatory paragraphs, ordinary lists, and isolated WHEN/THEN lines are not extra items. Acceptance IDs are sequential, such as `A1`, `A2`, and `A3`; they associate verdicts with items, are not content hashes, and do not identify files. Runtime saves the complete acceptance text and its source when Shape is confirmed.
62
+ Acceptance criteria must be specific, verifiable, and nonduplicative. Runtime generates items only from top-level brief acceptance examples and complete Spec scenarios explicitly marked `Scenario:`. Explanatory paragraphs, ordinary lists, and isolated WHEN/THEN lines are not extra items. Acceptance IDs are sequential, such as `A1`, `A2`, and `A3`; they associate verdicts with items, are not content hashes, and do not identify files. Runtime saves the complete acceptance text and its source when Shape is confirmed. When a Spec scenario corresponds to an existing brief acceptance item, write `Acceptance: A1` (or Chinese `验收:A1`) on its own line in the scenario body, referencing the current brief's actual ID to avoid a duplicate acceptance item. Each scenario may reference one ID; new behavior needs a separate scenario. References combine acceptance identity only: complete Specs remain bound to confirmation and must be checked by the Verifier.
63
63
 
64
64
  New `children.yaml` files use `comet.native.children.v2`:
65
65
 
@@ -62,7 +62,7 @@ After all questions, explain how to reply, for example `Q1: A; Q2: A+C; Q3: addi
62
62
  1. Investigate prerequisite facts; pause only branches whose prerequisites are unresolved.
63
63
  2. Select one ready question and first save `- [blocking] <question>` in the brief's `# Open questions`.
64
64
  3. Ask only this question, including the recommendation, reason, and each option's impact, then wait.
65
- 4. Immediately record confirmed decisions in Decisions, the brief, and complete target Specs, then remove resolved blockers. Save follow-up answers in the same change.
65
+ 4. Immediately record confirmed decisions in the relevant brief sections and complete target Specs, then remove resolved blockers. Save follow-up answers in the same change.
66
66
  5. Update question dependencies, determine what is ready next, and begin the next round.
67
67
 
68
68
  #### Batch mode
@@ -70,7 +70,7 @@ After all questions, explain how to reply, for example `Q1: A; Q2: A+C; Q3: addi
70
70
  1. Identify every question ready for this round: prerequisite facts and decisions are resolved, and answers are independent. Keep each independent decision as a separate question.
71
71
  2. Before asking, save all questions in the brief's `# Open questions` with stable IDs such as `- [blocking] Q1: <question>` and `- [blocking] Q2: <question>`. Never reuse an existing ID for another question in later rounds.
72
72
  3. Ask the complete current set at once, giving each question its own recommendation, reason, and impacts, then wait. Do not split the round because of tool limits.
73
- 4. Update Decisions, the brief, and complete target Specs item by item, removing resolved questions. Preserve original IDs and `[blocking]` for partial, ambiguous, or unanswered questions.
73
+ 4. Update the relevant brief sections and complete target Specs item by item, removing resolved questions. Preserve original IDs and `[blocking]` for partial, ambiguous, or unanswered questions.
74
74
  5. Update answered and unanswered questions and their dependencies, then determine the next complete set.
75
75
 
76
76
  In either mode, if no questions are ready, continue investigating unresolved facts and checking for missed cases. An empty current list does not mean clarification is complete; all final-confirmation conditions still apply.
@@ -104,9 +104,9 @@ Whenever mode selection is needed, show both A and B. Use this table for text; i
104
104
 
105
105
  An explicit request for “multiple sessions,” “independent sessions,” “cross-session coordination,” or an “Agent Team” selects A; do not ask again. Before confirmation, do not create child changes, worktrees, Codex independent sessions, a Claude Code Agent Team, or dispatch tasks.
106
106
 
107
- When final Shape contains a Supervisor Change with two or more children, explicitly record the Supervisor Change and every Child in Decisions. Before preparing final requirements confirmation, require a choice between multi-session coordination and single-session progression; a generic confirmation is insufficient and there is no implicit default. Then execute `prepare-shape-confirmation` from continuation. Runtime saves the mode and separately waits for confirmation of the complete Shape. The user must explicitly confirm the complete Shape before the `--confirmed` alternative is executed.
107
+ When the final Shape contains two or more children, record the Supervisor Change and each child in Decisions, then execute `prepare-shape-confirmation` from continuation so Runtime binds the complete Shape and child plan. Present the saved complete summary, decomposition, and both A and B choices. The user may explicitly confirm the complete plan and choose its mode in one reply; execute the matching complete `--confirmed` alternative afterward. A generic confirmation is insufficient to select a mode, and selecting a mode alone does not confirm the plan. The separate mode-selection path remains available: after selecting only a mode, the user must explicitly confirm the complete Shape. If formal artifacts or children change, prepare and confirm again instead of reusing prior authorization.
108
108
 
109
- During confirmation preparation, Runtime writes `coordination_mode` to `comet-state.yaml`. Only after the user confirms the complete Shape does Runtime enter Build, create the Supervisor integration branch and worktree, and prepare child task packages from the integration branch's current commit, including role, worktree, baseline commit, and `runId`. Do not write the mode into `children.yaml`; it does not change Runtime's `readyChildren`, `runId`, verification, or integration rules. Start only children listed in `readyChildren`. Mode A starts at most two children without unresolved dependencies concurrently; mode B runs them sequentially. Every child stays within confirmed Supervisor scope. Return to Supervisor Shape for a new decision that changes user-visible outcomes.
109
+ Runtime writes the selected mode to `coordination_mode` in `comet-state.yaml`. Only after both the complete Shape and coordination mode are explicitly confirmed does Runtime enter Build, create the Supervisor integration branch and worktree, and prepare child task packages from the integration branch's current commit, including role, worktree, baseline commit, and `runId`. Do not write the mode into `children.yaml`; it does not change Runtime's `readyChildren`, `runId`, verification, or integration rules. Start only children listed in `readyChildren`. Mode A starts at most two children without unresolved dependencies concurrently; mode B runs them sequentially. Every child stays within confirmed Supervisor scope. Return to Supervisor Shape for a new decision that changes user-visible outcomes.
110
110
 
111
111
  On `/comet-native` resume, use Runtime's saved `coordination_mode`; do not duplicate children or worktrees and do not ask for the coordination mode again. `multi-session` continues multi-session coordination with automatic subagent fallback when independent sessions or Agent Teams are unavailable. `single-session` continues sequentially in the current session. If original Codex sessions or a Claude Code Agent Team no longer exist, reread Runtime state first. Do not infer child completion from old session or team state, and do not automatically switch to single-session progression.
112
112
 
@@ -22,6 +22,16 @@ Save memory as follows:
22
22
  - Only when the user has not requested long-term memory, but a collaboration practice is stable and reusable in future tasks, call `comet memory observe <project-root> --text "<collaboration practice>" --workflow <workflow> --change <change-id> --candidate-key <stable-topic-key> --json`.
23
23
  - Neither command may save task summaries, implementation progress, command output, or test results.
24
24
 
25
+ ### Project memory
26
+
27
+ Project Memory is separate from Personal Memory. Before task completion, write facts, decisions, patterns, procedures, constraints, or failure resolutions that are verified in the current project and reusable in future tasks:
28
+
29
+ ```text
30
+ comet knowledge remember <project-root> --title "<concise title>" --text "<symptom, approach, verification>" --type <fact|decision|pattern|procedure|constraint|failure-resolution> --json
31
+ ```
32
+
33
+ The same title updates the existing entry instead of creating a duplicate. Skip the write when nothing is reusable. The project memory index is injected with task context; expand one entry with the same task arguments plus `--expand-context "project-memory:<slug>"`. Never save task summaries, one-off command output, or unverified guesses in Project Memory.
34
+
25
35
  Before every task ends, complete one learning check. When the task contains a user correction, preference, or collaboration practice with a clear reuse condition, call `comet memory observe` and then pass `--learning-check submitted` to `comet task ... --complete`. If you checked and found no qualifying observation, pass `--learning-check no-observation`; if no check was performed, pass `--learning-check not-run`. The first observation creates only a `trial` candidate; a second independent successful observation from a different change may promote it. Do not submit a task summary or test result to manufacture a record.
26
36
 
27
37
  The observation JSON `learning.result` reports `candidate-created`, `candidate-promoted`, `deduplicated`, `ignored`, or `skipped`. `status.learning.lastCheck: not-run` means the host did not submit this task's learning check; it does not mean that no reusable information exists.
@@ -53,7 +63,7 @@ When the user explicitly wants to revoke a capability association, first run `co
53
63
 
54
64
  Read this section before first filling a Runtime template or returning a result through `returnAction`.
55
65
 
56
- Copy `inputOptions.template` into a system temporary JSON file, replace only the requested values, then execute `continuation.commandArgs` or the selected `commandAlternative.commandArgs`. Delete the temporary file after the command completes. Preserve all supplied iteration, attempt, state-version, and task identifiers exactly; fill only fields exposed by the template.
66
+ Copy `inputOptions.template` into a system temporary JSON file, replace only the requested values, then execute `continuation.commandArgs` or the selected `commandAlternative.commandArgs`. Delete the temporary file after Runtime accepts the input; retain rejected input, correct it as directed, and resubmit. Preserve all supplied iteration, attempt, state-version, and task identifiers exactly; fill only fields exposed by the template.
57
67
 
58
68
  Options in the same `exclusiveGroup` within `inputOptions` are mutually exclusive. Choose one and place its `template` in the temporary file as a single JSON object. On validation failure, correct that file using the JSON paths, missing fields, and unknown fields listed in `error.issues`.
59
69
 
@@ -69,11 +79,9 @@ An ordinary change or Supervisor parent does not need an additional read-only re
69
79
 
70
80
  The Builder handoff must describe this iteration's changes, acceptance items addressed, development checks actually run and not run, and known limitations. An earlier review cannot replace the formal Verifier, which still independently assesses every acceptance item.
71
81
 
72
- During development, prefer focused checks that give fast feedback on the current change. At handoff, put the final check plan in `builder-handoff.verification_checks` instead of running the same complete plan directly in the Builder first. Runtime freezes the candidate, executes the plan, and returns `runtimeCheckExecution.disposition`:
82
+ Before handoff, review the current brief, complete target Specs, and every confirmed acceptance ID, then fill `acceptance_review` from the Runtime template with exactly one entry per ID. Preserve each template `id`, report `status` truthfully, list specific file locations, check records, or reproducible observations in `evidence`, and explain how the implementation satisfies the item in `note`. Submit only when every item is `implemented-with-evidence` with nonempty evidence. `implemented-no-evidence`, `not-implemented`, and `known-fail` mean missing evidence, missing implementation, and a known failure; complete that work before submitting a candidate for verification. Repair templates retain prior evidence for unaffected items and request current evidence for affected items. Before submission, check every entry still applies and update invalidated evidence. Prefilled content is not a new check result. Self-review does not establish independent acceptance.
73
83
 
74
- - `executed`: Runtime ran the plan in this action. After every check passes, Runtime enters Verify and prefills the same plan in the next `dispatch-verifier` action.
75
- - A failed check, or an interrupted nonrepeatable check: Runtime returns the change to Build; repair the implementation and submit a new Builder handoff.
76
- - An interrupted repeatable check: retry only the checks named by the latest `retry-checks` template.
84
+ Run focused development checks and put the final plan in `builder-handoff.verification_checks`; do not run the same complete plan beforehand. Runtime freezes the candidate and runs the plan: passing returns the `verifierDispatch` task package directly for immediate handoff to a new read-only Verifier; failure or an interrupted nonrepeatable check returns to Build; retry an interrupted repeatable check only through `retry-checks`. `runtimeCheckExecution.disposition` distinguishes execution from reuse.
77
85
 
78
86
  Use `verification_checks` only for commands Runtime can safely manage on the current candidate. Leave external-service checks, one-time operations, and steps that cannot be safely retried for the Verifier to assess under their existing execution constraints. `builder-handoff.checks` continues to report development checks the Builder did or did not run; it never becomes formal Runtime check evidence.
79
87
 
@@ -96,22 +104,22 @@ Pass these `verifierDispatch` directories and references unchanged when launchin
96
104
  - `changeDir`: the base directory for relative `briefRef` and `specRefs[].ref` paths.
97
105
  - `supervisorStateRef`: the local state file with child acceptance and integration records; `null` for ordinary changes.
98
106
 
99
- Pass any `recoveryContext` unchanged too; it contains the latest recovery or user-provided information. `detailsPageArgs` already includes `--project-root`; retain it when querying from any directory. After additional checks run, return Runtime's results and handoff information to the current Verifier and keep waiting for its final result.
107
+ Pass optional `recoveryContext` unchanged. Read all inline `acceptance` text when provided; otherwise follow `detailsPageArgs` and subsequent pages through every `scopeId`, retaining `--project-root`. The Verifier that actually starts copies `startupInput` to submit its receipt. After additional checks, return Runtime's results and handoff to the current Verifier and keep waiting.
100
108
 
101
109
  When Runtime requests `dispatch-verifier`:
102
110
 
103
- 1. Prefer submitting the final check plan through Builder handoff `verification_checks`; if it was not submitted there, add it to the `dispatch-verifier` `inputOptions.template`. Runtime evidence binds the result to the candidate, workspace, machine, command, inputs, and tool environment. After handoff checks pass, execute Runtime's prefilled `dispatch-verifier` unchanged. When the binding is unchanged, Runtime returns `runtimeCheckExecution.disposition=reused` and does not execute the same plan again; a changed plan or binding is executed again. Ordinary Builder logs cannot replace this evidence.
111
+ 1. Submit the final check plan through handoff `verification_checks` and use the returned task package directly. Fill the dispatch template only for older handoffs without a plan. Unchanged evidence bindings return `runtimeCheckExecution.disposition=reused`; otherwise checks execute. Builder logs are not Runtime evidence.
104
112
  2. After interrupted checks, retry only the repeatable checks listed by the latest `continuation` action `retry-checks`. Do not treat assertion failures or nonrepeatable checks as environment failures to retry automatically.
105
- 3. Read `verifierDispatch` for workspace and check-record locations, `scopeIds`, `scopeCount`, total acceptance count, brief/Spec references, detail pagination arguments, optional review summary, and check results. The package does not inline every acceptance text; use pagination to read every scenario covered by `scopeIds`.
113
+ 3. Read `verifierDispatch` for the workspace, check records, `scopeIds`, `scopeCount`, brief/Spec references, review summary, and check results. Read acceptance text as specified above. Consult the handoff in `builderSummary` after independent inspection.
106
114
  4. Immediately launch a new read-only Verifier subagent through the platform's native capability. Pass directories, check-record locations, and any `recoveryContext` unchanged. If the launch call is rejected or fails, handle it as `verifier-execution-error` immediately; the dispatch is complete only after the platform accepts the launch. If subagents are unavailable, a separate Agent session from the Builder is allowed only when the user selected multi-session coordination and the platform can manage independent sessions. Otherwise report Verifier unavailability as specified below and follow the latest `continuation`.
107
115
 
108
116
  `dispatch-verifier` registers the attempt and returns its package and attempt identifier. It does not start an independent service or process, and requires no service address or callback. Verifier results must return this package's `candidateId` and `verifierExecutionRef` unchanged. Runtime rejects late results for older candidates or Verifier tasks.
109
117
 
110
- While waiting, use `status` and read `localExecution.verifierStartup`: `unconfirmed` means only the dispatch record exists and the Verifier has not yet contacted Runtime; `confirmed` means the Verifier reported startup or already added checks. When unconfirmed and the subagent is unresponsive, verify the dispatch actually succeeded. A missing receipt alone is not an execution failure; recording an error still follows the conditions under "Independent acceptance and results" below.
118
+ Follow Runtime's instructions based on `localExecution.verifierStartup`: for `unconfirmed`, check whether the platform accepted the launch and locate the original task. Launch only if no startup call was made; otherwise keep waiting. `confirmed` means the Verifier reported startup or requested checks; wait for that same Verifier. A missing receipt or wait-tool timeout alone is not an execution failure. Report errors only under the conditions below, without dispatching another task.
111
119
 
112
120
  ### Independent acceptance and results
113
121
 
114
- The Verifier remains read-only throughout. Its first Runtime action is the `verifier-started` startup receipt (`candidateId` and `verifierExecutionRef` copied unchanged from the package; repeated submission has no side effect). It then reads the scenarios identified by current `scopeIds`, the brief, complete target Specs, actual implementation, and Runtime check results. Confirm that recorded checks match the current candidate, workspace, and inputs and cover the current scope. Add only missing or invalidated checks through `inputOptions.template` for Runtime to execute. Independently assess every acceptance item in `scopeIds` regardless of check reuse.
122
+ The Verifier remains read-only and first submits the package's `startupInput` (`verifier-started`; repeated submission has no side effect). Read all current-scope acceptance text, the brief, complete target Specs, implementation, and Runtime check results. Confirm the candidate, workspace, inputs, and coverage. Add only missing or invalidated checks through `inputOptions.template` for Runtime to execute; independently assess each item in `scopeIds`.
115
123
 
116
124
  Read the Builder handoff last, as investigation leads. The Builder provides only implementation locations, acceptance IDs and references, check-record locations, known limitations, and relevant file locations. Read log bodies on demand.
117
125
 
@@ -119,6 +127,8 @@ A wait-tool timeout means keep waiting for the same Verifier. Record an executio
119
127
 
120
128
  For `verifier-response`, submit this result shape: The response lists only the current `scopeIds` and marks each scenario exactly once as `passed`, `failed`, or `blocked`. Give a concrete reason for failed or blocked items. Runtime filters a known superset only when the extra criteria already passed and still report `passed`; nonexistent IDs, duplicates, missing scope IDs, and out-of-scope `failed` or `blocked` results remain invalid.
121
129
 
130
+ When a result or check request is rejected for fields, format, or acceptance scope, correct the original input and resubmit through the returned current `continuation`. Keep the active Verifier; do not report `verifier-execution-error` or redispatch it. For candidate or execution-binding mismatches, first check the dispatch identity; never relabel an old task's result as a new task's result. Actual check execution failures and platform task failures still follow their corresponding failure paths.
131
+
122
132
  After submission of a repaired implementation, Runtime retains still-valid check receipts and a new formal Verifier assesses every scenario in one round. Once all pass, wait directly for user acceptance. Do not automatically clear results and add another identical full verification round.
123
133
 
124
134
  Distinguish inability to complete verification:
@@ -135,6 +145,8 @@ After a Skill-launched final Verifier passes and Runtime waits for a decision, u
135
145
 
136
146
  - `retry-checks`: retry only checks Runtime marks interrupted and repeatable for the current candidate. Copy the latest continuation's `check_ids`; do not replace commands or the candidate. Each check may execute at most three times. Successful results and valid logs are retained.
137
147
 
148
+ When the Verifier explicitly requests a failed check rerun, submit the original check plan through the current `request-checks` template. Only repeatable checks with matching candidate and execution bindings and remaining attempts can rerun. Runtime reuses passed results that remain valid. Assertion failures remain failures and do not trigger automatic retry loops; resolve blockers before repeating nonrepeatable operations.
149
+
138
150
  Complete when Runtime accepts the complete Verifier result and explicitly enters Build, Archive, `await-user`, `blocked`, or `done`.
139
151
 
140
152
  ## Supervisor coordination
@@ -186,7 +198,7 @@ Runtime can validate saved check records and their candidate, workspace, machine
186
198
 
187
199
  ## Command inputs and exceptions
188
200
 
189
- For normal flow, execute the commands Runtime supplies in `continuation`. This section explains returned fields and handles rejected input, unavailable Verifiers, Verifier execution errors, missing external information, and decisions to accept results without complete independent verification. `continuation.disposition` says whether to continue, wait for the user, address a blocker, or finish. Execute commands containing `--confirmed` only after explicit user confirmation. CLI text begins with a readable `summary`, a single `NEXT:`, and optional `RELAY TO USER:`. Use `--json` for structured responses; use `--verbose` only to diagnose local execution.
201
+ For normal flow, execute the commands Runtime supplies in `continuation`. During recovery, `next --summary` without action input returns the current template and preserves the phase; fill that template directly without another status query. This section explains returned fields and handles rejected input, unavailable Verifiers, Verifier execution errors, missing external information, and decisions to accept results without complete independent verification. `continuation.disposition` says whether to continue, wait for the user, address a blocker, or finish. Execute commands containing `--confirmed` only after explicit user confirmation. CLI text begins with a readable `summary`, a single `NEXT:`, and optional `RELAY TO USER:`. Use `--json` for structured responses; use `--verbose` only to diagnose local execution.
190
202
 
191
203
  The CLI is authoritative for command signatures and current arguments:
192
204
 
@@ -198,7 +210,7 @@ comet native <group> <command> --help
198
210
 
199
211
  ### Runtime's next action
200
212
 
201
- - `disposition`: whether to continue, wait, handle blockers, or finish. When `userCommunication.required` is true, relay the message and wait before any confirmation command.
213
+ - `disposition`: whether to continue, wait, handle blockers, or finish. When `requiresUserDecision` is true, relay the message and wait for a decision. `userCommunication.required` may only require a recovery update; then continue within existing authorization.
202
214
  - `commandArgs` / `commandAlternatives`: complete command arguments. Each alternative represents a mutually exclusive user decision; execute the matching one and do not combine alternatives.
203
215
  - `inputOptions`: fields and JSON templates to fill for this command.
204
216
  - `workspace` / `preparation`: the actual working directory and change-creation result.
@@ -207,7 +219,7 @@ comet native <group> <command> --help
207
219
  - `verifierDispatch`: workspace and evidence locations, current `scopeIds`, `scopeCount`, total acceptance count, content references, detail pagination arguments, review summary, and check results for independent Verifier dispatch. Pass any `recoveryContext` directly as the latest recovery or user-provided information.
208
220
  - `workspaceFinishResult` / `recoveryArgs`: post-Archive workspace result and recovery commands.
209
221
 
210
- At Archive-ready, first execute continuation's `archive --dry-run`. If an isolated workspace has no selected finish action, wait for the user, then either run `comet native archive <change-name> --confirmed --finish <chosen mode>` in one step (Runtime records the choice and revalidates inside the transaction; no second dry-run is needed) or use the matching `--dry-run --finish` command in `commandAlternatives` for a preview. Do not append other arguments yourself. The dry-run checks both archive content and the branches and files involved in Git finishing. On `ready: false`, address `blockers` and the complete `workspaceFinishBlockers[].paths` list from that same response; do not add a `status` query or manually commit change state/verification files. Only on `ready: true` execute the single returned `archive --confirmed` command.
222
+ At Archive-ready, follow the latest continuation. If an isolated workspace has no selected finish action, wait for the user, then execute the complete matching `--confirmed --finish` alternative, retaining its state-version arguments. Runtime saves the choice, performs preflight, and revalidates inside the transaction. Reuse a finish choice made when accepting the result without asking again. Use dry-run when a custom commit message needs previewing; do not append other arguments yourself. The dry-run checks both archive content and the branches and files involved in Git finishing. On `ready: false`, address `blockers` and the complete `workspaceFinishBlockers[].paths` list from that same response; do not add a `status` query or manually commit change state/verification files. Only on `ready: true` execute the single returned `archive --confirmed` command.
211
223
 
212
224
  Angle brackets in templates mark values to fill. `await-user` means wait for the user's decision before running advancement commands. If `commandArgs` is `null` and `commandAlternatives` is present, obtain the decision first, then execute the selected alternative's complete `commandArgs`, retaining `--expected-state-version` and `--expected-action`. If stale state or a mismatched action rejects the command, reread the latest `continuation` and follow current state; do not construct commands without state guards. `localExecution: absent` means this machine has no currently running execution, not that the change is damaged.
213
225
 
@@ -12,7 +12,7 @@ Read only the section relevant to the current blocker.
12
12
 
13
13
  ## Fault recovery
14
14
 
15
- On Runtime failure, stop editing the project and rerun `status --details --json` and read-only `doctor`. Execute only recovery actions explicitly returned by `continuation` or `doctor`. Runtime owns portable state, local execution state, locks, and transactions. If safe automatic recovery cannot be established, preserve the workspace and wait for a user decision.
15
+ When a command returns a specific error, current state, and recovery action, use that response first. Correct Runner validation errors through [filling command inputs](commands.md#filling-command-inputs), and continue commit or merge failures through [Archive completion](workspace.md#archive-completion), keeping active task identifiers. If current state is missing, the call was interrupted, or the workspace disagrees with the response, stop dependent edits and run `status --details --json` once. Run read-only `doctor` only if the state or recovery records still do not explain the issue. Execute only recovery actions explicitly returned by `continuation` or `doctor`. Runtime owns portable state, local execution state, locks, and transactions. If safe automatic recovery cannot be established, preserve the workspace and wait for a user decision.
16
16
 
17
17
  For external-input waits, follow [external input and monitoring](#external-input-and-monitoring) while independent work continues. Before redispatching Supervisor tasks after recovery, read [Supervisor coordination](commands.md#supervisor-coordination). Before relaunching a Verifier, read [Verify protocol](commands.md#verify-protocol) and check current task identifiers, candidate version, and execution state.
18
18
 
@@ -31,8 +31,8 @@ Runtime blocks writes if project root, branch, workspace type, or Git state disa
31
31
  - Verify (`verify-ready`): rerun checks needed for the current implementation and launch a new Verifier; do not reuse passes from the old device.
32
32
  - Archive (`archive-ready`): safely return to Verify, reset acceptance to `pending`, and assess the implementation synchronized to this device.
33
33
  - `await-user` / `blocked`: restore the original blocker, responsible party, and allowed actions, then wait for the corresponding condition.
34
- - `done` in active: finish only directory moves and cleanup that can be established safely.
35
- - `done` in archive: show it read-only; the change is finished.
34
+ - `done` in active: perform directory moves and cleanup only as Runtime directs; continue pending Git finishing through [Archive completion](workspace.md#archive-completion).
35
+ - `done` in archive: check workspace finishing in public status first. Show it read-only only after the selected finish is completed or kept; continue incomplete or `blocked` finishing through [Archive completion](workspace.md#archive-completion).
36
36
 
37
37
  Treat old processes, log connections, and Agent sessions as lost; leftover files do not prove success. If checks ended before YAML recorded them, rerun only safe repeatable checks. Operations that could duplicate external effects require a user decision.
38
38
 
@@ -32,7 +32,7 @@ Only when a branch was renamed, repurposed by the user, or has unclear ownership
32
32
 
33
33
  ## Archive completion
34
34
 
35
- Continue only when `continuation` permits Archive. Reuse the accepted verification result. A `current` workspace needs no finish choice: show the current branch and directory, explain that no merge, push, or PR creation will occur, and follow the latest `continuation`.
35
+ Continue only when `continuation` permits Archive. Reuse the accepted verification result. Reuse any finish choice made when accepting the result without asking again. A `current` workspace needs no finish choice: show the current branch and directory, explain that no merge, push, or PR creation will occur, and follow the latest `continuation`.
36
36
 
37
37
  For `branch` or `worktree` isolation that needs a finish choice, show the actual change branch, target branch, and directory once and offer all options below as single choice. Text questions must use this table. Structured questions must use the mode as the short label and its actual impact as the description, not just `merge`, `push`, `pull-request`, or `keep`.
38
38
 
@@ -46,15 +46,21 @@ For `branch` or `worktree` isolation that needs a finish choice, show the actual
46
46
 
47
47
  After A, B, C, or D, execute Runtime's complete command for `keep`, `merge`, `push`, or `pull-request` respectively. Stop after E. At Archive-ready:
48
48
 
49
- 1. First execute Runtime's complete `archive --dry-run` command. If an isolated workspace has no finish choice, wait for the user, then either run `comet native archive <change-name> --confirmed --finish <chosen mode>` to archive in one step (Runtime records the choice and revalidates verification freshness inside the transaction, so no second dry-run is needed) or use the matching `--dry-run --finish` command in `commandAlternatives` for a preview. Do not add other arguments yourself.
49
+ Before archiving, prepare the archive commit message under the project's active commit conventions; also prepare a merge message for local merge. Apply known format, language, and body requirements directly without asking again. Ask only for conflicting requirements or missing information the user must supply. Pass the archive message through public `--commit-message <text>` and use `--merge-message <text>` only for merge. Pass Unicode or multiline text as one complete argument. Keep default messages when neither project conventions nor the user require a custom message. Project Git hooks continue to run normally.
50
+
51
+ 1. Follow the latest continuation. If an isolated workspace has no finish choice, wait for the user, then execute the complete matching `--confirmed --finish` alternative, retaining its state-version arguments. Runtime performs preflight and transaction rechecks in one command. To preview custom messages, run `archive --dry-run --finish` with the message options above, checking both archive and merge messages for merge. Apart from the public message options, preserve Runtime's workspace, confirmation, and serial-archive arguments without adding or removing them.
50
52
  2. On `ready: false`, address only blockers in that response. Do not first query `status`, repeat Archive, or manually commit Native state and verification files.
51
- 3. Only after `ready: true`, execute the single returned `archive --confirmed` command.
53
+ 3. After `ready: true`, check that `commitMessages` match the prepared messages, then execute the single returned `archive --confirmed` command with its message contents intact. Do not repeat the same preview or reconstruct the messages.
52
54
  4. If dry-run or confirmed fails, follow only the latest structured `continuation` and `workspaceFinishResult.recoveryArgs`; do not infer commands from error text.
53
55
 
54
56
  A retains the branch and directory; do not remove that worktree during the same Archive. For other ordinary changes, offer cleanup for an archived worktree with no uncommitted changes. Do not ask again if Runtime already cleaned it. Run `git worktree remove` only after user confirmation; keep any worktree with uncommitted changes or active use.
55
57
 
56
58
  After final Supervisor delivery, Runtime automatically cleans only child and integration worktrees and their branches that are clean and unused. Uncommitted files, a current process inside a directory, or unfinished Git steps cause Runtime to preserve them and return a blocker; never force deletion.
57
59
 
58
- Commit only this change's implementation and formal artifacts; preserve other user edits. After executing Runtime's `commandArgs`, inspect `workspaceFinishResult`. If `blocked`, preserve the workspace and execute the recovery commands in `recoveryArgs`.
60
+ Commit only this change's implementation and formal artifacts; preserve other user edits. After executing Runtime's `commandArgs`, inspect `workspaceFinishResult`. If `blocked`, preserve the workspace, explain the cause, repair within existing delivery authorization, and execute `recoveryArgs` without asking the user to say "retry". Ask only when recovery needs new permissions, user information, or a delivery choice. Keep delivery pending until recovery completes.
61
+
62
+ When a project Git hook rejects a commit or merge message, the archive may already be sealed while Git finishing remains incomplete. Correct the message for the unfinished step under project conventions and pass its message option in the returned recovery command. Runtime reuses completed archive commits, Spec publication, and merges. Recovery cannot rewrite a completed commit's message; do not bypass hooks or create a replacement change.
63
+
64
+ If the user or Agent already completed Git finishing manually under valid authorization, first run `comet native doctor <change-name> --json`, then follow its returned recovery command. For local keep / merge finishing, `doctor --repair` clears records only after proving archive identity, committed contents, and the selected completion conditions. Push / pull-request still use Archive to check remote and platform results. Doctor does not commit, merge, or push. An archive directory or portable `done` state does not establish workspace delivery completion; use public status and finish results.
59
65
 
60
66
  Complete when state is `done` and authorized workspace finishing is `completed` or `kept`; otherwise follow `continuation`. At task completion, call `comet task --complete` with the original request saved at startup, workflow, change, and the same session. Do not run `printenv COMET_TASK` or inspect undeclared environment variables to guess the task.