@apralabs/apra-fleet 0.3.6 → 0.4.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (569) hide show
  1. package/README.md +245 -257
  2. package/dist/agents/backlog-groomer.md +287 -0
  3. package/dist/agents/deployer.md +52 -11
  4. package/dist/agents/doer.md +32 -7
  5. package/dist/agents/integ-test-runner.md +151 -61
  6. package/dist/agents/plan-reviewer.md +74 -17
  7. package/dist/agents/planner.md +94 -14
  8. package/dist/agents/regression-test-runner.md +216 -0
  9. package/dist/agents/reviewer.md +21 -6
  10. package/dist/agents/schemas/backlog-groomer-input.json +35 -0
  11. package/dist/agents/schemas/backlog-groomer-output.json +164 -0
  12. package/dist/agents/schemas/integ-test-runner-output.json +34 -0
  13. package/dist/agents/schemas/regression-test-runner-input.json +15 -0
  14. package/dist/agents/schemas/regression-test-runner-output.json +50 -0
  15. package/dist/agents/schemas/reviewer-output.json +4 -0
  16. package/dist/cli/agent-transform.d.ts +1 -0
  17. package/dist/cli/agent-transform.d.ts.map +1 -1
  18. package/dist/cli/agent-transform.js +77 -0
  19. package/dist/cli/agent-transform.js.map +1 -1
  20. package/dist/cli/auth.d.ts +3 -0
  21. package/dist/cli/auth.d.ts.map +1 -1
  22. package/dist/cli/auth.js +181 -5
  23. package/dist/cli/auth.js.map +1 -1
  24. package/dist/cli/config.d.ts +5 -1
  25. package/dist/cli/config.d.ts.map +1 -1
  26. package/dist/cli/config.js +7 -1
  27. package/dist/cli/config.js.map +1 -1
  28. package/dist/cli/dolt-install.d.ts +91 -0
  29. package/dist/cli/dolt-install.d.ts.map +1 -0
  30. package/dist/cli/dolt-install.js +334 -0
  31. package/dist/cli/dolt-install.js.map +1 -0
  32. package/dist/cli/install-guard.d.ts +29 -0
  33. package/dist/cli/install-guard.d.ts.map +1 -0
  34. package/dist/cli/install-guard.js +153 -0
  35. package/dist/cli/install-guard.js.map +1 -0
  36. package/dist/cli/install.d.ts +44 -3
  37. package/dist/cli/install.d.ts.map +1 -1
  38. package/dist/cli/install.js +604 -108
  39. package/dist/cli/install.js.map +1 -1
  40. package/dist/cli/join.d.ts +12 -0
  41. package/dist/cli/join.d.ts.map +1 -0
  42. package/dist/cli/join.js +87 -0
  43. package/dist/cli/join.js.map +1 -0
  44. package/dist/cli/register-member.d.ts +2 -0
  45. package/dist/cli/register-member.d.ts.map +1 -0
  46. package/dist/cli/register-member.js +193 -0
  47. package/dist/cli/register-member.js.map +1 -0
  48. package/dist/cli/restart.d.ts +2 -0
  49. package/dist/cli/restart.d.ts.map +1 -0
  50. package/dist/cli/restart.js +7 -0
  51. package/dist/cli/restart.js.map +1 -0
  52. package/dist/cli/spoke.d.ts +35 -0
  53. package/dist/cli/spoke.d.ts.map +1 -0
  54. package/dist/cli/spoke.js +212 -0
  55. package/dist/cli/spoke.js.map +1 -0
  56. package/dist/cli/start.d.ts +2 -0
  57. package/dist/cli/start.d.ts.map +1 -0
  58. package/dist/cli/start.js +92 -0
  59. package/dist/cli/start.js.map +1 -0
  60. package/dist/cli/status.d.ts +2 -0
  61. package/dist/cli/status.d.ts.map +1 -0
  62. package/dist/cli/status.js +83 -0
  63. package/dist/cli/status.js.map +1 -0
  64. package/dist/cli/stop.d.ts +2 -0
  65. package/dist/cli/stop.d.ts.map +1 -0
  66. package/dist/cli/stop.js +51 -0
  67. package/dist/cli/stop.js.map +1 -0
  68. package/dist/cli/uninstall.d.ts.map +1 -1
  69. package/dist/cli/uninstall.js +109 -9
  70. package/dist/cli/uninstall.js.map +1 -1
  71. package/dist/cli/update.d.ts.map +1 -1
  72. package/dist/cli/update.js +10 -1
  73. package/dist/cli/update.js.map +1 -1
  74. package/dist/cli/workflow-assets.d.ts +37 -0
  75. package/dist/cli/workflow-assets.d.ts.map +1 -0
  76. package/dist/cli/workflow-assets.js +144 -0
  77. package/dist/cli/workflow-assets.js.map +1 -0
  78. package/dist/cli/workflow.d.ts +75 -0
  79. package/dist/cli/workflow.d.ts.map +1 -0
  80. package/dist/cli/workflow.js +397 -0
  81. package/dist/cli/workflow.js.map +1 -0
  82. package/dist/hub-service/activity.d.ts +30 -0
  83. package/dist/hub-service/activity.d.ts.map +1 -0
  84. package/dist/hub-service/activity.js +20 -0
  85. package/dist/hub-service/activity.js.map +1 -0
  86. package/dist/hub-service/audit-log.d.ts +19 -0
  87. package/dist/hub-service/audit-log.d.ts.map +1 -0
  88. package/dist/hub-service/audit-log.js +13 -0
  89. package/dist/hub-service/audit-log.js.map +1 -0
  90. package/dist/hub-service/db/pool.d.ts +19 -0
  91. package/dist/hub-service/db/pool.d.ts.map +1 -0
  92. package/dist/hub-service/db/pool.js +53 -0
  93. package/dist/hub-service/db/pool.js.map +1 -0
  94. package/dist/hub-service/enrollment.d.ts +29 -0
  95. package/dist/hub-service/enrollment.d.ts.map +1 -0
  96. package/dist/hub-service/enrollment.js +48 -0
  97. package/dist/hub-service/enrollment.js.map +1 -0
  98. package/dist/hub-service/envelope-routes.d.ts +46 -0
  99. package/dist/hub-service/envelope-routes.d.ts.map +1 -0
  100. package/dist/hub-service/envelope-routes.js +94 -0
  101. package/dist/hub-service/envelope-routes.js.map +1 -0
  102. package/dist/hub-service/handlers/installers.d.ts +14 -0
  103. package/dist/hub-service/handlers/installers.d.ts.map +1 -0
  104. package/dist/hub-service/handlers/installers.js +8 -0
  105. package/dist/hub-service/handlers/installers.js.map +1 -0
  106. package/dist/hub-service/http-server.d.ts +31 -0
  107. package/dist/hub-service/http-server.d.ts.map +1 -0
  108. package/dist/hub-service/http-server.js +711 -0
  109. package/dist/hub-service/http-server.js.map +1 -0
  110. package/dist/hub-service/hub-jwt.d.ts +30 -0
  111. package/dist/hub-service/hub-jwt.d.ts.map +1 -0
  112. package/dist/hub-service/hub-jwt.js +105 -0
  113. package/dist/hub-service/hub-jwt.js.map +1 -0
  114. package/dist/hub-service/jwt-revocation.d.ts +20 -0
  115. package/dist/hub-service/jwt-revocation.d.ts.map +1 -0
  116. package/dist/hub-service/jwt-revocation.js +18 -0
  117. package/dist/hub-service/jwt-revocation.js.map +1 -0
  118. package/dist/hub-service/machines.d.ts +16 -0
  119. package/dist/hub-service/machines.d.ts.map +1 -0
  120. package/dist/hub-service/machines.js +14 -0
  121. package/dist/hub-service/machines.js.map +1 -0
  122. package/dist/hub-service/main.d.ts +3 -0
  123. package/dist/hub-service/main.d.ts.map +1 -0
  124. package/dist/hub-service/main.js +64 -0
  125. package/dist/hub-service/main.js.map +1 -0
  126. package/dist/hub-service/member-tokens.d.ts +11 -0
  127. package/dist/hub-service/member-tokens.d.ts.map +1 -0
  128. package/dist/hub-service/member-tokens.js +49 -0
  129. package/dist/hub-service/member-tokens.js.map +1 -0
  130. package/dist/hub-service/member-view.d.ts +20 -0
  131. package/dist/hub-service/member-view.d.ts.map +1 -0
  132. package/dist/hub-service/member-view.js +76 -0
  133. package/dist/hub-service/member-view.js.map +1 -0
  134. package/dist/hub-service/members.d.ts +44 -0
  135. package/dist/hub-service/members.d.ts.map +1 -0
  136. package/dist/hub-service/members.js +25 -0
  137. package/dist/hub-service/members.js.map +1 -0
  138. package/dist/hub-service/presence.d.ts +40 -0
  139. package/dist/hub-service/presence.d.ts.map +1 -0
  140. package/dist/hub-service/presence.js +44 -0
  141. package/dist/hub-service/presence.js.map +1 -0
  142. package/dist/hub-service/project-view.d.ts +12 -0
  143. package/dist/hub-service/project-view.d.ts.map +1 -0
  144. package/dist/hub-service/project-view.js +34 -0
  145. package/dist/hub-service/project-view.js.map +1 -0
  146. package/dist/hub-service/projects.d.ts +48 -0
  147. package/dist/hub-service/projects.d.ts.map +1 -0
  148. package/dist/hub-service/projects.js +80 -0
  149. package/dist/hub-service/projects.js.map +1 -0
  150. package/dist/hub-service/relay-queue.d.ts +115 -0
  151. package/dist/hub-service/relay-queue.d.ts.map +1 -0
  152. package/dist/hub-service/relay-queue.js +148 -0
  153. package/dist/hub-service/relay-queue.js.map +1 -0
  154. package/dist/hub-service/session-jwt.d.ts +10 -0
  155. package/dist/hub-service/session-jwt.d.ts.map +1 -0
  156. package/dist/hub-service/session-jwt.js +69 -0
  157. package/dist/hub-service/session-jwt.js.map +1 -0
  158. package/dist/hub-service/usage.d.ts +40 -0
  159. package/dist/hub-service/usage.d.ts.map +1 -0
  160. package/dist/hub-service/usage.js +25 -0
  161. package/dist/hub-service/usage.js.map +1 -0
  162. package/dist/hub-service/users.d.ts +56 -0
  163. package/dist/hub-service/users.d.ts.map +1 -0
  164. package/dist/hub-service/users.js +72 -0
  165. package/dist/hub-service/users.js.map +1 -0
  166. package/dist/hub-service/workspaces.d.ts +20 -0
  167. package/dist/hub-service/workspaces.d.ts.map +1 -0
  168. package/dist/hub-service/workspaces.js +10 -0
  169. package/dist/hub-service/workspaces.js.map +1 -0
  170. package/dist/index.js +179 -132
  171. package/dist/index.js.map +1 -1
  172. package/dist/os/linux.d.ts +13 -0
  173. package/dist/os/linux.d.ts.map +1 -1
  174. package/dist/os/linux.js +62 -3
  175. package/dist/os/linux.js.map +1 -1
  176. package/dist/os/os-commands.d.ts +10 -0
  177. package/dist/os/os-commands.d.ts.map +1 -1
  178. package/dist/os/windows.d.ts +1 -0
  179. package/dist/os/windows.d.ts.map +1 -1
  180. package/dist/os/windows.js +21 -1
  181. package/dist/os/windows.js.map +1 -1
  182. package/dist/paths.d.ts +23 -0
  183. package/dist/paths.d.ts.map +1 -1
  184. package/dist/paths.js +29 -0
  185. package/dist/paths.js.map +1 -1
  186. package/dist/providers/agy.d.ts +9 -2
  187. package/dist/providers/agy.d.ts.map +1 -1
  188. package/dist/providers/agy.js +127 -14
  189. package/dist/providers/agy.js.map +1 -1
  190. package/dist/providers/claude.d.ts +6 -2
  191. package/dist/providers/claude.d.ts.map +1 -1
  192. package/dist/providers/claude.js +248 -9
  193. package/dist/providers/claude.js.map +1 -1
  194. package/dist/providers/codex.d.ts +3 -2
  195. package/dist/providers/codex.d.ts.map +1 -1
  196. package/dist/providers/codex.js +7 -0
  197. package/dist/providers/codex.js.map +1 -1
  198. package/dist/providers/copilot.d.ts +3 -2
  199. package/dist/providers/copilot.d.ts.map +1 -1
  200. package/dist/providers/copilot.js +8 -1
  201. package/dist/providers/copilot.js.map +1 -1
  202. package/dist/providers/gemini.d.ts +3 -2
  203. package/dist/providers/gemini.d.ts.map +1 -1
  204. package/dist/providers/gemini.js +11 -1
  205. package/dist/providers/gemini.js.map +1 -1
  206. package/dist/providers/index.d.ts +1 -0
  207. package/dist/providers/index.d.ts.map +1 -1
  208. package/dist/providers/index.js +3 -0
  209. package/dist/providers/index.js.map +1 -1
  210. package/dist/providers/none.d.ts +51 -0
  211. package/dist/providers/none.d.ts.map +1 -0
  212. package/dist/providers/none.js +103 -0
  213. package/dist/providers/none.js.map +1 -0
  214. package/dist/providers/opencode.d.ts +4 -2
  215. package/dist/providers/opencode.d.ts.map +1 -1
  216. package/dist/providers/opencode.js +49 -0
  217. package/dist/providers/opencode.js.map +1 -1
  218. package/dist/providers/provider.d.ts +79 -1
  219. package/dist/providers/provider.d.ts.map +1 -1
  220. package/dist/providers/provider.js +10 -0
  221. package/dist/providers/provider.js.map +1 -1
  222. package/dist/services/agent-provisioner.d.ts +3 -3
  223. package/dist/services/agent-provisioner.d.ts.map +1 -1
  224. package/dist/services/agent-provisioner.js +12 -5
  225. package/dist/services/agent-provisioner.js.map +1 -1
  226. package/dist/services/auth-socket.d.ts.map +1 -1
  227. package/dist/services/auth-socket.js +2 -0
  228. package/dist/services/auth-socket.js.map +1 -1
  229. package/dist/services/budget-awareness.d.ts +130 -0
  230. package/dist/services/budget-awareness.d.ts.map +1 -0
  231. package/dist/services/budget-awareness.js +129 -0
  232. package/dist/services/budget-awareness.js.map +1 -0
  233. package/dist/services/cloud-sync.d.ts +35 -0
  234. package/dist/services/cloud-sync.d.ts.map +1 -0
  235. package/dist/services/cloud-sync.js +90 -0
  236. package/dist/services/cloud-sync.js.map +1 -0
  237. package/dist/services/context-admission.d.ts +82 -0
  238. package/dist/services/context-admission.d.ts.map +1 -0
  239. package/dist/services/context-admission.js +136 -0
  240. package/dist/services/context-admission.js.map +1 -0
  241. package/dist/services/event-bus.d.ts +27 -0
  242. package/dist/services/event-bus.d.ts.map +1 -0
  243. package/dist/services/event-bus.js +20 -0
  244. package/dist/services/event-bus.js.map +1 -0
  245. package/dist/services/file-chunker.d.ts +38 -0
  246. package/dist/services/file-chunker.d.ts.map +1 -0
  247. package/dist/services/file-chunker.js +63 -0
  248. package/dist/services/file-chunker.js.map +1 -0
  249. package/dist/services/file-transfer-relay.d.ts +55 -0
  250. package/dist/services/file-transfer-relay.d.ts.map +1 -0
  251. package/dist/services/file-transfer-relay.js +115 -0
  252. package/dist/services/file-transfer-relay.js.map +1 -0
  253. package/dist/services/github-app.d.ts.map +1 -1
  254. package/dist/services/github-app.js +1 -0
  255. package/dist/services/github-app.js.map +1 -1
  256. package/dist/services/http-transport.d.ts +25 -0
  257. package/dist/services/http-transport.d.ts.map +1 -0
  258. package/dist/services/http-transport.js +371 -0
  259. package/dist/services/http-transport.js.map +1 -0
  260. package/dist/services/hub-client.d.ts +69 -0
  261. package/dist/services/hub-client.d.ts.map +1 -0
  262. package/dist/services/hub-client.js +217 -0
  263. package/dist/services/hub-client.js.map +1 -0
  264. package/dist/services/jwt.d.ts +17 -0
  265. package/dist/services/jwt.d.ts.map +1 -0
  266. package/dist/services/jwt.js +70 -0
  267. package/dist/services/jwt.js.map +1 -0
  268. package/dist/services/known-sessions.d.ts +8 -0
  269. package/dist/services/known-sessions.d.ts.map +1 -0
  270. package/dist/services/known-sessions.js +40 -0
  271. package/dist/services/known-sessions.js.map +1 -0
  272. package/dist/services/model-pricing.d.ts +36 -0
  273. package/dist/services/model-pricing.d.ts.map +1 -0
  274. package/dist/services/model-pricing.js +108 -0
  275. package/dist/services/model-pricing.js.map +1 -0
  276. package/dist/services/orphan-recovery.d.ts +56 -0
  277. package/dist/services/orphan-recovery.d.ts.map +1 -0
  278. package/dist/services/orphan-recovery.js +111 -0
  279. package/dist/services/orphan-recovery.js.map +1 -0
  280. package/dist/services/pending-responses.d.ts +23 -0
  281. package/dist/services/pending-responses.d.ts.map +1 -0
  282. package/dist/services/pending-responses.js +52 -0
  283. package/dist/services/pending-responses.js.map +1 -0
  284. package/dist/services/registry.d.ts +4 -1
  285. package/dist/services/registry.d.ts.map +1 -1
  286. package/dist/services/registry.js +3 -0
  287. package/dist/services/registry.js.map +1 -1
  288. package/dist/services/relay-context.d.ts +23 -0
  289. package/dist/services/relay-context.d.ts.map +1 -0
  290. package/dist/services/relay-context.js +8 -0
  291. package/dist/services/relay-context.js.map +1 -0
  292. package/dist/services/relay-executor.d.ts +68 -0
  293. package/dist/services/relay-executor.d.ts.map +1 -0
  294. package/dist/services/relay-executor.js +70 -0
  295. package/dist/services/relay-executor.js.map +1 -0
  296. package/dist/services/relay-request.d.ts +65 -0
  297. package/dist/services/relay-request.d.ts.map +1 -0
  298. package/dist/services/relay-request.js +129 -0
  299. package/dist/services/relay-request.js.map +1 -0
  300. package/dist/services/relay-strategy.d.ts +19 -0
  301. package/dist/services/relay-strategy.d.ts.map +1 -0
  302. package/dist/services/relay-strategy.js +111 -0
  303. package/dist/services/relay-strategy.js.map +1 -0
  304. package/dist/services/service-manager/index.d.ts +5 -0
  305. package/dist/services/service-manager/index.d.ts.map +1 -0
  306. package/dist/services/service-manager/index.js +66 -0
  307. package/dist/services/service-manager/index.js.map +1 -0
  308. package/dist/services/service-manager/linux.d.ts +10 -0
  309. package/dist/services/service-manager/linux.d.ts.map +1 -0
  310. package/dist/services/service-manager/linux.js +97 -0
  311. package/dist/services/service-manager/linux.js.map +1 -0
  312. package/dist/services/service-manager/macos.d.ts +10 -0
  313. package/dist/services/service-manager/macos.d.ts.map +1 -0
  314. package/dist/services/service-manager/macos.js +94 -0
  315. package/dist/services/service-manager/macos.js.map +1 -0
  316. package/dist/services/service-manager/types.d.ts +18 -0
  317. package/dist/services/service-manager/types.d.ts.map +1 -0
  318. package/dist/services/service-manager/types.js +5 -0
  319. package/dist/services/service-manager/types.js.map +1 -0
  320. package/dist/services/service-manager/windows.d.ts +10 -0
  321. package/dist/services/service-manager/windows.d.ts.map +1 -0
  322. package/dist/services/service-manager/windows.js +73 -0
  323. package/dist/services/service-manager/windows.js.map +1 -0
  324. package/dist/services/session-registry.d.ts +70 -0
  325. package/dist/services/session-registry.d.ts.map +1 -0
  326. package/dist/services/session-registry.js +91 -0
  327. package/dist/services/session-registry.js.map +1 -0
  328. package/dist/services/singleton.d.ts +15 -0
  329. package/dist/services/singleton.d.ts.map +1 -0
  330. package/dist/services/singleton.js +98 -0
  331. package/dist/services/singleton.js.map +1 -0
  332. package/dist/services/sprint-coordination.d.ts +168 -0
  333. package/dist/services/sprint-coordination.d.ts.map +1 -0
  334. package/dist/services/sprint-coordination.js +564 -0
  335. package/dist/services/sprint-coordination.js.map +1 -0
  336. package/dist/services/ssh.d.ts.map +1 -1
  337. package/dist/services/ssh.js +70 -3
  338. package/dist/services/ssh.js.map +1 -1
  339. package/dist/services/stall/log-path-resolver.d.ts.map +1 -1
  340. package/dist/services/stall/log-path-resolver.js +8 -1
  341. package/dist/services/stall/log-path-resolver.js.map +1 -1
  342. package/dist/services/stall/stall-detector.d.ts.map +1 -1
  343. package/dist/services/stall/stall-detector.js +39 -7
  344. package/dist/services/stall/stall-detector.js.map +1 -1
  345. package/dist/services/stall/stall-poller.d.ts +16 -0
  346. package/dist/services/stall/stall-poller.d.ts.map +1 -1
  347. package/dist/services/stall/stall-poller.js +136 -18
  348. package/dist/services/stall/stall-poller.js.map +1 -1
  349. package/dist/services/strategy.d.ts.map +1 -1
  350. package/dist/services/strategy.js +55 -5
  351. package/dist/services/strategy.js.map +1 -1
  352. package/dist/services/substitution-engine.d.ts +20 -0
  353. package/dist/services/substitution-engine.d.ts.map +1 -0
  354. package/dist/services/substitution-engine.js +92 -0
  355. package/dist/services/substitution-engine.js.map +1 -0
  356. package/dist/services/task-cleanup.d.ts.map +1 -1
  357. package/dist/services/task-cleanup.js +1 -9
  358. package/dist/services/task-cleanup.js.map +1 -1
  359. package/dist/services/token-issuer.d.ts +47 -0
  360. package/dist/services/token-issuer.d.ts.map +1 -0
  361. package/dist/services/token-issuer.js +63 -0
  362. package/dist/services/token-issuer.js.map +1 -0
  363. package/dist/services/tool-registry.d.ts +3 -0
  364. package/dist/services/tool-registry.d.ts.map +1 -0
  365. package/dist/services/tool-registry.js +139 -0
  366. package/dist/services/tool-registry.js.map +1 -0
  367. package/dist/services/user-config.d.ts +14 -0
  368. package/dist/services/user-config.d.ts.map +1 -1
  369. package/dist/services/user-config.js +55 -0
  370. package/dist/services/user-config.js.map +1 -1
  371. package/dist/services/vcs/types.d.ts +44 -0
  372. package/dist/services/vcs/types.d.ts.map +1 -1
  373. package/dist/skills/fleet-sprint-cli/SKILL.md +103 -0
  374. package/dist/skills/pm/cost.md +3 -0
  375. package/dist/skills/pm/doer-reviewer-loop.md +3 -0
  376. package/dist/tools/check-status.d.ts.map +1 -1
  377. package/dist/tools/check-status.js +8 -2
  378. package/dist/tools/check-status.js.map +1 -1
  379. package/dist/tools/child-id-allocator.d.ts +45 -0
  380. package/dist/tools/child-id-allocator.d.ts.map +1 -0
  381. package/dist/tools/child-id-allocator.js +62 -0
  382. package/dist/tools/child-id-allocator.js.map +1 -0
  383. package/dist/tools/cloud-control.d.ts +2 -2
  384. package/dist/tools/compose-permissions.d.ts +12 -4
  385. package/dist/tools/compose-permissions.d.ts.map +1 -1
  386. package/dist/tools/compose-permissions.js +55 -4
  387. package/dist/tools/compose-permissions.js.map +1 -1
  388. package/dist/tools/dolt-push-mutex.d.ts +44 -0
  389. package/dist/tools/dolt-push-mutex.d.ts.map +1 -0
  390. package/dist/tools/dolt-push-mutex.js +83 -0
  391. package/dist/tools/dolt-push-mutex.js.map +1 -0
  392. package/dist/tools/execute-command.d.ts +11 -1
  393. package/dist/tools/execute-command.d.ts.map +1 -1
  394. package/dist/tools/execute-command.js +22 -5
  395. package/dist/tools/execute-command.js.map +1 -1
  396. package/dist/tools/execute-prompt.d.ts +72 -12
  397. package/dist/tools/execute-prompt.d.ts.map +1 -1
  398. package/dist/tools/execute-prompt.js +962 -71
  399. package/dist/tools/execute-prompt.js.map +1 -1
  400. package/dist/tools/get-member-model-pricing.d.ts +21 -0
  401. package/dist/tools/get-member-model-pricing.d.ts.map +1 -0
  402. package/dist/tools/get-member-model-pricing.js +29 -0
  403. package/dist/tools/get-member-model-pricing.js.map +1 -0
  404. package/dist/tools/list-members.d.ts.map +1 -1
  405. package/dist/tools/list-members.js +39 -7
  406. package/dist/tools/list-members.js.map +1 -1
  407. package/dist/tools/member-detail.d.ts +1 -1
  408. package/dist/tools/member-detail.d.ts.map +1 -1
  409. package/dist/tools/member-detail.js +8 -2
  410. package/dist/tools/member-detail.js.map +1 -1
  411. package/dist/tools/member-reservation.d.ts +29 -0
  412. package/dist/tools/member-reservation.d.ts.map +1 -0
  413. package/dist/tools/member-reservation.js +71 -0
  414. package/dist/tools/member-reservation.js.map +1 -0
  415. package/dist/tools/provision-vcs-auth.d.ts +7 -7
  416. package/dist/tools/provision-vcs-auth.js +1 -1
  417. package/dist/tools/provision-vcs-auth.js.map +1 -1
  418. package/dist/tools/register-member.d.ts +17 -5
  419. package/dist/tools/register-member.d.ts.map +1 -1
  420. package/dist/tools/register-member.js +144 -5
  421. package/dist/tools/register-member.js.map +1 -1
  422. package/dist/tools/remove-member.d.ts +1 -1
  423. package/dist/tools/report-status.d.ts +33 -0
  424. package/dist/tools/report-status.d.ts.map +1 -0
  425. package/dist/tools/report-status.js +44 -0
  426. package/dist/tools/report-status.js.map +1 -0
  427. package/dist/tools/respond-to-message.d.ts +21 -0
  428. package/dist/tools/respond-to-message.d.ts.map +1 -0
  429. package/dist/tools/respond-to-message.js +23 -0
  430. package/dist/tools/respond-to-message.js.map +1 -0
  431. package/dist/tools/send-files.d.ts +3 -0
  432. package/dist/tools/send-files.d.ts.map +1 -1
  433. package/dist/tools/send-files.js +88 -8
  434. package/dist/tools/send-files.js.map +1 -1
  435. package/dist/tools/send-message.d.ts +27 -0
  436. package/dist/tools/send-message.d.ts.map +1 -0
  437. package/dist/tools/send-message.js +46 -0
  438. package/dist/tools/send-message.js.map +1 -0
  439. package/dist/tools/shutdown-server.d.ts +2 -0
  440. package/dist/tools/shutdown-server.d.ts.map +1 -1
  441. package/dist/tools/shutdown-server.js +19 -0
  442. package/dist/tools/shutdown-server.js.map +1 -1
  443. package/dist/tools/update-member.d.ts +8 -8
  444. package/dist/tools/update-member.d.ts.map +1 -1
  445. package/dist/tools/update-member.js +11 -1
  446. package/dist/tools/update-member.js.map +1 -1
  447. package/dist/types.d.ts +15 -3
  448. package/dist/types.d.ts.map +1 -1
  449. package/dist/utils/agent-helpers.d.ts +2 -1
  450. package/dist/utils/agent-helpers.d.ts.map +1 -1
  451. package/dist/utils/agent-helpers.js +7 -2
  452. package/dist/utils/agent-helpers.js.map +1 -1
  453. package/dist/utils/pid-helpers.d.ts +15 -0
  454. package/dist/utils/pid-helpers.d.ts.map +1 -1
  455. package/dist/utils/pid-helpers.js +23 -0
  456. package/dist/utils/pid-helpers.js.map +1 -1
  457. package/dist/utils/process-utils.d.ts +3 -0
  458. package/dist/utils/process-utils.d.ts.map +1 -0
  459. package/dist/utils/process-utils.js +42 -0
  460. package/dist/utils/process-utils.js.map +1 -0
  461. package/dist/utils/prompt-errors.d.ts +2 -1
  462. package/dist/utils/prompt-errors.d.ts.map +1 -1
  463. package/dist/utils/prompt-errors.js +13 -3
  464. package/dist/utils/prompt-errors.js.map +1 -1
  465. package/dist/utils/workspace-trust.d.ts +24 -0
  466. package/dist/utils/workspace-trust.d.ts.map +1 -0
  467. package/dist/utils/workspace-trust.js +35 -0
  468. package/dist/utils/workspace-trust.js.map +1 -0
  469. package/dist/workflows/auto-sprint.js +450 -87
  470. package/package.json +35 -7
  471. package/packages/apra-fleet-client/package.json +26 -0
  472. package/packages/apra-fleet-client/src/client/api.mjs +396 -0
  473. package/packages/apra-fleet-client/src/client/client.mjs +146 -0
  474. package/packages/apra-fleet-client/src/client/errors.mjs +71 -0
  475. package/packages/apra-fleet-client/src/client/factory.mjs +65 -0
  476. package/packages/apra-fleet-client/src/client/server-resolution.mjs +265 -0
  477. package/packages/apra-fleet-client/src/client/transport.mjs +327 -0
  478. package/packages/apra-fleet-se/apra-pm/agents/backlog-groomer.md +287 -0
  479. package/packages/apra-fleet-se/apra-pm/agents/ci-watcher.md +89 -0
  480. package/packages/apra-fleet-se/apra-pm/agents/deployer.md +139 -0
  481. package/packages/apra-fleet-se/apra-pm/agents/doer.md +225 -0
  482. package/packages/apra-fleet-se/apra-pm/agents/harvester.md +169 -0
  483. package/packages/apra-fleet-se/apra-pm/agents/integ-test-runner.md +316 -0
  484. package/packages/apra-fleet-se/apra-pm/agents/plan-reviewer.md +188 -0
  485. package/packages/apra-fleet-se/apra-pm/agents/planner.md +239 -0
  486. package/packages/apra-fleet-se/apra-pm/agents/regression-test-runner.md +216 -0
  487. package/packages/apra-fleet-se/apra-pm/agents/reviewer.md +174 -0
  488. package/packages/apra-fleet-se/apra-pm/agents/schemas/backlog-groomer-input.json +35 -0
  489. package/packages/apra-fleet-se/apra-pm/agents/schemas/backlog-groomer-output.json +164 -0
  490. package/packages/apra-fleet-se/apra-pm/agents/schemas/ci-watcher-input.json +26 -0
  491. package/packages/apra-fleet-se/apra-pm/agents/schemas/ci-watcher-output.json +18 -0
  492. package/packages/apra-fleet-se/apra-pm/agents/schemas/deployer-input.json +19 -0
  493. package/packages/apra-fleet-se/apra-pm/agents/schemas/deployer-output.json +17 -0
  494. package/packages/apra-fleet-se/apra-pm/agents/schemas/doer-input.json +25 -0
  495. package/packages/apra-fleet-se/apra-pm/agents/schemas/doer-output.json +22 -0
  496. package/packages/apra-fleet-se/apra-pm/agents/schemas/harvester-input.json +31 -0
  497. package/packages/apra-fleet-se/apra-pm/agents/schemas/harvester-output.json +18 -0
  498. package/packages/apra-fleet-se/apra-pm/agents/schemas/integ-test-runner-input.json +20 -0
  499. package/packages/apra-fleet-se/apra-pm/agents/schemas/integ-test-runner-output.json +61 -0
  500. package/packages/apra-fleet-se/apra-pm/agents/schemas/plan-reviewer-input.json +15 -0
  501. package/packages/apra-fleet-se/apra-pm/agents/schemas/plan-reviewer-output.json +30 -0
  502. package/packages/apra-fleet-se/apra-pm/agents/schemas/regression-test-runner-input.json +15 -0
  503. package/packages/apra-fleet-se/apra-pm/agents/schemas/regression-test-runner-output.json +50 -0
  504. package/packages/apra-fleet-se/apra-pm/agents/schemas/reviewer-input.json +25 -0
  505. package/packages/apra-fleet-se/apra-pm/agents/schemas/reviewer-output.json +38 -0
  506. package/packages/apra-fleet-se/apra-pm/skills/pm/SKILL.md +402 -0
  507. package/packages/apra-fleet-se/apra-pm/skills/pm/beads.md +142 -0
  508. package/packages/apra-fleet-se/apra-pm/skills/pm/cost.md +157 -0
  509. package/packages/apra-fleet-se/apra-pm/skills/pm/doer-reviewer-loop.md +300 -0
  510. package/packages/apra-fleet-se/apra-pm/skills/pm/fleet-addendum.md +121 -0
  511. package/packages/apra-fleet-se/apra-pm/skills/pm/simple-sprint.md +65 -0
  512. package/packages/apra-fleet-se/apra-pm/skills/pm/sprint.md +327 -0
  513. package/packages/apra-fleet-se/apra-pm/skills/pm/worktrees.md +92 -0
  514. package/packages/apra-fleet-se/apra-pm/version.json +3 -0
  515. package/packages/apra-fleet-se/bin/cli.mjs +857 -0
  516. package/packages/apra-fleet-se/bin/serve.mjs +383 -0
  517. package/packages/apra-fleet-se/fleet-sprint/conflict-ladder.mjs +178 -0
  518. package/packages/apra-fleet-se/fleet-sprint/contracts.mjs +914 -0
  519. package/packages/apra-fleet-se/fleet-sprint/dispatch-safety-guard.mjs +177 -0
  520. package/packages/apra-fleet-se/fleet-sprint/docs/README.md +191 -0
  521. package/packages/apra-fleet-se/fleet-sprint/docs/coach-role-design.md +151 -0
  522. package/packages/apra-fleet-se/fleet-sprint/docs/dolt-tier2-runbook.md +127 -0
  523. package/packages/apra-fleet-se/fleet-sprint/docs/fleet-sprint-diagram.md +94 -0
  524. package/packages/apra-fleet-se/fleet-sprint/docs/stabilization-log.md +1125 -0
  525. package/packages/apra-fleet-se/fleet-sprint/dolt-literal-guard.mjs +72 -0
  526. package/packages/apra-fleet-se/fleet-sprint/dolt-recovery-path-b.mjs +339 -0
  527. package/packages/apra-fleet-se/fleet-sprint/dolt-recovery-tier2.mjs +360 -0
  528. package/packages/apra-fleet-se/fleet-sprint/dolt-recovery.mjs +363 -0
  529. package/packages/apra-fleet-se/fleet-sprint/dolt-sync.mjs +1244 -0
  530. package/packages/apra-fleet-se/fleet-sprint/errors.mjs +572 -0
  531. package/packages/apra-fleet-se/fleet-sprint/full-db-fetch-guard.mjs +118 -0
  532. package/packages/apra-fleet-se/fleet-sprint/runner.js +9447 -0
  533. package/packages/apra-fleet-se/fleet-sprint/skills/fleet-sprint-cli/SKILL.md +103 -0
  534. package/packages/apra-fleet-se/fleet-sprint/sprint-lock.mjs +140 -0
  535. package/packages/apra-fleet-se/fleet-sprint/vcs-module.mjs +435 -0
  536. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/azure-devops.mjs +68 -0
  537. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/bitbucket.mjs +53 -0
  538. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/dolt.mjs +146 -0
  539. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/generic-git.mjs +176 -0
  540. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/github.mjs +222 -0
  541. package/packages/apra-fleet-se/fleet-sprint/vcs-providers/index.mjs +218 -0
  542. package/packages/apra-fleet-se/fleet-sprint/viewer-extensions.mjs +882 -0
  543. package/packages/apra-fleet-se/package.json +28 -0
  544. package/packages/apra-fleet-se/workflow.json +1 -0
  545. package/packages/apra-fleet-workflow/package.json +30 -0
  546. package/packages/apra-fleet-workflow/src/viewer/command-output-cap.mjs +112 -0
  547. package/packages/apra-fleet-workflow/src/viewer/debounced-writer.mjs +153 -0
  548. package/packages/apra-fleet-workflow/src/viewer/html-utils.mjs +31 -0
  549. package/packages/apra-fleet-workflow/src/viewer/index.mjs +1325 -0
  550. package/packages/apra-fleet-workflow/src/viewer/lean-state.mjs +237 -0
  551. package/packages/apra-fleet-workflow/src/viewer/run-state-paths.mjs +130 -0
  552. package/packages/apra-fleet-workflow/src/viewer/sprint-state-paths.mjs +29 -0
  553. package/packages/apra-fleet-workflow/src/workflow/engine.mjs +200 -0
  554. package/packages/apra-fleet-workflow/src/workflow/errors.mjs +146 -0
  555. package/packages/apra-fleet-workflow/src/workflow/index.mjs +1723 -0
  556. package/packages/apra-fleet-workflow/src/workflow/journal.mjs +291 -0
  557. package/packages/apra-fleet-workflow/src/workflow/pricing.mjs +99 -0
  558. package/packages/apra-fleet-workflow/src/workflow/vetting.mjs +111 -0
  559. package/scripts/agy-transcript-reader.js +1 -0
  560. package/scripts/fleet-statusline.sh +17 -13
  561. package/skills/fleet/SKILL.md +37 -2
  562. package/skills/fleet/beads.md +14 -7
  563. package/skills/fleet/permissions.md +50 -0
  564. package/skills/fleet/profiles/base-dev.json +1 -0
  565. package/skills/fleet/profiles/base-reviewer.json +1 -0
  566. package/skills/fleet/profiles/node.json +2 -2
  567. package/version.json +1 -1
  568. package/scripts/agy-settings-merge.js +0 -27
  569. /package/{dist → packages/apra-fleet-se/apra-pm}/agents/_shared/GRAPH-SEMANTICS.md +0 -0
@@ -0,0 +1,157 @@
1
+ # Cost Quoting and Calibration
2
+
3
+ pm produces sprint cost quotes and post-sprint analyses using the same pure
4
+ JavaScript functions as the auto-sprint workflow. No arithmetic is done in prose --
5
+ all numbers come from deterministic JS. Node.js (v18+) is required; the orchestrator
6
+ checks for it at setup and warns the user if it is absent (cost features are skipped,
7
+ everything else runs normally).
8
+
9
+ ## Source of the functions
10
+
11
+ `install.mjs` extracts the pure functions from `auto-sprint.js` at install time
12
+ and writes them as a self-contained CommonJS module to `cost.js` in the same
13
+ directory as this file (`<configDir>/skills/pm/cost.js`). It is refreshed
14
+ automatically on every install run -- no `--force` needed. On Claude Code, the
15
+ full `auto-sprint.js` is additionally copied to `~/.claude/workflows/auto-sprint.js`
16
+ so the `/auto-sprint` workflow works natively.
17
+
18
+ The orchestrator loads the functions with a plain `require()` using `__SKILL_DIR__`
19
+ (the absolute path to the skill directory, available in the orchestrator context):
20
+
21
+ ```bash
22
+ node -e "
23
+ const { computeSprintQuote } = require('__SKILL_DIR__/cost.js');
24
+ const fs = require('fs');
25
+ const calib = JSON.parse(fs.readFileSync('sprint-logs/calibration.json', 'utf8'));
26
+ const ta = JSON.parse(process.env.TASK_ASSIGNMENTS);
27
+ process.stdout.write(JSON.stringify(computeSprintQuote(ta, calib)));
28
+ "
29
+ ```
30
+
31
+ Replace `__SKILL_DIR__` with the absolute path for your provider:
32
+
33
+ | Provider | `__SKILL_DIR__` |
34
+ |----------|----------------|
35
+ | Claude | `~/.claude/skills/pm` |
36
+ | Gemini | `~/.gemini/skills/pm` |
37
+ | AGY | `~/.gemini/antigravity-cli/skills/pm` |
38
+ | OpenCode | `~/.config/opencode/skills/pm` |
39
+
40
+ Pass inputs via `process.env` (for JSON blobs) or inline literals. Capture stdout
41
+ and parse as JSON.
42
+
43
+ ## Calibration file
44
+
45
+ `sprint-logs/calibration.json` is the single source of truth for pricing, bucket
46
+ sizes, cycle assumptions, and historical averages. It is shared between auto-sprint
47
+ and pm -- they update the same file so historical data accumulates regardless of
48
+ which harness ran the sprint.
49
+
50
+ Keys in `model_prices_per_1m_output_tokens` and `role_models` are **tier names**
51
+ (`cheap`, `standard`, `premium`), not provider-specific model IDs. The mapping from
52
+ tier name to actual model ID lives exclusively in `TIER_TO_MODEL` inside
53
+ `auto-sprint.js` -- the single place to update when models change. This makes
54
+ `calibration.json` provider-agnostic and reusable across Claude, AGY, OpenCode, etc.
55
+
56
+ - **On first run** (file absent): the setup step bootstraps it from `DEFAULT_CALIBRATION`
57
+ inside `cost.js` (same object the workflow uses). Write it with:
58
+ ```bash
59
+ node -e "
60
+ const { DEFAULT_CALIBRATION } = require('__SKILL_DIR__/cost.js');
61
+ const fs = require('fs');
62
+ fs.mkdirSync('sprint-logs', { recursive: true });
63
+ fs.writeFileSync('sprint-logs/calibration.json', JSON.stringify(DEFAULT_CALIBRATION, null, 2));
64
+ "
65
+ ```
66
+ Replace `__SKILL_DIR__` with the absolute path to the pm skill directory.
67
+ - **On subsequent runs**: read it, deep-merge with `DEFAULT_CALIBRATION` (so new
68
+ fields added to the source always have a default), and use the result.
69
+
70
+ ## Setup check (once per sprint)
71
+
72
+ ```bash
73
+ node --version 2>/dev/null || echo MISSING
74
+ ```
75
+
76
+ If `MISSING`: warn the user that cost quoting is unavailable, skip all cost steps,
77
+ continue the sprint normally. If Node.js is present, proceed with calibration load /
78
+ bootstrap as above.
79
+
80
+ ## Phase 3 -- quote after plan APPROVED
81
+
82
+ After the plan-reviewer returns APPROVED with `taskAssignments` (id + bucket + model
83
+ per task -- see `doer-reviewer-loop.md` plan-reviewer template), the orchestrator:
84
+
85
+ 1. Writes `taskAssignments` to a temp env var or file.
86
+ 2. Runs `computeSprintQuote(taskAssignments, calibration)` via the node snippet above.
87
+ 3. Logs the quote (opt/expected/pessimistic scenarios, calibration source).
88
+ 4. Writes per-task cost estimates back to beads notes:
89
+ ```bash
90
+ bd update <id> --notes="cost-estimate: bucket=<B> model=<M> doer_tokens=<N> reviewer_tokens=<N> output_usd=<N>"
91
+ ```
92
+ Dispatch a single cheap-tier agent to run all `bd update` commands in one pass.
93
+
94
+ The `taskAssignments` array shape (matches the plan-reviewer role schema,
95
+ `agents/schemas/plan-reviewer-output.json`).
96
+ The `model` field is the **tier name** (`cheap`/`standard`/`premium`) read from the
97
+ task's assigned model tier in its beads metadata (`--metadata '{"model": "<tier>"}'`)
98
+ -- never a provider-specific model ID:
99
+ ```json
100
+ [{ "id": "BD-10", "bucket": "M", "model": "standard" }, ...]
101
+ ```
102
+
103
+ ## Harvest -- analysis and calibration update
104
+
105
+ At sprint close, after the harvester runs:
106
+
107
+ 1. **Compute analysis**: run `computeSprintAnalysis(quote, logEntries, calibration,
108
+ actualCycles)`. `logEntries` is the sprint log JSONL (see Sprint log below) read
109
+ and parsed from `sprint-logs/<branch>-<startedAt>.jsonl`. If the log file does not
110
+ exist (pm ran without logging), pass `[]` for logEntries -- the analysis degrades
111
+ gracefully to estimates-only.
112
+ 2. **Build summary**: run `buildSprintSummary(analysis, quote, calibration, opts)`
113
+ where `opts = { branch, goal, goalMet, cycleCount, tasksCompleted, tasksOpen,
114
+ startedAt }`. Write the returned `summaryText` to
115
+ `sprint-logs/<branch>-<startedAt>.analysis.md` and commit it.
116
+ 3. **Update calibration**: run `computeUpdatedCalibration(calibration, analysis,
117
+ startedAt, taskAssignments, logEntries)` and write the result back to
118
+ `sprint-logs/calibration.json`. Commit it. This blends the actual token counts
119
+ into the historical averages so future quotes improve.
120
+ 4. **Update fleet token memory**: Parse the `tuple_averages` from the updated
121
+ calibration object into a flat array: `const tokenEstimates = { averages: Object.values(updatedCalibration.historical?.tuple_averages || {}) }`. Then write it to the beads persistent memory:
122
+ `bd remember '<json_string>' --key token-estimates-json`
123
+
124
+ ## Sprint log
125
+
126
+ pm appends one JSONL record per dispatch to `sprint-logs/<branch>-<startedAt>.jsonl`
127
+ so `computeSprintAnalysis` can compare estimates to actuals. Each record:
128
+
129
+ ```json
130
+ { "ts": "<ISO timestamp>", "cycle": 1, "phase": "Develop", "label": "doer-c1-i1",
131
+ "model": "<tier: cheap|standard|premium>", "context": "tasks BD-10, BD-11",
132
+ "outTokens": 1234, "costUsd": 0.0185 }
133
+ ```
134
+
135
+ The orchestrator appends after each dispatch using a cheap-tier agent (same
136
+ `appendNewEntries` pattern as auto-sprint). The `label` format `<role>-c<N>-i<M>`
137
+ is required -- `computeSprintAnalysis` strips the suffix to recover the role name.
138
+
139
+ If the harness/CLI does not expose per-subagent token counts, record
140
+ `outTokens: 0` and `costUsd: 0` -- the analysis will show actuals as zero but the
141
+ estimate side still works and calibration is not corrupted.
142
+
143
+ ## Node.js invocation pattern for all functions
144
+
145
+ All five functions follow the same extract-and-call pattern. Reference:
146
+
147
+ | Function | When | Inputs | Output |
148
+ |---|---|---|---|
149
+ | `computeSprintQuote` | After plan APPROVED | `taskAssignments`, `calibration` | quote object with scenarios |
150
+ | `computeSprintAnalysis` | Harvest | `quote`, `logEntries`, `calibration`, `actualCycles` | analysisText + byRole |
151
+ | `buildSprintSummary` | Harvest | `analysis`, `quote`, `calibration`, `opts` | summaryText markdown |
152
+ | `computeUpdatedCalibration` | Harvest | `calibration`, `analysis`, `startedAt`, `taskAssignments`, `logEntries` | updated calibration object |
153
+ | `DEFAULT_CALIBRATION` | Setup bootstrap | -- | JSON to write to calibration.json |
154
+
155
+ Never hardcode prices, bucket sizes, or cycle assumptions -- always read them from
156
+ `calibration.json` (or `DEFAULT_CALIBRATION` on first bootstrap). The file is the
157
+ single source of truth shared across all sprint harnesses.
@@ -0,0 +1,300 @@
1
+ # Doer-Reviewer Loop
2
+
3
+ The execute-phase loop for one track. Every dispatch is an inline subagent call (its
4
+ result returns in the same turn). Beads is the task store; the reviewer's verdict is
5
+ returned as structured output directly to the orchestrator. The orchestrator finds
6
+ ready work with `bd ready` and hands the doer explicit bead ids; the doer claims each
7
+ with `bd update --claim` and closes it with `bd close`. The
8
+ reviewer reads acceptance criteria with `bd show` and returns its verdict
9
+ (`verdict`/`notes`/`reopenIds`/`newTasks`) as structured output -- it never touches
10
+ beads. On CHANGES NEEDED the orchestrator reads that structured output and runs the
11
+ beads updates itself. There is no PLAN.md and no progress.json -- code is committed to
12
+ the branch; all task state lives in beads. (The plan loop that precedes this -- planner /
13
+ plan-reviewer -- is in `sprint.md`.)
14
+
15
+ ## Pre-flight checks
16
+
17
+ ### Before any dispatch
18
+
19
+ Verify the agent's worktree is on the correct branch with a clean working tree:
20
+
21
+ 1. `git -C <worktree> status --porcelain` -- must be empty.
22
+ 2. `git -C <worktree> branch --show-current` -- must match the track's branch.
23
+
24
+ Do not dispatch into a worktree on the wrong branch or with uncommitted changes.
25
+
26
+ ### Before review dispatch (SHA matching)
27
+
28
+ Verify the reviewer sees the same state the doer pushed:
29
+
30
+ 1. `git -C <reviewer-worktree> rev-parse HEAD` -- record the SHA.
31
+ 2. Compare with the doer's last pushed HEAD on the track branch.
32
+ 3. If SHA does not match: `git -C <reviewer-worktree> fetch origin &&
33
+ git -C <reviewer-worktree> reset --hard origin/<branch>`, then re-verify.
34
+
35
+ Never dispatch a reviewer against stale code -- SHA mismatch means the review
36
+ covers the wrong diff.
37
+
38
+ ## The loop (the Develop phase of one cycle)
39
+
40
+ ```
41
+ Plan APPROVED, beads tasks created (each with acceptance criteria + model tier in
42
+ metadata). goal priority = the sprint's exit threshold.
43
+ loop:
44
+ 1. Find ready tasks: bd ready (tasks with no open blockers, in the sprint-root subtree).
45
+ If none AND bd list --status=open at the goal priority is empty AND the last
46
+ reviewer verdict was APPROVED -> Develop phase done; exit to Test (or Harvest
47
+ on the final cycle).
48
+ 2. Group ready tasks into model streaks (consecutive tasks sharing a model tier,
49
+ in priority/dependency order). For each streak the doer, on dispatch:
50
+ - reads each task with `bd show <id>` (description, acceptance, model tier),
51
+ - claims it with `bd update <id> --claim`,
52
+ - implements the work and commits,
53
+ - closes it with `bd close <id>`.
54
+ The doer works its streak in order and STOPS at the VERIFY checkpoint.
55
+ 3. On doer completion: assert the worktree is clean (`git status --porcelain`
56
+ empty); a dirty tree means the doer left uncommitted work -- send it back to
57
+ commit before any review. If a task is blocked, the doer leaves it open with a
58
+ blocker note -- handle the blocker (see Safeguards). The doer has already
59
+ closed the tasks it finished; the orchestrator does not touch beads here.
60
+ 4. Dispatch ONE reviewer (inline) over all tasks worked this iteration. It reads
61
+ each task's acceptance criteria with `bd show <id>` and the diff. The reviewer
62
+ runs standard-tier, escalated to premium-tier if any streak this iteration ran
63
+ premium-tier. It returns its verdict as structured output only (see reviewer
64
+ template) -- it never writes a file. Read the returned verdict:
65
+ APPROVED -> tasks stay closed. `reopenIds` and `newTasks` absent or
66
+ empty -- no beads action needed.
67
+ CHANGES NEEDED -> the orchestrator parses `reopenIds` and `newTasks` from
68
+ the reviewer's structured output (see "Orchestrator reads
69
+ the reviewer's output" below) and runs the beads updates.
70
+ Dispatch the doer to fix.
71
+ -> back to step 1.
72
+ ```
73
+
74
+ **Exit condition.** The Develop phase exits when `bd ready` returns nothing, no open
75
+ task at the goal priority remains, **and** the last reviewer verdict was APPROVED --
76
+ a single APPROVED is sufficient because beads tracks discrete tasks, not a
77
+ convergence streak. Any CHANGES NEEDED reopens tasks, so the loop continues until
78
+ they are cleared and a clean APPROVED lands on an empty ready queue. Never advance to
79
+ Test or Harvest with open tasks at the goal priority.
80
+
81
+ The orchestrator does nothing between a doer finishing and a reviewer starting except
82
+ read state from beads + git and dispatch. Never pause for the user mid-loop.
83
+
84
+ ## Model per task
85
+
86
+ The planner assigned each task a model **tier** -- cheap-tier for mechanical tasks,
87
+ standard-tier for standard implementation, premium-tier for hard design -- and wrote it
88
+ into the task's beads metadata (`--metadata '{"model": "<tier>"}'`). Read it with `bd show <id>`
89
+ at dispatch time and dispatch the doer on that tier. An iteration may span tiers
90
+ across its ready tasks; run one doer dispatch per model streak (a run of consecutive
91
+ tasks sharing a tier), in priority/dependency order. An iteration with three tier
92
+ groups becomes up to three doer dispatches, each on its own tier. The dispatch whose
93
+ streak reaches the VERIFY task carries through it.
94
+
95
+ The reviewer runs standard-tier, escalated to premium-tier whenever any streak in the
96
+ iteration ran premium-tier -- review must never be weaker than the work it judges. Cheap
97
+ bounded state checks (`bd ready` queries, feedback commits, log appends) run
98
+ cheap-tier.
99
+
100
+ ## Dispatch and wait inline
101
+
102
+ Dispatch each subagent and wait for its result in the same turn -- the dispatch
103
+ returns the agent's result to you inline. Run the loop sequentially this way: one
104
+ dispatch, its result, the next. NEVER end your turn while a dispatched agent still
105
+ owes you a result; in a headless run nothing re-invokes you, so ending the turn
106
+ parks the sprint. With multiple tracks, dispatch one per active track and poll them
107
+ to completion within the turn -- track A's reviewer can run while track B's doer
108
+ works -- but keep the turn alive until you have collected their results.
109
+
110
+ After each agent returns, the orchestrator's FIRST action is to read state from beads
111
+ and git (`bd ready`, `bd list --status=open`, `bd show <id>`, the reviewer's returned
112
+ structured output, `git log <base>..<branch>`); those are the source of truth for what
113
+ was dispatched and where it landed.
114
+
115
+ ## Commit identity
116
+
117
+ So the branch history shows who did what, each role commits under its OWN git
118
+ identity rather than the ambient one. Pass it inline on every commit -- do not rely
119
+ on global config:
120
+
121
+ ```
122
+ git -c user.name='pm-<role>' -c user.email='<role>@pm.local' commit -m "<msg>"
123
+ ```
124
+
125
+ Identities: `pm-planner`, `pm-plan-reviewer`, `pm-doer`,
126
+ `pm-reviewer`. The orchestrator's own git plumbing (requirements.md, design.md,
127
+ and the completion scaffolding drop) commits as `pm`. Each template below restates
128
+ its identity so the dispatched agent uses it.
129
+
130
+ ## Per-role prompt templates
131
+
132
+ Everything the agent needs is in the prompt plus the committed files in its
133
+ worktree. Every prompt pins the worktree path. `<transport line>` is decided once
134
+ per repo (see `worktrees.md` Transport): remote present => "Push your commits." ;
135
+ local-only => "This is a local-only worktree -- commit every turn and skip pushing."
136
+
137
+ ### planner
138
+
139
+ ```
140
+ You are planning a track. Your worktree is <abs worktree path> on branch <branch>
141
+ (base <base>). cd there first; use absolute paths. Read requirements.md (and
142
+ design.md if present). Follow your planner instructions: explore, draft, front-load
143
+ foundations, self-critique, refine. Then write the plan into beads under sprint root
144
+ <sprint-id> -- do NOT write PLAN.md. For each task: bd create "<title>" -p <priority>
145
+ --parent <sprint-id> --assignee <track> --acceptance="<what must be true to be done>"
146
+ --metadata '{"model": "<tier>"}', where <tier> is cheap-tier for mechanical work, standard-tier
147
+ for standard implementation, premium-tier for hard design (pick from the tiers available
148
+ in this environment). Wire dependencies with bd dep add <task> <blocker>. Verify
149
+ with bd ready that leaf tasks (not features/sprint roots) are unblocked. <transport line>.
150
+ The worktree and branch already exist -- do not create or switch branches.
151
+ ```
152
+
153
+ ### plan-reviewer
154
+
155
+ ```
156
+ You are reviewing a plan. Your worktree is <abs worktree path> on branch <branch>.
157
+ cd there; use absolute paths. Read requirements.md and design.md (if present), then
158
+ inspect the beads DAG under sprint root <sprint-id>: bd graph --compact <sprint-id>, bd ready,
159
+ and bd show <id> on each task to read its acceptance criteria and model-tier metadata.
160
+ Follow your plan-reviewer instructions: check coverage, task size, acceptance
161
+ criteria, dependency direction, and model-tier assignment. Classify each task with a
162
+ complexity bucket (S = 1 file/narrow scope, M = 2-3 files/moderate logic,
163
+ L = 3+ files/non-trivial design) and read its assigned model tier from its metadata.
164
+
165
+ Return your structured output ONLY -- do not write a file:
166
+ - `verdict`: APPROVED or CHANGES_NEEDED (exact strings; the machine value is the
167
+ underscore form from agents/schemas/plan-reviewer-output.json)
168
+ - `notes`: specific findings with issue IDs and exact bd commands to fix any
169
+ dependency direction problems
170
+ - `taskAssignments`: one entry per task in JSON array form:
171
+ [{"id":"<beads-id>","bucket":"S|M|L","model":"<tier: cheap|standard|premium>"},...]
172
+
173
+ You never write feedback.md or mutate beads -- the orchestrator reads your structured
174
+ output. <transport line>.
175
+ ```
176
+
177
+ The orchestrator reads `taskAssignments` from the plan-reviewer's structured output
178
+ after APPROVED to run `computeSprintQuote` (see `cost.md`). The array shape must match
179
+ exactly: `[{ "id": "<beads task id>", "bucket": "S|M|L", "model": "<tier: cheap|standard|premium>" }]`.
180
+
181
+ ### doer
182
+
183
+ ```
184
+ You are executing tasks. Your worktree is <abs worktree path> on branch <branch>
185
+ (base <base>). cd there; use absolute paths. Work ONLY task(s) <task scope>, in
186
+ order. For each task: bd show <id> to read its description and acceptance criteria;
187
+ bd update <id> --claim to claim it; implement the work; run fast tests; commit as
188
+ identity pm-doer (git -c user.name='pm-doer' -c user.email='doer@pm.local' commit);
189
+ then bd close <id>. Never close a type=feature or type=bug issue -- only type=task.
190
+ (Verify-set beads -- a parent of ANY issue_type whose children are all closed while
191
+ it is still open -- are closed later by integ-test-runner, not by you; this is not a
192
+ feature-only path.)
193
+ If your scope reaches the VERIFY checkpoint, run it -- build, linter, and full test
194
+ suite -- then stop. Otherwise stop after the last task in <task scope>. If a task is
195
+ blocked, leave it open, add a blocker note (bd update <id> --notes="blocked: ..."),
196
+ and stop. <transport line>. Do not start anything beyond <task scope>. The worktree
197
+ and branch already exist -- do not create branches.
198
+ [If fixing review findings:] the reopened task(s) carry review notes (bd show <id>).
199
+ Address every finding, commit, bd close <id> again, then stop for re-review.
200
+ ```
201
+
202
+ ### reviewer
203
+
204
+ ```
205
+ You are reviewing code. Your worktree is <abs worktree path> on branch <branch>
206
+ (base <base>). cd there; use absolute paths. Review ONLY the tasks worked this
207
+ iteration: <task ids>. Run bd show <id> on each to read its acceptance criteria, and
208
+ git diff <base>...<branch> to see the changes. Read requirements.md and design.md (if
209
+ present). Run the build, linter, and full test suite. Read the prior review history
210
+ (git log <base>..<branch>) so you account for how earlier findings were addressed.
211
+ Judge each task against its acceptance criteria.
212
+
213
+ Do NOT run any bd commands. Return your structured output ONLY -- do not write a file:
214
+
215
+ verdict: "APPROVED" or "CHANGES_NEEDED"
216
+ notes: <human-readable notes: one finding per task, referencing the task ID and what
217
+ acceptance criterion was not met>
218
+ reopenIds: ["<id1>", "<id2>"]
219
+ newTasks: [{"title": "fix: <finding>", "description": "<detail>", "priority": "P2"}]
220
+ (newTasks entries use {title, description, priority-as-string} -- the exact shape in
221
+ agents/schemas/reviewer-output.json)
222
+
223
+ On APPROVED: reopenIds and newTasks are both empty arrays.
224
+ On CHANGES NEEDED: reopenIds lists every task that failed its acceptance criteria;
225
+ newTasks lists out-of-scope findings that need a new tracked task (may be []).
226
+ Both fields must be present and contain valid JSON arrays when verdict is CHANGES NEEDED.
227
+
228
+ You never write feedback.md or mutate beads -- the orchestrator reads your structured
229
+ output and applies the reopen/create transitions. <transport line>.
230
+ ```
231
+
232
+ ### Orchestrator reads the reviewer's output
233
+
234
+ After the reviewer returns, the orchestrator reads its structured output and runs all
235
+ beads updates itself -- the reviewer is a pure reader of beads:
236
+
237
+ 1. Read the `verdict` field (`APPROVED` or `CHANGES_NEEDED`).
238
+ 2. On CHANGES NEEDED:
239
+ - Read `reopenIds`. For each id: `bd update <id> --status=open --notes="<relevant finding from notes>"`.
240
+ Reopened tasks return to `bd ready` next iteration.
241
+ - Read `newTasks`. For each entry:
242
+ `bd create "<title>" -p <priority> --parent <sprint-id> --assignee <track> --description="<description>"`.
243
+ Empty array `[]` means no new tasks to create.
244
+ 3. On APPROVED: no beads action needed.
245
+
246
+ ## Continuity between dispatches
247
+
248
+ Each dispatch is a fresh subagent run; continuity comes from the files it reads.
249
+ Two ways to continue:
250
+
251
+ - **Default: fresh dispatch.** Dispatch a new agent with the right tags (tags: ['doer'] / tags: ['reviewer']); it
252
+ reconstructs context from beads (`bd ready`, `bd show <id>`) and `git log`. This is
253
+ robust and the right choice across phase boundaries and tag switches -- the agents
254
+ begin with a context-recovery `git log` and `bd` query.
255
+ - **Tight iteration: continue the same agent.** For a quick same-worktree, same-tag
256
+ turn -- e.g. the doer addressing review findings it just produced context for --
257
+ continue the SAME agent instance, which keeps its context. Use this within one
258
+ worktree and one tag; switch tags with a fresh dispatch.
259
+
260
+ ## Resume rules
261
+
262
+ Resume is data-driven from beads task state, not manually reasoned. A fresh dispatch
263
+ re-derives everything from `bd ready` / `bd show <id>` and `git log`; resume keeps an
264
+ agent's in-session context for a tight follow-up turn.
265
+
266
+ ### Doer dispatches
267
+
268
+ | Condition | resume |
269
+ |-----------|--------|
270
+ | Next ready task is the same kind of work the agent just finished | `true` (continue session) |
271
+ | Next ready task is a context switch (different feature/area) | `false` (fresh) |
272
+ | After reviewer CHANGES NEEDED -> same doer fixes the reopened task it just built | `true` |
273
+ | Role switch (doer -> reviewer) | `false` |
274
+
275
+ ### All dispatches
276
+
277
+ | Dispatch | resume |
278
+ |----------|--------|
279
+ | Initial plan generation | `false` |
280
+ | Plan revision (any feedback iteration) | `true` |
281
+ | Initial review dispatch | `false` |
282
+ | Re-review after CHANGES NEEDED + doer fixes | `true` |
283
+ | Tag switch (tags: ['doer'] -> tags: ['reviewer'], or vice versa) | `false` |
284
+ | After stop_prompt cancellation (fleet mode) | `false` -- session state unreliable after kill; start fresh |
285
+ | After session timed out mid-grant (fleet mode) | `true` -- fleet auto-recovers but member restarts without prior context |
286
+
287
+ A tag switch always requires sending the new context (context file or inline
288
+ prompt). Never resume across a tag switch.
289
+
290
+ ## Safeguards
291
+
292
+ | Safeguard | Trigger | Orchestrator action | Limit |
293
+ |----------------------|----------------------------------|------------------------------------------------------------------|------------------|
294
+ | Dispatch retry | Agent errors or returns nothing | Re-dispatch the same tag fresh; after 3 fails, pause + flag user | 3 per dispatch |
295
+ | Doer-reviewer cycle | Reviewer lists task in `reopenIds`; orchestrator reopens it | Doer fixes, re-review; if 3 cycles leave it open, pause + flag | 3 cycles/task |
296
+ | Zero progress | Two fresh dispatches, no commits | Escalate to a stronger model; still stuck after the strongest => flag | 2 per model |
297
+ | Blocked task | Doer left a task open with a `blocked:` note | Read the blocker note (bd show <id>); resolve if mechanical, else flag user | -- |
298
+
299
+ Escalate to the user only after a safeguard limit trips, on a genuine requirements
300
+ ambiguity, or on a risky/irreversible action. Otherwise keep the loop running.
@@ -0,0 +1,121 @@
1
+ # Fleet Execution Addendum
2
+
3
+ This document covers features that apply ONLY when the pm skill runs in fleet
4
+ mode (dispatching to remote fleet members via execute_prompt). Fleet mode is an
5
+ integration with the separate apra-fleet product; apra-pm never requires it --
6
+ local subagent mode is the complete feature set, and the tools named here
7
+ (execute_prompt, compose_permissions, list_members, ...) exist only when
8
+ apra-fleet is installed. In local subagent
9
+ mode, everything here is skipped -- local subagents inherit the orchestrator's
10
+ permissions and context.
11
+
12
+ ## Mode detection
13
+
14
+ At sprint start, detect the execution mode:
15
+
16
+ 1. Probe for fleet MCP tools (fleet_status, list_members). If the tool is absent
17
+ or returns no members, fall to local mode.
18
+ 2. An explicit --local or --fleet flag from the user overrides the probe.
19
+ 3. Default: if fleet members are available AND the task tier has a matching
20
+ member, use fleet mode. Otherwise use local mode.
21
+
22
+ Record the mode in status.md at sprint init so recovery/resume uses the same
23
+ mode.
24
+
25
+ ## Permissions
26
+
27
+ Compose and deliver permissions per the fleet skill permissions.md for each
28
+ member using tag-based selection: tags: ['doer'] for doer members, tags: ['reviewer']
29
+ for reviewer members. Recompose when switching tags (doer to reviewer or vice versa).
30
+ Each provider gets its native permission config -- compose_permissions handles the
31
+ format automatically.
32
+
33
+ Call compose_permissions before EVERY dispatch regardless of unattended mode.
34
+
35
+ ### Unattended modes
36
+
37
+ - `update_member(unattended='auto')` -- safer auto-approval scoped to
38
+ explicitly listed operations.
39
+ - `update_member(unattended='dangerous')` -- full permission bypass. Prefer
40
+ auto over dangerous.
41
+ - Do NOT pass dangerously_skip_permissions to execute_prompt -- it is deprecated
42
+ and ignored.
43
+
44
+ ### Mid-sprint permission denial
45
+
46
+ If a member is blocked by a permission denial:
47
+
48
+ 1. Call compose_permissions with `grant: [<denied permission>]` and
49
+ project_folder.
50
+ 2. This grants the missing permission, delivers the updated config, and appends
51
+ to the ledger so future phases start with it included.
52
+ 3. Resume the member with resume=true.
53
+ 4. Never bypass by running the denied command yourself via execute_command.
54
+
55
+ Act on the grant promptly -- the inactivity timer fires on stdout silence. If
56
+ it fires while you are composing permissions, resume=true still succeeds via
57
+ stale-session auto-recovery, but the member restarts without its in-progress
58
+ context.
59
+
60
+ ## stop_prompt
61
+
62
+ Use stop_prompt (a fleet MCP tool) when a member is working on the wrong thing,
63
+ stuck in a loop, or dispatched with incorrect instructions. Always follow
64
+ immediately with resume=false to start a clean session.
65
+
66
+ stop_prompt kills the member's LLM process. This is distinct from stopping a
67
+ background orchestration sub-task within the PM's own session.
68
+
69
+ ## Doer-reviewer pairing (fleet mode)
70
+
71
+ 1. Record pair in status.md. Multiple pairs per project is normal.
72
+ 2. Override icons via update_member -- doer gets circle, reviewer gets square,
73
+ same color.
74
+ 3. Compose and deliver permissions for each member using tag-based selection:
75
+ tags: ['doer'] for the doer, tags: ['reviewer'] for the reviewer.
76
+ 4. Send the role-specific agent context file via send_files before dispatch.
77
+
78
+ ## Agent context file delivery
79
+
80
+ Each fleet member needs a provider-specific agent context file in their
81
+ work_folder root. It is the member's persistent execution model and survives
82
+ across session resumes.
83
+
84
+ ### Provider filename table
85
+
86
+ Use member_detail -> llmProvider to determine the target filename:
87
+
88
+ | Provider | Filename |
89
+ |----------|----------|
90
+ | Claude | CLAUDE.md |
91
+ | Antigravity (agy) | AGY.md |
92
+ | Gemini | GEMINI.md |
93
+ | Codex | AGENTS.md |
94
+ | Copilot | COPILOT.md |
95
+ | OpenCode | OPENCODE.md |
96
+
97
+ ### Rules
98
+
99
+ - Pick the correct template based on role and filename based on provider.
100
+ - Fill in {{branch}} and {{base_branch}} with the sprint branch and base branch
101
+ before delivering.
102
+ - Send to member via send_files to the member's work_folder root before dispatch.
103
+ - Never commit to git -- add the filename to the member's .gitignore.
104
+ - On role switch: send the new context file before dispatch.
105
+ - Remove before merge: use the cleanup procedure in sprint.md Completion.
106
+
107
+ ## Secrets and credentials
108
+
109
+ Never pass raw secrets in execute_prompt prompts -- reference the credential by
110
+ name only (e.g. "authenticate using credential github_pat"). The member then
111
+ uses {{secure.github_pat}} in its own execute_command calls. The
112
+ {{secure.NAME}} token is substituted at launch time -- reference it bare in
113
+ execute_command, not inside your own quotes (it is already single-quote-escaped
114
+ by the fleet).
115
+
116
+ ## Fleet call batching (R8)
117
+
118
+ When executing a sequence of fleet calls -- any combination of send_files,
119
+ execute_command, execute_prompt, receive_files -- club them into a single
120
+ background Agent rather than issuing individual calls or multiple background
121
+ agents. This reduces orchestrator turn overhead.
@@ -0,0 +1,65 @@
1
+ # Simple Sprint
2
+
3
+ A lightweight flow for small, single-session tasks. Use this when the work is
4
+ small enough that the full plan/plan-review loop is unnecessary overhead. State
5
+ still lives in beads + git -- there is no PLAN.md or progress.json in any sprint.
6
+
7
+ ## When to use
8
+
9
+ - 1-3 tasks, completable in a single session
10
+ - No complex phasing or cross-phase dependencies
11
+ - Low risk, well-understood scope
12
+
13
+ Use the full sprint lifecycle (sprint.md) for anything larger. If the work turns
14
+ out bigger than expected, promote to a full sprint.
15
+
16
+ ## Flow
17
+
18
+ 1. Write a concise requirements.md on the branch and commit.
19
+ 2. Create a small beads sprint root + a task per item, each with `--acceptance="..."` and a
20
+ `--metadata '{"model": "<tier>"}'` tag.
21
+ 3. Dispatch the doer (tier sized to the work). The doer claims each task
22
+ (`bd update <id> --claim`), implements, commits, pushes, and closes it
23
+ (`bd close <id>`).
24
+ 4. Dispatch the reviewer (premium-tier, fresh session). It reads each task's acceptance
25
+ criteria (`bd show <id>`) + the diff and outputs a verdict: APPROVED or
26
+ CHANGES NEEDED.
27
+ 5. On APPROVED: close the delivered source issues, clean sprint scaffolding from the
28
+ PR (see sprint.md Completion step 7 -- git rm sprint-created narrative files,
29
+ restore any the repo already shipped, then verify the net diff is product only),
30
+ commit as pm, raise the PR (or report the diff for local-only), remove the
31
+ worktree.
32
+ 6. On CHANGES NEEDED: the orchestrator reads `reopenIds` and `newTasks` from the
33
+ reviewer's structured output (the reviewer never writes feedback.md); runs
34
+ `bd update <id> --status=open` for each reopen ID and `bd create` for each new
35
+ task; then re-dispatches the doer from step 3.
36
+ 7. Create low-priority beads tasks for any unresolved findings or deferred items.
37
+ 8. STOP: do not merge. Surface the PR URL and CI status to the user and await
38
+ explicit instruction.
39
+
40
+ ## Rules
41
+
42
+ - Still requires pre-flight checks (see doer-reviewer-loop.md Pre-flight checks).
43
+ - In fleet mode: still requires permissions and doer/reviewer pairing (see
44
+ fleet-addendum.md).
45
+ - Beads + git carry the state -- as in every sprint, no progress.json or PLAN.md.
46
+ - Branch naming: choose a name that makes the purpose clear --
47
+ feat/<description>, fix/<description>, chore/<description>.
48
+
49
+ ## Recovery after orchestrator restart
50
+
51
+ Recovery relies on beads and git history.
52
+
53
+ 1. `bd list --tree <sprint-id>` (and `bd ready`) -- what is open, in progress, closed.
54
+ 2. Per worktree: `git log --oneline -5` -- any commits since last known state?
55
+ 3. `git status` -- uncommitted changes?
56
+
57
+ Then act:
58
+
59
+ - **Task in_progress with no matching commits** -> reset it
60
+ (`bd update <id> --status=open`) and re-dispatch the doer.
61
+ - **Work committed, next step clear** -> dispatch reviewer or continue doer.
62
+ - **At review checkpoint** -> dispatch reviewer (fresh session).
63
+ - **Uncommitted changes of unknown origin** -> escalate to user: "commit and
64
+ resume, or discard?"
65
+ - **No progress** -> re-dispatch from scratch.