typeclaw 0.48.13 → 0.50.0

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 (572) hide show
  1. package/dist/types/agent/attention-escalation.d.ts +18 -0
  2. package/dist/types/agent/auth.d.ts +7 -0
  3. package/dist/types/agent/compaction.d.ts +8 -0
  4. package/dist/types/agent/git-nudge.d.ts +6 -0
  5. package/dist/types/agent/guard-types.d.ts +25 -0
  6. package/dist/types/agent/guards.d.ts +7 -0
  7. package/dist/types/agent/idle-compaction.d.ts +7 -0
  8. package/dist/types/agent/index.d.ts +163 -0
  9. package/dist/types/agent/live-subagents.d.ts +113 -0
  10. package/dist/types/agent/llm-replay-sanitizer.d.ts +11 -0
  11. package/dist/types/agent/loop-guard.d.ts +67 -0
  12. package/dist/types/agent/model-fallback.d.ts +29 -0
  13. package/dist/types/agent/model-overrides.d.ts +8 -0
  14. package/dist/types/agent/multimodal/index.d.ts +2 -0
  15. package/dist/types/agent/multimodal/look-at.d.ts +60 -0
  16. package/dist/types/agent/multimodal/looker.d.ts +61 -0
  17. package/dist/types/agent/multimodal/read-redirect.d.ts +6 -0
  18. package/dist/types/agent/network/response-body.d.ts +1 -0
  19. package/dist/types/agent/network/safe-http.d.ts +46 -0
  20. package/dist/types/agent/plugin-tools.d.ts +129 -0
  21. package/dist/types/agent/proactive-next-step-nudge.d.ts +2 -0
  22. package/dist/types/agent/prompt-cache-key.d.ts +3 -0
  23. package/dist/types/agent/provider-error.d.ts +23 -0
  24. package/dist/types/agent/reload-tool.d.ts +10 -0
  25. package/dist/types/agent/restart/index.d.ts +31 -0
  26. package/dist/types/agent/restart-handoff/index.d.ts +32 -0
  27. package/dist/types/agent/restart-handoff/paths.d.ts +1 -0
  28. package/dist/types/agent/retry-same-ref.d.ts +41 -0
  29. package/dist/types/agent/reviewer-activity.d.ts +3 -0
  30. package/dist/types/agent/reviewer-bash-policy.d.ts +8 -0
  31. package/dist/types/agent/self.d.ts +1 -0
  32. package/dist/types/agent/session-meta.d.ts +28 -0
  33. package/dist/types/agent/session-origin.d.ts +65 -0
  34. package/dist/types/agent/subagent-cache-keepalive.d.ts +38 -0
  35. package/dist/types/agent/subagent-completion-reminder.d.ts +29 -0
  36. package/dist/types/agent/subagent-drain.d.ts +25 -0
  37. package/dist/types/agent/subagent-session.d.ts +3 -0
  38. package/dist/types/agent/subagents.d.ts +153 -0
  39. package/dist/types/agent/system-prompt.d.ts +7 -0
  40. package/dist/types/agent/throttle-circuit.d.ts +25 -0
  41. package/dist/types/agent/todo/continuation-policy.d.ts +54 -0
  42. package/dist/types/agent/todo/continuation-state.d.ts +11 -0
  43. package/dist/types/agent/todo/continuation-wiring.d.ts +32 -0
  44. package/dist/types/agent/todo/continuation.d.ts +21 -0
  45. package/dist/types/agent/todo/scope.d.ts +6 -0
  46. package/dist/types/agent/todo/store.d.ts +16 -0
  47. package/dist/types/agent/tool-error-remediation.d.ts +1 -0
  48. package/dist/types/agent/tool-file-safety.d.ts +78 -0
  49. package/dist/types/agent/tool-not-found-nudge.d.ts +9 -0
  50. package/dist/types/agent/tool-result-budget.d.ts +15 -0
  51. package/dist/types/agent/tools/channel-disengage.d.ts +18 -0
  52. package/dist/types/agent/tools/channel-edit.d.ts +19 -0
  53. package/dist/types/agent/tools/channel-fetch-attachment.d.ts +36 -0
  54. package/dist/types/agent/tools/channel-handoff.d.ts +13 -0
  55. package/dist/types/agent/tools/channel-history-render.d.ts +3 -0
  56. package/dist/types/agent/tools/channel-history.d.ts +25 -0
  57. package/dist/types/agent/tools/channel-log.d.ts +5 -0
  58. package/dist/types/agent/tools/channel-react.d.ts +23 -0
  59. package/dist/types/agent/tools/channel-read.d.ts +27 -0
  60. package/dist/types/agent/tools/channel-reply.d.ts +47 -0
  61. package/dist/types/agent/tools/channel-send.d.ts +42 -0
  62. package/dist/types/agent/tools/channel-session-search.d.ts +10 -0
  63. package/dist/types/agent/tools/curl-impersonate.d.ts +37 -0
  64. package/dist/types/agent/tools/ddg.d.ts +18 -0
  65. package/dist/types/agent/tools/grant-role.d.ts +25 -0
  66. package/dist/types/agent/tools/keyed-semaphore.d.ts +10 -0
  67. package/dist/types/agent/tools/post-github-review.d.ts +38 -0
  68. package/dist/types/agent/tools/restart.d.ts +25 -0
  69. package/dist/types/agent/tools/runtime-notice.d.ts +3 -0
  70. package/dist/types/agent/tools/search-retry.d.ts +16 -0
  71. package/dist/types/agent/tools/skip-response.d.ts +17 -0
  72. package/dist/types/agent/tools/spawn-subagent.d.ts +77 -0
  73. package/dist/types/agent/tools/stream-snapshot.d.ts +23 -0
  74. package/dist/types/agent/tools/subagent-access.d.ts +20 -0
  75. package/dist/types/agent/tools/subagent-cancel.d.ts +23 -0
  76. package/dist/types/agent/tools/subagent-output.d.ts +50 -0
  77. package/dist/types/agent/tools/todo/index.d.ts +14 -0
  78. package/dist/types/agent/tools/webfetch/adaptive-html.d.ts +29 -0
  79. package/dist/types/agent/tools/webfetch/fetch.d.ts +26 -0
  80. package/dist/types/agent/tools/webfetch/html-lexical.d.ts +27 -0
  81. package/dist/types/agent/tools/webfetch/index.d.ts +1 -0
  82. package/dist/types/agent/tools/webfetch/inline-style-visibility.d.ts +3 -0
  83. package/dist/types/agent/tools/webfetch/strategies/grep.d.ts +11 -0
  84. package/dist/types/agent/tools/webfetch/strategies/jq.d.ts +4 -0
  85. package/dist/types/agent/tools/webfetch/strategies/raw.d.ts +1 -0
  86. package/dist/types/agent/tools/webfetch/strategies/readability.d.ts +16 -0
  87. package/dist/types/agent/tools/webfetch/strategies/selector.d.ts +4 -0
  88. package/dist/types/agent/tools/webfetch/strategies/snapshot.d.ts +1 -0
  89. package/dist/types/agent/tools/webfetch/tool.d.ts +15 -0
  90. package/dist/types/agent/tools/webfetch/types.d.ts +32 -0
  91. package/dist/types/agent/tools/websearch.d.ts +17 -0
  92. package/dist/types/agent/tools/wikipedia.d.ts +7 -0
  93. package/dist/types/agent/tools-without-local-file-operands.d.ts +1 -0
  94. package/dist/types/agent/turn-runner.d.ts +36 -0
  95. package/dist/types/agent-messenger/config-dir.d.ts +19 -0
  96. package/dist/types/browser/dockerfile.d.ts +2 -0
  97. package/dist/types/browser/registry.d.ts +49 -0
  98. package/dist/types/bundled-plugins/bun-hygiene/policy.d.ts +18 -0
  99. package/dist/types/bundled-plugins/github-cli-auth/git-askpass.d.ts +5 -0
  100. package/dist/types/bundled-plugins/guard/keys.d.ts +6 -0
  101. package/dist/types/bundled-plugins/guard/policies/managed-config.d.ts +7 -0
  102. package/dist/types/bundled-plugins/guard/policies/memory-topics-delete.d.ts +11 -0
  103. package/dist/types/bundled-plugins/guard/policies/memory-topics-write.d.ts +7 -0
  104. package/dist/types/bundled-plugins/guard/policies/non-workspace-write.d.ts +9 -0
  105. package/dist/types/bundled-plugins/guard/policies/skill-authoring.d.ts +20 -0
  106. package/dist/types/bundled-plugins/guard/policies/uncommitted-changes.d.ts +12 -0
  107. package/dist/types/bundled-plugins/guard/policy.d.ts +11 -0
  108. package/dist/types/bundled-plugins/memory/secret-detector.d.ts +11 -0
  109. package/dist/types/bundled-plugins/memory/slug.d.ts +4 -0
  110. package/dist/types/bundled-plugins/security/permissions.d.ts +20 -0
  111. package/dist/types/bundled-plugins/security/policies/private-surface-read.d.ts +34 -0
  112. package/dist/types/bundled-plugins/security/policies/ssrf.d.ts +18 -0
  113. package/dist/types/bundled-plugins/security/policy.d.ts +4 -0
  114. package/dist/types/channels/adapter-namespace.d.ts +2 -0
  115. package/dist/types/channels/adapters/mention-hints.d.ts +12 -0
  116. package/dist/types/channels/adapters/webex-id-ref.d.ts +5 -0
  117. package/dist/types/channels/background-handoff.d.ts +56 -0
  118. package/dist/types/channels/background-obligations.d.ts +135 -0
  119. package/dist/types/channels/catch-up.d.ts +26 -0
  120. package/dist/types/channels/commands.d.ts +2 -0
  121. package/dist/types/channels/completion-claim.d.ts +20 -0
  122. package/dist/types/channels/continuation-willingness.d.ts +1 -0
  123. package/dist/types/channels/continuity-types.d.ts +75 -0
  124. package/dist/types/channels/describe-error.d.ts +1 -0
  125. package/dist/types/channels/engagement.d.ts +34 -0
  126. package/dist/types/channels/github-false-receipt.d.ts +20 -0
  127. package/dist/types/channels/github-repo.d.ts +2 -0
  128. package/dist/types/channels/github-rereview-guard.d.ts +24 -0
  129. package/dist/types/channels/github-review-check.d.ts +135 -0
  130. package/dist/types/channels/github-review-claim.d.ts +3 -0
  131. package/dist/types/channels/github-review-turn-ledger.d.ts +71 -0
  132. package/dist/types/channels/github-review-verdict-coordinator.d.ts +117 -0
  133. package/dist/types/channels/github-token-bridge.d.ts +20 -0
  134. package/dist/types/channels/github-verdict-activity.d.ts +12 -0
  135. package/dist/types/channels/inbound-journal.d.ts +272 -0
  136. package/dist/types/channels/last-alive.d.ts +3 -0
  137. package/dist/types/channels/membership-cache.d.ts +27 -0
  138. package/dist/types/channels/membership.d.ts +21 -0
  139. package/dist/types/channels/outbound-flood-filter.d.ts +7 -0
  140. package/dist/types/channels/participants.d.ts +4 -0
  141. package/dist/types/channels/persistence.d.ts +32 -0
  142. package/dist/types/channels/recovery-notice.d.ts +26 -0
  143. package/dist/types/channels/recovery-outbox.d.ts +58 -0
  144. package/dist/types/channels/router.d.ts +798 -0
  145. package/dist/types/channels/schema.d.ts +501 -0
  146. package/dist/types/channels/session-handoff.d.ts +58 -0
  147. package/dist/types/channels/session-search.d.ts +29 -0
  148. package/dist/types/channels/types.d.ts +414 -0
  149. package/dist/types/channels/waiting-reactions.d.ts +25 -0
  150. package/dist/types/cli/builtins.d.ts +2 -0
  151. package/dist/types/cli/command-meta.d.ts +9 -0
  152. package/dist/types/coding-agent/broker-body.d.ts +7 -0
  153. package/dist/types/coding-agent/broker.d.ts +22 -0
  154. package/dist/types/coding-agent/claude.d.ts +22 -0
  155. package/dist/types/coding-agent/codex.d.ts +19 -0
  156. package/dist/types/coding-agent/index.d.ts +71 -0
  157. package/dist/types/coding-agent/omo-state.d.ts +2 -0
  158. package/dist/types/coding-agent/omo-stream.d.ts +22 -0
  159. package/dist/types/coding-agent/omo.d.ts +133 -0
  160. package/dist/types/coding-agent/runner.d.ts +35 -0
  161. package/dist/types/coding-agent/tool.d.ts +27 -0
  162. package/dist/types/coding-agent/worktree.d.ts +22 -0
  163. package/dist/types/commands/index.d.ts +45 -0
  164. package/dist/types/config/config.d.ts +814 -0
  165. package/dist/types/config/index.d.ts +3 -0
  166. package/dist/types/config/model-overrides.d.ts +3 -0
  167. package/dist/types/config/providers.d.ts +1456 -0
  168. package/dist/types/config/reloadable.d.ts +9 -0
  169. package/dist/types/container/agent-operation-lock.d.ts +31 -0
  170. package/dist/types/container/build-cache.d.ts +39 -0
  171. package/dist/types/container/compose-project.d.ts +1 -0
  172. package/dist/types/container/controller.d.ts +25 -0
  173. package/dist/types/container/docker-app.d.ts +23 -0
  174. package/dist/types/container/exit-reason.d.ts +10 -0
  175. package/dist/types/container/index.d.ts +14 -0
  176. package/dist/types/container/log-archive.d.ts +68 -0
  177. package/dist/types/container/log-colors.d.ts +3 -0
  178. package/dist/types/container/log-timestamps.d.ts +9 -0
  179. package/dist/types/container/logs.d.ts +37 -0
  180. package/dist/types/container/memory-limit.d.ts +13 -0
  181. package/dist/types/container/memory-oversubscription.d.ts +20 -0
  182. package/dist/types/container/oom-events.d.ts +18 -0
  183. package/dist/types/container/port.d.ts +18 -0
  184. package/dist/types/container/require-running.d.ts +16 -0
  185. package/dist/types/container/restart.d.ts +32 -0
  186. package/dist/types/container/shared.d.ts +93 -0
  187. package/dist/types/container/shell.d.ts +35 -0
  188. package/dist/types/container/start.d.ts +144 -0
  189. package/dist/types/container/stats.d.ts +31 -0
  190. package/dist/types/container/status.d.ts +25 -0
  191. package/dist/types/container/stop.d.ts +24 -0
  192. package/dist/types/container/tool-oom-priority.d.ts +4 -0
  193. package/dist/types/container/verify-running.d.ts +49 -0
  194. package/dist/types/cron/consumer.d.ts +42 -0
  195. package/dist/types/cron/count-state.d.ts +24 -0
  196. package/dist/types/cron/index.d.ts +21 -0
  197. package/dist/types/cron/list.d.ts +33 -0
  198. package/dist/types/cron/reloadable.d.ts +11 -0
  199. package/dist/types/cron/scheduler.d.ts +51 -0
  200. package/dist/types/cron/schema.d.ts +108 -0
  201. package/dist/types/dependencies/bin-resolver.d.ts +28 -0
  202. package/dist/types/dependencies/bins.d.ts +22 -0
  203. package/dist/types/dependencies/index.d.ts +2 -0
  204. package/dist/types/git/bound-gc-config.d.ts +4 -0
  205. package/dist/types/git/hookless.d.ts +1 -0
  206. package/dist/types/git/reconcile-ignored.d.ts +13 -0
  207. package/dist/types/git/resolve-agent-git.d.ts +8 -0
  208. package/dist/types/git/resolve-identity.d.ts +10 -0
  209. package/dist/types/git/run.d.ts +24 -0
  210. package/dist/types/git/system-commit.d.ts +2 -0
  211. package/dist/types/hostd/client.d.ts +13 -0
  212. package/dist/types/hostd/models.d.ts +1 -0
  213. package/dist/types/hostd/paths.d.ts +17 -0
  214. package/dist/types/hostd/protocol.d.ts +93 -0
  215. package/dist/types/hostd/spawn.d.ts +17 -0
  216. package/dist/types/hostd/version.d.ts +18 -0
  217. package/dist/types/init/auto-upgrade.d.ts +41 -0
  218. package/dist/types/init/cli-version.d.ts +7 -0
  219. package/dist/types/init/dockerfile.d.ts +34 -0
  220. package/dist/types/init/ensure-deps.d.ts +17 -0
  221. package/dist/types/init/env-file.d.ts +3 -0
  222. package/dist/types/init/gitignore.d.ts +5 -0
  223. package/dist/types/init/local-source-stamp.d.ts +3 -0
  224. package/dist/types/init/packagejson.d.ts +7 -0
  225. package/dist/types/init/paths.d.ts +3 -0
  226. package/dist/types/init/reconcile-plugin-deps.d.ts +13 -0
  227. package/dist/types/init/run-bun-install.d.ts +16 -0
  228. package/dist/types/init/windows-dev-link.d.ts +8 -0
  229. package/dist/types/mcp/catalog.d.ts +7 -0
  230. package/dist/types/mcp/client.d.ts +58 -0
  231. package/dist/types/mcp/manager.d.ts +56 -0
  232. package/dist/types/mcp/tools.d.ts +23 -0
  233. package/dist/types/models/embedding-model.d.ts +20 -0
  234. package/dist/types/models/transformers-version.d.ts +1 -0
  235. package/dist/types/network/internal-destinations.d.ts +29 -0
  236. package/dist/types/operations/classify-tool-outcome.d.ts +35 -0
  237. package/dist/types/operations/incidents.d.ts +47 -0
  238. package/dist/types/operations/index.d.ts +3 -0
  239. package/dist/types/operations/remediations.d.ts +25 -0
  240. package/dist/types/path-safety/real-intended-path.d.ts +9 -0
  241. package/dist/types/permissions/builtins.d.ts +25 -0
  242. package/dist/types/permissions/grant.d.ts +19 -0
  243. package/dist/types/permissions/index.d.ts +6 -0
  244. package/dist/types/permissions/match-rule.d.ts +28 -0
  245. package/dist/types/permissions/permissions.d.ts +42 -0
  246. package/dist/types/permissions/resolve.d.ts +25 -0
  247. package/dist/types/permissions/schema.d.ts +14 -0
  248. package/dist/types/plugin/context.d.ts +20 -0
  249. package/dist/types/plugin/core-tool-names.d.ts +4 -0
  250. package/dist/types/plugin/define.d.ts +46 -0
  251. package/dist/types/plugin/hooks.d.ts +52 -0
  252. package/dist/types/plugin/index.d.ts +11 -0
  253. package/dist/types/plugin/load-skill.d.ts +14 -0
  254. package/dist/types/plugin/loader.d.ts +25 -0
  255. package/dist/types/plugin/manager.d.ts +42 -0
  256. package/dist/types/plugin/registry.d.ts +73 -0
  257. package/dist/types/plugin/skills.d.ts +11 -0
  258. package/dist/types/plugin/types.d.ts +336 -0
  259. package/dist/types/plugin/zod-introspect.d.ts +11 -0
  260. package/dist/types/reload/client.d.ts +11 -0
  261. package/dist/types/reload/docker-exec-client.d.ts +11 -0
  262. package/dist/types/reload/format.d.ts +2 -0
  263. package/dist/types/reload/index.d.ts +6 -0
  264. package/dist/types/reload/recover.d.ts +22 -0
  265. package/dist/types/reload/registry.d.ts +10 -0
  266. package/dist/types/reload/types.d.ts +25 -0
  267. package/dist/types/role-claim/client.d.ts +20 -0
  268. package/dist/types/role-claim/code.d.ts +4 -0
  269. package/dist/types/role-claim/controller.d.ts +59 -0
  270. package/dist/types/role-claim/index.d.ts +6 -0
  271. package/dist/types/role-claim/match-rule.d.ts +9 -0
  272. package/dist/types/role-claim/pending.d.ts +35 -0
  273. package/dist/types/role-claim/reload-after-claim.d.ts +16 -0
  274. package/dist/types/run/llm-fetch-observer.d.ts +50 -0
  275. package/dist/types/sandbox/availability.d.ts +21 -0
  276. package/dist/types/sandbox/build.d.ts +7 -0
  277. package/dist/types/sandbox/canonical-secrets.d.ts +8 -0
  278. package/dist/types/sandbox/coding-agent.d.ts +20 -0
  279. package/dist/types/sandbox/dependency-bins.d.ts +19 -0
  280. package/dist/types/sandbox/dev-source.d.ts +4 -0
  281. package/dist/types/sandbox/env-exposure.d.ts +1 -0
  282. package/dist/types/sandbox/errors.d.ts +20 -0
  283. package/dist/types/sandbox/hidden-paths.d.ts +11 -0
  284. package/dist/types/sandbox/index.d.ts +14 -0
  285. package/dist/types/sandbox/package-install.d.ts +1 -0
  286. package/dist/types/sandbox/policy.d.ts +71 -0
  287. package/dist/types/sandbox/privileged-runtime.d.ts +13 -0
  288. package/dist/types/sandbox/quote.d.ts +2 -0
  289. package/dist/types/sandbox/runtime-private.d.ts +1 -0
  290. package/dist/types/sandbox/session-tmp.d.ts +6 -0
  291. package/dist/types/sandbox/symlinks.d.ts +6 -0
  292. package/dist/types/sandbox/writable-zones.d.ts +15 -0
  293. package/dist/types/secrets/coding-agent-auth.d.ts +25 -0
  294. package/dist/types/secrets/defaults.d.ts +21 -0
  295. package/dist/types/secrets/export-github-cli-store.d.ts +22 -0
  296. package/dist/types/secrets/hydrate.d.ts +7 -0
  297. package/dist/types/secrets/index.d.ts +7 -0
  298. package/dist/types/secrets/provision-github-cli-store.d.ts +45 -0
  299. package/dist/types/secrets/refresh-provider-oauth.d.ts +22 -0
  300. package/dist/types/secrets/resolve.d.ts +14 -0
  301. package/dist/types/secrets/schema.d.ts +840 -0
  302. package/dist/types/secrets/storage.d.ts +56 -0
  303. package/dist/types/shared/host-locale.d.ts +1 -0
  304. package/dist/types/shared/index.d.ts +5 -0
  305. package/dist/types/shared/local-time.d.ts +4 -0
  306. package/dist/types/shared/platform.d.ts +2 -0
  307. package/dist/types/shared/protocol.d.ts +424 -0
  308. package/dist/types/shared/style-text.d.ts +3 -0
  309. package/dist/types/shared/wsl.d.ts +17 -0
  310. package/dist/types/stream/broker.d.ts +2 -0
  311. package/dist/types/stream/index.d.ts +2 -0
  312. package/dist/types/stream/types.d.ts +71 -0
  313. package/package.json +13 -6
  314. package/scripts/build-plugin-types.ts +105 -0
  315. package/scripts/check-plugin-sdk.ts +111 -0
  316. package/scripts/dump-system-prompt.ts +52 -110
  317. package/scripts/require-parallel.ts +5 -32
  318. package/src/agent/guard-types.ts +3 -0
  319. package/src/agent/guards.ts +4 -0
  320. package/src/agent/idle-compaction.ts +174 -0
  321. package/src/agent/index.ts +210 -80
  322. package/src/agent/live-sessions.ts +2 -2
  323. package/src/agent/live-subagents.ts +20 -0
  324. package/src/agent/multimodal/look-at.ts +22 -9
  325. package/src/agent/plugin-tools.ts +66 -12
  326. package/src/agent/prompt-cache-key.ts +42 -0
  327. package/src/agent/provider-error.ts +179 -10
  328. package/src/agent/restart-handoff/index.ts +7 -14
  329. package/src/agent/reviewer-activity.ts +68 -0
  330. package/src/agent/reviewer-bash-policy.ts +41 -1
  331. package/src/agent/session-origin.ts +58 -9
  332. package/src/agent/subagent-cache-keepalive.ts +198 -0
  333. package/src/agent/subagent-session.ts +6 -0
  334. package/src/agent/subagents.ts +59 -9
  335. package/src/agent/system-prompt.ts +68 -202
  336. package/src/agent/todo/continuation-policy.ts +12 -4
  337. package/src/agent/tool-file-safety.ts +119 -15
  338. package/src/agent/tool-not-found-nudge.ts +1 -1
  339. package/src/agent/tools/channel-edit.ts +20 -10
  340. package/src/agent/tools/channel-fetch-attachment.ts +12 -5
  341. package/src/agent/tools/channel-handoff.ts +72 -0
  342. package/src/agent/tools/channel-history-render.ts +2 -1
  343. package/src/agent/tools/channel-read.ts +57 -10
  344. package/src/agent/tools/channel-reply.ts +55 -30
  345. package/src/agent/tools/channel-send.ts +69 -19
  346. package/src/agent/tools/channel-session-search.ts +35 -0
  347. package/src/agent/tools/curl-impersonate.ts +22 -6
  348. package/src/agent/tools/ddg.ts +13 -5
  349. package/src/agent/tools/post-github-review.ts +77 -15
  350. package/src/agent/tools/skip-response.ts +2 -2
  351. package/src/agent/tools/spawn-subagent.ts +190 -23
  352. package/src/agent/tools/stream-snapshot.ts +15 -2
  353. package/src/agent/tools/subagent-access.ts +11 -4
  354. package/src/agent/tools/subagent-output.ts +8 -0
  355. package/src/browser/bsk-shim.mjs +74 -0
  356. package/src/browser/dockerfile.ts +84 -0
  357. package/src/browser/egress.ts +104 -0
  358. package/src/browser/managed-browser.ts +205 -0
  359. package/src/browser/registry.ts +110 -0
  360. package/src/browser/relay.ts +231 -0
  361. package/src/browser/sandbox-entry.ts +167 -0
  362. package/src/bundled-plugins/agent-browser/index.ts +2 -0
  363. package/src/bundled-plugins/backup/README.md +10 -9
  364. package/src/bundled-plugins/backup/runner.ts +140 -58
  365. package/src/bundled-plugins/backup/subagents.ts +58 -26
  366. package/src/bundled-plugins/browser-agent/browser-agent.ts +89 -0
  367. package/src/bundled-plugins/browser-agent/index.ts +12 -0
  368. package/src/bundled-plugins/doc-render/index.ts +0 -10
  369. package/src/bundled-plugins/doc-render/render.ts +83 -111
  370. package/src/bundled-plugins/doc-render/skills/typeclaw-render-pdf/SKILL.md +152 -257
  371. package/src/bundled-plugins/github-cli-auth/gh-command.ts +38 -2
  372. package/src/bundled-plugins/github-cli-auth/git-auth-diagnostic.ts +95 -0
  373. package/src/bundled-plugins/github-cli-auth/granted-push.ts +135 -0
  374. package/src/bundled-plugins/github-cli-auth/index.ts +158 -70
  375. package/src/bundled-plugins/github-cli-auth/review-account.ts +23 -0
  376. package/src/bundled-plugins/github-cli-auth/review-checkout.ts +9 -2
  377. package/src/bundled-plugins/github-cli-auth/review-recorder.ts +45 -7
  378. package/src/bundled-plugins/managed-browser/index.ts +74 -0
  379. package/src/bundled-plugins/managed-browser/skills/browser-use/browser-use/SKILL.md +27 -0
  380. package/src/bundled-plugins/managed-browser/skills/browserskill/browserskill/SKILL.md +28 -0
  381. package/src/bundled-plugins/managed-browser/skills/omowright/omowright/SKILL.md +40 -0
  382. package/src/bundled-plugins/operator/operator.ts +15 -4
  383. package/src/bundled-plugins/qa/index.ts +11 -0
  384. package/src/bundled-plugins/qa/qa.ts +160 -0
  385. package/src/bundled-plugins/qa/skills/qa-cli.ts +50 -0
  386. package/src/bundled-plugins/qa/skills/qa-general.ts +30 -0
  387. package/src/bundled-plugins/qa/skills/qa-http.ts +39 -0
  388. package/src/bundled-plugins/qa/skills/qa-pr.ts +51 -0
  389. package/src/bundled-plugins/qa/skills/qa-web.ts +49 -0
  390. package/src/bundled-plugins/reviewer/skills/code-review.ts +1 -1
  391. package/src/bundled-plugins/security/index.ts +103 -36
  392. package/src/bundled-plugins/security/policies/git-exfil.ts +24 -12
  393. package/src/bundled-plugins/security/policies/outbound-secret-scan.ts +1 -0
  394. package/src/channels/adapter-namespace.ts +13 -0
  395. package/src/channels/adapters/discord-bot-reactions.ts +18 -0
  396. package/src/channels/adapters/discord-bot.ts +219 -72
  397. package/src/channels/adapters/discord-reactions.ts +18 -0
  398. package/src/channels/adapters/discord.ts +114 -8
  399. package/src/channels/adapters/github/inbound.ts +103 -8
  400. package/src/channels/adapters/github/index.ts +267 -19
  401. package/src/channels/adapters/github/last-alive.ts +15 -0
  402. package/src/channels/adapters/github/outbound.ts +54 -18
  403. package/src/channels/adapters/github/pending-review-store.ts +141 -0
  404. package/src/channels/adapters/github/permission-guidance.ts +1 -0
  405. package/src/channels/adapters/github/reactions.ts +100 -0
  406. package/src/channels/adapters/github/reconcile-cooldown-store.ts +197 -67
  407. package/src/channels/adapters/github/reconcile-open-prs.ts +418 -33
  408. package/src/channels/adapters/github/recovery.ts +106 -0
  409. package/src/channels/adapters/github/review-check.ts +510 -0
  410. package/src/channels/adapters/github/review-submitter.ts +34 -2
  411. package/src/channels/adapters/instagram.ts +32 -14
  412. package/src/channels/adapters/kakaotalk.ts +22 -7
  413. package/src/channels/adapters/line.ts +18 -3
  414. package/src/channels/adapters/outbound-account.ts +20 -0
  415. package/src/channels/adapters/own-reactions.ts +66 -0
  416. package/src/channels/adapters/recovery-correlation.ts +214 -0
  417. package/src/channels/adapters/slack-attachment-download.ts +14 -3
  418. package/src/channels/adapters/slack-bot-author-resolver.ts +4 -3
  419. package/src/channels/adapters/slack-bot-channel-resolver.ts +9 -6
  420. package/src/channels/adapters/slack-bot-classify.ts +19 -11
  421. package/src/channels/adapters/slack-bot-dedupe.ts +111 -28
  422. package/src/channels/adapters/slack-bot-reactions.ts +18 -0
  423. package/src/channels/adapters/slack-bot-slash-commands.ts +3 -1
  424. package/src/channels/adapters/slack-bot.ts +424 -124
  425. package/src/channels/adapters/slack-classify.ts +26 -4
  426. package/src/channels/adapters/slack-inbound-revision.ts +96 -0
  427. package/src/channels/adapters/slack-reactions.ts +18 -0
  428. package/src/channels/adapters/slack-thread-ts.ts +6 -0
  429. package/src/channels/adapters/slack.ts +48 -3
  430. package/src/channels/adapters/teams.ts +14 -4
  431. package/src/channels/adapters/telegram-bot.ts +56 -10
  432. package/src/channels/adapters/webex-bot.ts +12 -2
  433. package/src/channels/adapters/webex.ts +12 -2
  434. package/src/channels/background-handoff.ts +434 -0
  435. package/src/channels/background-obligations.ts +893 -0
  436. package/src/channels/catch-up.ts +137 -0
  437. package/src/channels/continuation-willingness.ts +2 -2
  438. package/src/channels/continuity-types.ts +272 -0
  439. package/src/channels/engagement.ts +17 -1
  440. package/src/channels/github-false-receipt.ts +3 -3
  441. package/src/channels/github-rereview-guard.ts +9 -5
  442. package/src/channels/github-review-check.ts +598 -0
  443. package/src/channels/github-review-turn-ledger.ts +102 -28
  444. package/src/channels/github-review-verdict-coordinator.ts +72 -12
  445. package/src/channels/github-token-bridge.ts +35 -5
  446. package/src/channels/inbound-journal.ts +1738 -0
  447. package/src/channels/index.ts +6 -1
  448. package/src/channels/last-alive.ts +34 -0
  449. package/src/channels/manager.ts +25 -8
  450. package/src/channels/recovery-dispatcher.ts +264 -0
  451. package/src/channels/recovery-notice.ts +94 -0
  452. package/src/channels/recovery-outbox.ts +459 -0
  453. package/src/channels/router.ts +2951 -736
  454. package/src/channels/schema.ts +8 -1
  455. package/src/channels/session-handoff.fixture.ts +144 -0
  456. package/src/channels/session-handoff.ts +175 -0
  457. package/src/channels/session-search.ts +71 -0
  458. package/src/channels/subagent-completion-bridge.ts +17 -14
  459. package/src/channels/types.ts +133 -18
  460. package/src/channels/waiting-reactions.ts +664 -0
  461. package/src/cli/channel.ts +2 -3
  462. package/src/cli/compose.ts +7 -4
  463. package/src/cli/hostd.ts +49 -3
  464. package/src/cli/init.ts +2 -3
  465. package/src/cli/inspect.ts +11 -14
  466. package/src/cli/model.ts +132 -3
  467. package/src/coding-agent/broker-body.ts +38 -0
  468. package/src/coding-agent/broker.ts +290 -0
  469. package/src/coding-agent/claude.ts +95 -0
  470. package/src/coding-agent/codex.ts +104 -0
  471. package/src/coding-agent/index.ts +512 -0
  472. package/src/coding-agent/omo-state.ts +69 -0
  473. package/src/coding-agent/omo-stream.ts +138 -0
  474. package/src/coding-agent/omo.ts +192 -0
  475. package/src/coding-agent/runner.ts +285 -0
  476. package/src/coding-agent/tool.ts +62 -0
  477. package/src/coding-agent/worktree.ts +355 -0
  478. package/src/compose/restart.ts +10 -3
  479. package/src/compose/start.ts +20 -3
  480. package/src/config/config.ts +121 -25
  481. package/src/config/model-overrides.ts +59 -0
  482. package/src/config/models-mutation.ts +185 -2
  483. package/src/config/providers-mutation.ts +12 -4
  484. package/src/config/providers.ts +71 -9
  485. package/src/container/exit-reason.ts +44 -0
  486. package/src/container/index.ts +5 -0
  487. package/src/container/log-archive.ts +51 -13
  488. package/src/container/memory-limit.ts +1 -1
  489. package/src/container/memory-oversubscription.ts +36 -6
  490. package/src/container/oom-events.ts +86 -0
  491. package/src/container/restart.ts +2 -0
  492. package/src/container/shared.ts +8 -1
  493. package/src/container/start.ts +120 -17
  494. package/src/container/stop.ts +6 -2
  495. package/src/container/tool-oom-priority.ts +32 -0
  496. package/src/cron/consumer.ts +10 -4
  497. package/src/cron/reloadable.ts +23 -2
  498. package/src/doctor/checks.ts +22 -8
  499. package/src/hostd/daemon.ts +33 -11
  500. package/src/hostd/spawn.ts +106 -5
  501. package/src/hostd/version.ts +20 -0
  502. package/src/hostd/watchdog.ts +241 -0
  503. package/src/init/auto-upgrade.ts +7 -1
  504. package/src/init/dockerfile.ts +112 -602
  505. package/src/init/index.ts +6 -6
  506. package/src/init/local-source-stamp.ts +68 -0
  507. package/src/init/models-dev.ts +4 -0
  508. package/src/init/validate-api-key.ts +3 -0
  509. package/src/mcp/client.ts +3 -1
  510. package/src/operations/classify-tool-outcome.ts +45 -1
  511. package/src/operations/incidents.ts +10 -5
  512. package/src/permissions/builtins.ts +7 -0
  513. package/src/permissions/permissions.ts +18 -9
  514. package/src/permissions/resolve.ts +7 -1
  515. package/src/plugin/context.ts +2 -0
  516. package/src/plugin/core-tool-names.ts +1 -1
  517. package/src/plugin/define.ts +1 -0
  518. package/src/plugin/hooks.ts +137 -0
  519. package/src/plugin/index.ts +7 -0
  520. package/src/plugin/manager.ts +4 -0
  521. package/src/plugin/types.ts +66 -1
  522. package/src/run/background-handoff-boot.ts +86 -0
  523. package/src/run/bundled-plugins.ts +6 -0
  524. package/src/run/index.ts +107 -97
  525. package/src/run/llm-fetch-observer.ts +12 -15
  526. package/src/run/resource-sample.ts +19 -0
  527. package/src/sandbox/build.ts +4 -0
  528. package/src/sandbox/canonical-secrets.ts +4 -0
  529. package/src/sandbox/coding-agent.ts +214 -0
  530. package/src/sandbox/dev-source.ts +26 -0
  531. package/src/sandbox/hidden-paths.ts +16 -2
  532. package/src/sandbox/policy.ts +1 -0
  533. package/src/sandbox/runtime-private.ts +4 -1
  534. package/src/secrets/coding-agent-auth.ts +97 -0
  535. package/src/secrets/index.ts +0 -16
  536. package/src/secrets/storage.ts +13 -0
  537. package/src/server/command-runner.ts +6 -2
  538. package/src/server/index.ts +101 -40
  539. package/src/skills/typeclaw-channel-github/SKILL.md +15 -13
  540. package/src/skills/typeclaw-claude-code/SKILL.md +25 -260
  541. package/src/skills/typeclaw-claude-code/references/auth-flow.md +7 -23
  542. package/src/skills/typeclaw-claude-code/references/headless-sessions.md +11 -0
  543. package/src/skills/typeclaw-claude-code/references/results-and-worktrees.md +13 -0
  544. package/src/skills/typeclaw-codex-cli/SKILL.md +35 -270
  545. package/src/skills/typeclaw-codex-cli/references/auth-flow.md +13 -20
  546. package/src/skills/typeclaw-codex-cli/references/headless-sessions.md +26 -0
  547. package/src/skills/typeclaw-codex-cli/references/results-and-worktrees.md +28 -0
  548. package/src/skills/typeclaw-config/SKILL.md +60 -26
  549. package/src/skills/typeclaw-config/references/dockerfile.md +14 -13
  550. package/src/skills/typeclaw-cron/SKILL.md +1 -1
  551. package/src/skills/typeclaw-github-contributing/SKILL.md +3 -3
  552. package/src/skills/typeclaw-memory/SKILL.md +3 -3
  553. package/src/skills/typeclaw-monorepo/SKILL.md +1 -1
  554. package/src/skills/typeclaw-omo-cli/SKILL.md +52 -0
  555. package/src/skills/typeclaw-omo-cli/references/auth-flow.md +11 -0
  556. package/src/skills/typeclaw-omo-cli/references/headless-sessions.md +11 -0
  557. package/src/skills/typeclaw-omo-cli/references/results-and-worktrees.md +23 -0
  558. package/src/skills/typeclaw-permissions/SKILL.md +7 -7
  559. package/src/skills/typeclaw-plugins/SKILL.md +15 -15
  560. package/src/skills/typeclaw-skills/SKILL.md +19 -2
  561. package/src/skills/typeclaw-troubleshooting/SKILL.md +26 -2
  562. package/typeclaw.schema.json +448 -6
  563. package/src/bundled-plugins/doc-render/templates/lib.typ +0 -339
  564. package/src/bundled-plugins/github-cli-auth/gh-store.ts +0 -733
  565. package/src/secrets/claude-credentials-json.ts +0 -129
  566. package/src/secrets/codex-auth-json.ts +0 -102
  567. package/src/secrets/export-claude-credentials-file.ts +0 -270
  568. package/src/secrets/export-codex-auth-file.ts +0 -250
  569. package/src/skills/typeclaw-claude-code/references/stop-hook.md +0 -101
  570. package/src/skills/typeclaw-claude-code/references/tmux-driving.md +0 -243
  571. package/src/skills/typeclaw-codex-cli/references/stop-hook.md +0 -92
  572. package/src/skills/typeclaw-codex-cli/references/tmux-driving.md +0 -239
@@ -22,6 +22,7 @@ type GithubFetch = (input: RequestInfo | URL, init?: RequestInit) => Promise<Res
22
22
  export async function prepareReviewerCheckout(options: {
23
23
  repoSlug: string
24
24
  headSha: string
25
+ baseSha?: string
25
26
  sessionId: string
26
27
  resolveTokenForRepo: ResolveGithubTokenForRepo
27
28
  fetchImpl?: GithubFetch
@@ -30,6 +31,8 @@ export async function prepareReviewerCheckout(options: {
30
31
  }): Promise<{ path: string; repoSlug: string; headSha: string }> {
31
32
  if (!REPO_SLUG.test(options.repoSlug)) throw new Error('repository must be one owner/repo slug')
32
33
  if (!FULL_SHA.test(options.headSha)) throw new Error('headSha must be a full 40-character commit SHA')
34
+ if (options.baseSha !== undefined && !FULL_SHA.test(options.baseSha))
35
+ throw new Error('baseSha must be a full 40-character commit SHA')
33
36
  const tokenResult = await options.resolveTokenForRepo(options.repoSlug)
34
37
  if (tokenResult.kind !== 'token') throw new Error(tokenResult.reason)
35
38
  const token = tokenResult.token
@@ -61,6 +64,9 @@ export async function prepareReviewerCheckout(options: {
61
64
  '--depth=1',
62
65
  `https://github.com/${options.repoSlug}.git`,
63
66
  options.headSha,
67
+ // Without the base commit, `git diff <base> <head>` fails with "bad
68
+ // object" and the reviewer falls back to truncated per-file API patches.
69
+ ...(options.baseSha !== undefined ? [options.baseSha] : []),
64
70
  ],
65
71
  { env: { ...baseEnv, GIT_ASKPASS: askPass, TYPECLAW_GIT_TOKEN: token } },
66
72
  )
@@ -99,19 +105,20 @@ export function createReviewerCheckoutTool(resolveTokenForRepo: ResolveGithubTok
99
105
  return defineTool({
100
106
  description:
101
107
  'Prepare a runtime-owned token-safe scratch checkout of one allowlisted GitHub repository at an exact full commit SHA.',
102
- parameters: z.object({ repoSlug: z.string(), headSha: z.string() }),
108
+ parameters: z.object({ repoSlug: z.string(), headSha: z.string(), baseSha: z.string().optional() }),
103
109
  // `repoSlug` ("owner/repo") and `headSha` (40-hex) are remote GitHub
104
110
  // coordinates, never local paths — repoSlug is only ever interpolated into
105
111
  // `https://github.com/<slug>.git` and prepareReviewerCheckout rejects
106
112
  // anything not matching /^[A-Za-z0-9_.-]+\/[A-Za-z0-9_.-]+$/. Declared so the
107
113
  // required "/" in a slug isn't misread as a path separator. Scoped to THIS
108
114
  // tool: an undeclared reader passing a slug-shaped value still fails closed.
109
- fileOperands: { nonFile: ['repoSlug', 'headSha'] },
115
+ fileOperands: { nonFile: ['repoSlug', 'headSha', 'baseSha'] },
110
116
  async execute(args, ctx) {
111
117
  try {
112
118
  const receipt = await prepareReviewerCheckout({
113
119
  repoSlug: args.repoSlug,
114
120
  headSha: args.headSha,
121
+ ...(args.baseSha !== undefined ? { baseSha: args.baseSha } : {}),
115
122
  sessionId: ctx.sessionId,
116
123
  resolveTokenForRepo,
117
124
  })
@@ -1,6 +1,11 @@
1
1
  import { readFile } from 'node:fs/promises'
2
2
 
3
- import { recordReview, recordReviewOutput } from '@/channels/github-review-turn-ledger'
3
+ import {
4
+ captureReviewResultCoverage,
5
+ recordReview,
6
+ recordReviewOutput,
7
+ type ReviewResultCoverage,
8
+ } from '@/channels/github-review-turn-ledger'
4
9
  import type { ContentPart, ToolResult } from '@/plugin'
5
10
 
6
11
  import {
@@ -35,6 +40,18 @@ const submissionAttempts = new Map<string, ReviewSubmissionAttempt>()
35
40
  // credited to the false-receipt ledger via recordReview (which also fans to the
36
41
  // output observer), so only non-verdict COMMENTs need their own output credit.
37
42
  const pendingCommentOutput = new Map<string, DetectedReviewOutput>()
43
+ const pendingCoverage = new Map<string, ReviewResultCoverage>()
44
+
45
+ export function capturedReviewAccountIdentity(callId: string): string | undefined {
46
+ return pendingCoverage.get(callId)?.expectedAccountIdentity
47
+ }
48
+
49
+ export function discardReviewCommand(callId: string): void {
50
+ pending.delete(callId)
51
+ pendingCommentOutput.delete(callId)
52
+ submissionAttempts.delete(callId)
53
+ pendingCoverage.delete(callId)
54
+ }
38
55
 
39
56
  const MAX_INPUT_BYTES = 1_000_000
40
57
 
@@ -43,7 +60,12 @@ export type NoteReviewResult = {
43
60
  detected: DetectedReview | null
44
61
  }
45
62
 
46
- export async function noteReviewCommand(args: { callId: string; command: string }): Promise<NoteReviewResult> {
63
+ export async function noteReviewCommand(args: {
64
+ callId: string
65
+ command: string
66
+ sessionId?: string
67
+ }): Promise<NoteReviewResult> {
68
+ discardReviewCommand(args.callId)
47
69
  const inputFileContents = await readInputFile(args.command)
48
70
  const detected = detectReviewSubmission({ command: args.command, inputFileContents })
49
71
  if (detected !== null) pending.set(args.callId, detected)
@@ -65,6 +87,17 @@ export async function noteReviewCommand(args: { callId: string; command: string
65
87
  if (attempt !== null) submissionAttempts.set(args.callId, attempt)
66
88
  }
67
89
  }
90
+ if (pending.has(args.callId) || pendingCommentOutput.has(args.callId) || submissionAttempts.has(args.callId)) {
91
+ try {
92
+ pendingCoverage.set(
93
+ args.callId,
94
+ args.sessionId === undefined ? {} : await captureReviewResultCoverage(args.sessionId),
95
+ )
96
+ } catch (error) {
97
+ discardReviewCommand(args.callId)
98
+ throw error
99
+ }
100
+ }
68
101
  return { dump: detectReviewDump({ command: args.command, inputFileContents }), detected }
69
102
  }
70
103
 
@@ -82,21 +115,24 @@ export function dismissalMutationSucceeded(result: ToolResult): boolean {
82
115
  return !FAILURE_MARKERS.some((marker) => text.includes(marker)) && /"state"\s*:\s*"DISMISSED"/.test(text)
83
116
  }
84
117
 
85
- export function commitReviewIfSucceeded(args: {
118
+ export async function commitReviewIfSucceeded(args: {
86
119
  sessionId: string
87
120
  callId: string
88
121
  result: ToolResult
89
- }): CommitReviewResult {
122
+ }): Promise<CommitReviewResult> {
90
123
  const text = collectText(args.result.content)
124
+ const coverage = pendingCoverage.get(args.callId) ?? {}
125
+ pendingCoverage.delete(args.callId)
91
126
  const detected = pending.get(args.callId)
92
127
  if (detected !== undefined) {
93
128
  pending.delete(args.callId)
94
129
  if (!looksSucceeded(detected, text)) return { committed: false, landedFromResult: null }
95
- recordReview({
130
+ await recordReview({
96
131
  sessionId: args.sessionId,
97
132
  workspace: detected.workspace,
98
133
  prNumber: detected.prNumber,
99
134
  verdict: detected.verdict,
135
+ ...coverage,
100
136
  })
101
137
  return { committed: true, landedFromResult: null }
102
138
  }
@@ -107,11 +143,12 @@ export function commitReviewIfSucceeded(args: {
107
143
  if (comment !== undefined) {
108
144
  pendingCommentOutput.delete(args.callId)
109
145
  if (commentReviewSucceeded(text)) {
110
- recordReviewOutput({
146
+ await recordReviewOutput({
111
147
  sessionId: args.sessionId,
112
148
  workspace: comment.workspace,
113
149
  prNumber: comment.prNumber,
114
150
  state: 'COMMENT',
151
+ ...coverage,
115
152
  })
116
153
  }
117
154
  return { committed: false, landedFromResult: null }
@@ -130,11 +167,12 @@ export function commitReviewIfSucceeded(args: {
130
167
  if (landed.workspace !== attempt.workspace || landed.prNumber !== attempt.prNumber) {
131
168
  return { committed: false, landedFromResult: null }
132
169
  }
133
- recordReview({
170
+ await recordReview({
134
171
  sessionId: args.sessionId,
135
172
  workspace: landed.workspace,
136
173
  prNumber: landed.prNumber,
137
174
  verdict: landed.verdict,
175
+ ...coverage,
138
176
  })
139
177
  return { committed: true, landedFromResult: landed }
140
178
  }
@@ -0,0 +1,74 @@
1
+ import { join } from 'node:path'
2
+
3
+ import { TYPECLAW_INTERNAL_BASH_ENV } from '@/agent/plugin-tools'
4
+ import { type ManagedBrowser, startManagedBrowser } from '@/browser/managed-browser'
5
+ import { BROWSER_ENGINES, type BrowserConfig, needsManagedBrowser } from '@/browser/registry'
6
+ import { getConfig } from '@/config'
7
+ import type { PermissionService } from '@/permissions'
8
+ import { CORE_PERMISSIONS } from '@/permissions/builtins'
9
+ import { definePlugin, type Hooks } from '@/plugin'
10
+
11
+ const SKILLS_ROOT = join(import.meta.dir, 'skills')
12
+
13
+ export default definePlugin({
14
+ plugin: async (ctx) => {
15
+ // Boot snapshot: `browser` is restart-required, matching the image layers
16
+ // and the browser process started below.
17
+ const browser = getConfig().browser
18
+ const skillsDirs = browserSkillDirs(browser)
19
+ if (!needsManagedBrowser(browser)) return { skillsDirs }
20
+
21
+ let managed: ManagedBrowser
22
+ try {
23
+ managed = await startManagedBrowser({ agentDir: ctx.agentDir, browser, logger: ctx.logger })
24
+ } catch (error) {
25
+ ctx.logger.error(`managed browser failed to start; browser engines are unavailable: ${String(error)}`)
26
+ return { skillsDirs }
27
+ }
28
+ const overlay = browserEnvOverlay(browser, managed)
29
+
30
+ return {
31
+ skillsDirs,
32
+ hooks: { 'tool.before': sharedBrowserBashHook(ctx.permissions, overlay) },
33
+ onDispose: () => managed.stop(),
34
+ }
35
+ },
36
+ })
37
+
38
+ export function sharedBrowserBashHook(
39
+ permissions: PermissionService,
40
+ overlay: Record<string, string>,
41
+ ): NonNullable<Hooks['tool.before']> {
42
+ return (event) => {
43
+ if (event.tool !== 'bash') return
44
+ if (!permissions.has(event.origin, CORE_PERMISSIONS.browserShared)) return
45
+ const existing = event.args[TYPECLAW_INTERNAL_BASH_ENV]
46
+ const current = existing !== null && typeof existing === 'object' ? (existing as Record<string, string>) : {}
47
+ event.args[TYPECLAW_INTERNAL_BASH_ENV] = { ...current, ...overlay }
48
+ }
49
+ }
50
+
51
+ export function browserSkillDirs(browser: BrowserConfig): string[] {
52
+ const ids = [...browser.engines.filter((id) => id !== 'agent-browser'), ...browser.extensions]
53
+ // One root per id (skills/<id>/<skill>/SKILL.md) so disabled engines stay out of the catalog.
54
+ return ids.map((id) => join(SKILLS_ROOT, id))
55
+ }
56
+
57
+ export function browserEnvOverlay(
58
+ browser: BrowserConfig,
59
+ managed: Pick<ManagedBrowser, 'cdpUrl' | 'filesDir' | 'bskRelay' | 'token'>,
60
+ ) {
61
+ const env: Record<string, string> = {
62
+ TYPECLAW_BROWSER_CDP_URL: managed.cdpUrl,
63
+ TYPECLAW_BROWSER_FILES: managed.filesDir,
64
+ }
65
+ // agent-browser is in every image and bundled skills (qa, PDF) call it
66
+ // directly, so it always joins the shared browser while one is running.
67
+ for (const id of new Set(['agent-browser' as const, ...browser.engines])) {
68
+ Object.assign(env, BROWSER_ENGINES[id].env(managed.cdpUrl))
69
+ }
70
+ if (managed.bskRelay !== undefined) {
71
+ Object.assign(env, { TYPECLAW_BSK_RELAY: managed.bskRelay, TYPECLAW_BSK_TOKEN: managed.token })
72
+ }
73
+ return env
74
+ }
@@ -0,0 +1,27 @@
1
+ ---
2
+ name: browser-use
3
+ description: Browser Use CLI for driving the shared TypeClaw browser from Python snippets over CDP. Use for browsing, clicking, scraping and screenshots when you prefer Python helpers.
4
+ ---
5
+
6
+ # browser-use
7
+
8
+ The `browser-use` CLI runs a Python snippet with browser helpers pre-imported. `$BU_CDP_WS` points it at the shared browser. If the variable is unset, this session's role cannot use the shared browser (it needs the `browser.shared` permission) or the browser failed to start; the operator log (`typeclaw logs`) names which. Do not try to start Chrome yourself.
9
+
10
+ ```bash
11
+ browser-use <<'PY'
12
+ new_tab("https://example.com")
13
+ wait_for_load()
14
+ print(page_info())
15
+ PY
16
+ ```
17
+
18
+ Run `browser-use skill` once to read the full helper reference (`ensure_real_tab`, `click_at_xy`, `cdp(...)`, screenshots). Its tabs are marked with a horse emoji in the title; keep to those.
19
+
20
+ Never run `browser-use --update`, `browser-use auth login`, or remote/cloud daemons (`start_remote_daemon`): the version is pinned by TypeClaw and cloud browsers are not part of this setup.
21
+
22
+ ## Shared browser rules
23
+
24
+ - One TypeClaw-managed Chrome serves every engine and every session that has `browser.shared`. Logins and tabs persist across bash calls and sessions.
25
+ - Work only in tabs you opened. Never navigate or close a tab another engine or session is using.
26
+ - Files: downloads land in `$TYPECLAW_BROWSER_FILES`; to upload a file, copy it there first and pass that path. Chrome cannot see other paths.
27
+ - The browser runs in its own sandbox with no view of the agent folder, so `file://` URLs outside `$TYPECLAW_BROWSER_FILES` fail by design.
@@ -0,0 +1,28 @@
1
+ ---
2
+ name: browserskill
3
+ description: BrowserSkill (bsk) CLI driving the shared TypeClaw browser through its extension. Use for semantic page observation, table/list extraction, recording traces, and website debugging evidence.
4
+ ---
5
+
6
+ # browserskill
7
+
8
+ The `bsk` CLI talks to a daemon TypeClaw runs next to the shared browser; the BrowserSkill extension is preloaded and connected. `bsk` reaches that daemon through `$TYPECLAW_BSK_RELAY`; do not set `BSK_HOME` yourself. If the variable is unset, this session's role cannot use the shared browser (it needs the `browser.shared` permission) or the browser failed to start; the operator log (`typeclaw logs`) names which. Do not try to start Chrome yourself.
9
+
10
+ ```bash
11
+ bsk status --json # a connected browser must be listed
12
+ SID=$(bsk session start --json | jq -r .session_id)
13
+ bsk navigate https://example.com --session "$SID"
14
+ bsk snapshot --session "$SID" # @eN refs
15
+ bsk click @e3 --session "$SID"
16
+ bsk session stop "$SID" # always stop, success or failure
17
+ ```
18
+
19
+ - Sessions open their own Agent Window. Never borrow an existing tab: borrowing asks a human to confirm, and nobody can click it here.
20
+ - `bsk --help` and `bsk <command> --help` list everything (`extract`, `observe`, `record`, `debug`).
21
+ - Never run `bsk update`, `bsk daemon stop`, or `bsk daemon restart`; TypeClaw owns the daemon and its version.
22
+
23
+ ## Shared browser rules
24
+
25
+ - One TypeClaw-managed Chrome serves every engine and every session that has `browser.shared`. Logins and tabs persist across bash calls and sessions.
26
+ - Work only in tabs you opened. Never navigate or close a tab another engine or session is using.
27
+ - Files: downloads land in `$TYPECLAW_BROWSER_FILES`; to upload a file, copy it there first and pass that path. Chrome cannot see other paths.
28
+ - The browser runs in its own sandbox with no view of the agent folder, so `file://` URLs outside `$TYPECLAW_BROWSER_FILES` fail by design.
@@ -0,0 +1,40 @@
1
+ ---
2
+ name: omowright
3
+ description: Playwright-shaped browser automation library (OmOWright) for the shared TypeClaw browser. Use for browsing, clicking, form filling, accessibility snapshots, scrolling feeds and collecting JSON from network responses while scrolling.
4
+ ---
5
+
6
+ # omowright
7
+
8
+ OmOWright is a JavaScript library with the Playwright API shape (`page.goto`, `page.locator(ref).click()`, `page.mouse.wheel`, `page.keyboard.type`, `page.screenshot`). Write a short Bun script and run it with `bun`. Each script connects to the shared browser, so state survives between scripts.
9
+
10
+ `$OMOWRIGHT_CDP_URL` holds the connection URL. If the variable is unset, this session's role cannot use the shared browser (it needs the `browser.shared` permission) or the browser failed to start; the operator log (`typeclaw logs`) names which. Do not try to start Chrome yourself.
11
+
12
+ ```js
13
+ // save as /tmp/browse.mjs, run: bun /tmp/browse.mjs
14
+ import { compactSnapshot, connect, createAgentTabs } from '/usr/local/lib/typeclaw-browser/omowright/index.js'
15
+
16
+ const url = process.env.OMOWRIGHT_CDP_URL
17
+ // The relay requires the token on every request; omowright drops the URL query
18
+ // for HTTP discovery (listTargets), so pass it as an Authorization header too.
19
+ const token = new URL(url).searchParams.get('token')
20
+ const browser = await connect(url, {
21
+ client: { authorizer: { resolveAuthorization: async () => `Bearer ${token}` } },
22
+ })
23
+ const tabs = createAgentTabs(browser) // background tabs that never steal focus
24
+ const page = await tabs.open('https://example.com')
25
+ console.log(compactSnapshot(await page.snapshot())) // refs like [ref=e1]
26
+ await page.locator('e1').click()
27
+ process.exit(0) // leaves the tab and browser running
28
+ ```
29
+
30
+ - Prefer `compactSnapshot(await page.snapshot())` over screenshots; re-snapshots come back as a diff.
31
+ - Long lists: `createNetworkSnoop` + `collectWhileScrolling` read the page's JSON instead of scraping rows.
32
+ - Canvas or custom widgets: `createCua(page)` gives coordinate-level clicks and scrolls.
33
+ - `connectPipe()` and `connectBrowserSkill()` launch or find other browsers; do not use them here.
34
+
35
+ ## Shared browser rules
36
+
37
+ - One TypeClaw-managed Chrome serves every engine and every session that has `browser.shared`. Logins and tabs persist across bash calls and sessions.
38
+ - Work only in tabs you opened. Never navigate or close a tab another engine or session is using.
39
+ - Files: downloads land in `$TYPECLAW_BROWSER_FILES`; to upload a file, copy it there first and pass that path. Chrome cannot see other paths.
40
+ - The browser runs in its own sandbox with no view of the agent folder, so `file://` URLs outside `$TYPECLAW_BROWSER_FILES` fail by design.
@@ -1,6 +1,16 @@
1
1
  import { z } from 'zod'
2
2
 
3
- import { bashTool, editTool, findTool, grepTool, lsTool, readTool, type Subagent, writeTool } from '@/plugin'
3
+ import {
4
+ bashTool,
5
+ codingAgentTool,
6
+ editTool,
7
+ findTool,
8
+ grepTool,
9
+ lsTool,
10
+ readTool,
11
+ type Subagent,
12
+ writeTool,
13
+ } from '@/plugin'
4
14
 
5
15
  export const OPERATOR_SYSTEM_PROMPT = `You are an operator subagent running inside TypeClaw. Your job: execute a multi-step task on behalf of the main agent and report what happened.
6
16
 
@@ -17,6 +27,7 @@ You have a full tool set: read, write, edit, grep, find, ls, bash. You can:
17
27
  - Modify files (write/edit)
18
28
  - Run shell commands with side effects (bash without the read-only restriction)
19
29
  - Use any tool available to a normal operator session
30
+ - Delegate Claude Code, Codex, or OmO Native through \`coding_agent\`: one foreground headless run, persistent isolated worktree, explicit task-ID resume. Review its complete diff before applying; never drive credentialed tmux sessions.
20
31
 
21
32
  You CAN delegate, but rarely should:
22
33
  - You may \`spawn_subagent\` to hand a clearly separable, context-heavy chunk to a fresh worker — e.g. a focused read-only investigation of a large area you don't want to load into your own context. Spawn only when delegation clearly pays for itself; doing the work yourself is the default. The delegation chain is depth-limited, so a worker you spawn cannot spawn again — keep your own tree flat.
@@ -67,11 +78,11 @@ export function createOperatorSubagent(): Subagent<OperatorPayload> {
67
78
  return {
68
79
  systemPrompt: OPERATOR_SYSTEM_PROMPT,
69
80
  profile: 'default',
70
- tools: [readTool, grepTool, findTool, lsTool, bashTool, writeTool, editTool],
81
+ tools: [readTool, grepTool, findTool, lsTool, bashTool, writeTool, editTool, codingAgentTool],
71
82
  payloadSchema: operatorPayloadSchema,
72
83
  visibility: 'public',
73
84
  rosterDescription:
74
- 'write-capable: bash-with-side-effects, write, edit — for browser sessions, refactors, deploys, batch ops, and Claude Code / Codex CLI driving; runs at YOUR role, so it can do nothing here you could not already do inline',
85
+ 'write-capable: bash-with-side-effects, write, edit, coding_agent — for refactors, deploys, batch ops, and brokered Claude Code/Codex delegation; runs at YOUR role',
75
86
  // No `requiresSpecificPermission`: a subagent inherits `spawnedByRole`, and
76
87
  // main sessions are never narrowed to a subset of the builtin tools, so a
77
88
  // caller already holds every tool listed above. Gating the spawn bought no
@@ -84,7 +95,7 @@ export function createOperatorSubagent(): Subagent<OperatorPayload> {
84
95
  inFlightKey: (payload) => payload?.requestId ?? `anon-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`,
85
96
  toolResultBudget: {
86
97
  maxTotalBytes: 1_000_000,
87
- toolNames: ['read', 'grep', 'find', 'ls', 'bash', 'write', 'edit'],
98
+ toolNames: ['read', 'grep', 'find', 'ls', 'bash', 'write', 'edit', 'coding_agent'],
88
99
  },
89
100
  }
90
101
  }
@@ -0,0 +1,11 @@
1
+ import { definePlugin } from '@/plugin'
2
+
3
+ import { createQaSubagent } from './qa'
4
+
5
+ export default definePlugin({
6
+ plugin: async (ctx) => ({
7
+ subagents: {
8
+ qa: createQaSubagent({ resolveTokenForRepo: ctx.github.resolveTokenForRepo }),
9
+ },
10
+ }),
11
+ })
@@ -0,0 +1,160 @@
1
+ import { z } from 'zod'
2
+
3
+ import { createReviewerCheckoutTool } from '@/bundled-plugins/github-cli-auth/review-checkout'
4
+ import type { ResolveGithubTokenForRepo } from '@/channels/github-token-bridge'
5
+ import {
6
+ bashTool,
7
+ createLoadSkillTool,
8
+ findTool,
9
+ grepTool,
10
+ type LoadableSkill,
11
+ lsTool,
12
+ readTool,
13
+ type Subagent,
14
+ webFetchTool,
15
+ webSearchTool,
16
+ writeTool,
17
+ } from '@/plugin'
18
+
19
+ import { CLI_QA_SKILL } from './skills/qa-cli'
20
+ import { GENERAL_QA_SKILL } from './skills/qa-general'
21
+ import { HTTP_QA_SKILL } from './skills/qa-http'
22
+ import { PR_QA_SKILL } from './skills/qa-pr'
23
+ import { WEB_QA_SKILL } from './skills/qa-web'
24
+
25
+ // Ordered by expected frequency; `qa-general` stays last as the fallback, the same
26
+ // convention as REVIEWER_SKILLS.
27
+ export const QA_SKILLS: readonly LoadableSkill[] = [
28
+ PR_QA_SKILL,
29
+ WEB_QA_SKILL,
30
+ CLI_QA_SKILL,
31
+ HTTP_QA_SKILL,
32
+ GENERAL_QA_SKILL,
33
+ ]
34
+
35
+ // QA builds, installs dependencies, boots servers, and drives browsers, so it
36
+ // routinely outlasts a review. Same 30m ceiling as the other deep-profile
37
+ // subagents (reviewer, researcher): liveness for the parent, not cancellation.
38
+ export const QA_SPAWN_TIMEOUT_MS = 1_800_000
39
+
40
+ export const QA_SYSTEM_PROMPT = `You are a QA tester running inside TypeClaw. Your job: exercise a product, a repository, or a code change through its REAL user-facing surface and return an evidence-backed verdict on whether it works.
41
+
42
+ The \`reviewer\` reads code and judges it. You do the other half: you RUN it. A claim you did not observe through an executed command, a rendered page, or a captured response is not a QA result.
43
+
44
+ === WHAT YOU MAY CHANGE ===
45
+ You run builds, installs, servers, browsers, and test suites, so your bash is not read-only. Your side effects are bounded to a private scratch area:
46
+
47
+ - Work only inside a scratch directory under \`/tmp/qa/<run-id>/\` (pick a short unique run id). Clone, install, build, write helper scripts, and store evidence there.
48
+ - NEVER edit, commit, push, or "fix" the product under test. If you find a bug, you report it; the parent decides what happens next. A QA run that patches the code until it passes has destroyed its own evidence.
49
+ - NEVER weaken an assertion, skip a failing check, or mark a recovered step as a pass.
50
+ - NEVER post to GitHub, Slack, Discord, or any channel. The parent owns posting.
51
+ - Do not touch the agent's own workspace files, secrets, or config. Do not run destructive commands against shared or production systems; QA targets staging, local builds, or what the caller explicitly named.
52
+ - Stop every server, browser, and background process you started before you finish.
53
+
54
+ Each \`bash\` call runs in its own sandbox, and a process it backgrounds can die when that call returns. Files under \`/tmp\` persist between calls; processes do not. So start a server, wait for it, exercise it, and stop it inside ONE \`bash\` invocation, usually a script you \`write\` to your scratch directory and then run.
55
+
56
+ ## Tools
57
+
58
+ - \`read\`, \`grep\`, \`find\`, \`ls\` — inspect files and the target's source/docs
59
+ - \`bash\` — clone, install, build, run, drive CLIs, curl endpoints, drive a browser (the \`agent-browser\` CLI is always installed; other enabled engines ship their own skill)
60
+ - \`write\` — helper scripts and evidence notes inside your scratch directory only
61
+ - \`web_search\`, \`web_fetch\` — public docs or a deployed URL's static content
62
+ - \`load_skill\` — load a QA playbook for the surface under test (see below)
63
+ - \`github_prepare_review_checkout\` — when available, acquire a private GitHub repo at an exact SHA
64
+
65
+ ## Loading a QA skill
66
+
67
+ Surface-specific craft lives in skills you load on demand. First read the payload and decide what is being tested: a pull request or repository, a web app or URL, a CLI or TUI, an HTTP API, or something else. Then call \`load_skill\` with the matching name before you plan. Load a second skill when the target spans surfaces (a PR that changes a web page loads \`qa-pr\` and \`qa-web\`). If nothing fits, load \`qa-general\`. Keep the selection internal — do not narrate it in \`<summary>\`.
68
+
69
+ ## The QA loop
70
+
71
+ 1. **Plan.** Turn the request into explicit acceptance criteria: what must work, the riskiest edge (empty, boundary, malformed, unauthorized, concurrent), and one adjacent behavior that could regress. Every stated success criterion gets a scenario.
72
+ 2. **Prepare.** Isolate state in your scratch directory. Record the target identity (URL, binary version, commit SHA), the environment, and the exact commands used to install and start it. Wait for readiness on an observable signal (a port answering, a log line), never a fixed sleep.
73
+ 3. **Execute.** Exercise the behavior the way a user would: the real CLI, the real page, the real endpoint. Include at least one negative path.
74
+ 4. **Capture.** Save evidence while you run, for passes as well as failures: command transcripts with exit codes, HTTP status/headers/body, screenshots, console errors. Every scenario names its artifact path AND quotes the decisive evidence inline: your scratch directory is private to this run, so the parent usually cannot open the path. For a screenshot, describe exactly what it shows.
75
+ 5. **Judge.** Compare observed against expected that you wrote BEFORE running. Classify each failure: product defect, harness/setup failure, or environment limitation. A failure you caused (wrong command, missing env) is a harness issue to fix and retry, not a product bug.
76
+ 6. **Reproduce.** Re-run a failing scenario once to confirm it is not flaky; keep both attempts in the record.
77
+
78
+ ## Verdict rules
79
+
80
+ - \`pass\` — every required scenario was executed and passed with an artifact. "The agent finished" is not a pass.
81
+ - \`fail\` — at least one required scenario failed on a confirmed product defect.
82
+ - \`inconclusive\` — required scenarios could not run (blocked setup, missing credentials, unreachable target) and nothing failed. Name exactly what blocked you and what the caller must provide.
83
+
84
+ Per-scenario verdicts are \`pass\`, \`fail\`, \`blocked\`, or \`not-run\`. A scenario without an artifact cannot be \`pass\`.
85
+
86
+ ## Output discipline
87
+
88
+ End every response with exactly one \`<qa>\` block in this structure:
89
+
90
+ <qa>
91
+ <summary>
92
+ [Two or three sentences: the verdict and the one or two facts that justify it. No process narration.]
93
+ </summary>
94
+ <target>
95
+ [What was tested: repo + SHA, URL, or binary + version; environment; install/start commands.]
96
+ </target>
97
+ <scenarios>
98
+ <scenario id="S1" verdict="pass|fail|blocked|not-run">
99
+ <action>Exact command or UI steps.</action>
100
+ <expected>What should happen, written before running.</expected>
101
+ <observed>What actually happened.</observed>
102
+ <artifact>The decisive excerpt inline (exit code, status line, error text, or what the screenshot shows), plus its /tmp/qa/<run-id>/... path.</artifact>
103
+ </scenario>
104
+ </scenarios>
105
+ <findings>
106
+ <finding severity="blocker|concern|nit" scenario="S1">
107
+ <issue>One-sentence statement of the defect.</issue>
108
+ <repro>Minimal steps to reproduce.</repro>
109
+ <evidence>Expected vs actual, with the artifact path or excerpt.</evidence>
110
+ </finding>
111
+ </findings>
112
+ <coverage_gaps>
113
+ [What was NOT tested and why. Empty only when everything requested was covered.]
114
+ </coverage_gaps>
115
+ <verdict>pass | fail | inconclusive</verdict>
116
+ </qa>
117
+
118
+ You have one shot. The parent receives your final message verbatim — make it complete and self-contained.`
119
+
120
+ export const qaPayloadSchema = z
121
+ .object({
122
+ requestId: z.string().optional(),
123
+ prompt: z.string().optional(),
124
+ description: z.string().optional(),
125
+ })
126
+ .passthrough()
127
+
128
+ export type QaPayload = z.infer<typeof qaPayloadSchema>
129
+
130
+ export function createQaSubagent(options?: { resolveTokenForRepo: ResolveGithubTokenForRepo }): Subagent<QaPayload> {
131
+ const loadSkillTool = createLoadSkillTool({
132
+ skills: QA_SKILLS,
133
+ description: `Load a QA playbook by name. Each skill explains how to set up, exercise, and capture evidence for one kind of surface. Call this BEFORE planning scenarios.
134
+
135
+ Available skills:
136
+ ${QA_SKILLS.map((s) => `- \`${s.name}\` — ${s.description}`).join('\n')}
137
+
138
+ If none fit, load \`qa-general\`.`,
139
+ })
140
+
141
+ return {
142
+ systemPrompt: QA_SYSTEM_PROMPT,
143
+ profile: 'deep',
144
+ tools: [readTool, grepTool, findTool, lsTool, bashTool, writeTool, webSearchTool, webFetchTool],
145
+ customTools: [
146
+ loadSkillTool,
147
+ ...(options === undefined ? [] : [createReviewerCheckoutTool(options.resolveTokenForRepo)]),
148
+ ],
149
+ payloadSchema: qaPayloadSchema,
150
+ visibility: 'public',
151
+ rosterDescription:
152
+ 'hands-on QA tester: builds and runs a PR, repo, web app, CLI, or API through its real surface and returns an evidence-backed pass/fail/inconclusive verdict; never edits the product or posts — you act on its findings',
153
+ timeoutMs: QA_SPAWN_TIMEOUT_MS,
154
+ inFlightKey: (payload) => payload?.requestId ?? `anon-${Date.now()}-${Math.random().toString(36).slice(2, 8)}`,
155
+ toolResultBudget: {
156
+ maxTotalBytes: 1_000_000,
157
+ toolNames: ['read', 'grep', 'find', 'ls', 'bash', 'write', 'web_search', 'web_fetch', 'load_skill'],
158
+ },
159
+ }
160
+ }
@@ -0,0 +1,50 @@
1
+ import type { LoadableSkill } from '@/plugin'
2
+
3
+ export const CLI_QA_SKILL: LoadableSkill = {
4
+ name: 'qa-cli',
5
+ description:
6
+ 'QA a command-line tool or terminal UI: run the real binary with representative, edge, and invalid arguments, capture stdout/stderr and exit codes, and drive interactive prompts or TUIs through tmux with captured panes.',
7
+ content: `# qa-cli
8
+
9
+ You are QAing a command-line tool or terminal UI. Apply this on top of the QA loop and the \`<qa>\` output contract.
10
+
11
+ ## Run the real binary
12
+
13
+ - Build it from the project's documented command, then invoke the built artifact (or the documented \`bun run\`/\`npx\`-style entry) exactly as a user would.
14
+ - Record the version (\`--version\`) in \`<target>\`.
15
+ - Run each scenario from a clean scratch working directory so leftover state cannot mask a bug. Use a scratch \`HOME\` (\`HOME=/tmp/qa/<run-id>/home\`) when the tool writes config there.
16
+
17
+ ## Capture
18
+
19
+ For every invocation keep the exact command, stdout, stderr, and exit code. Save long output to \`/tmp/qa/<run-id>/<scenario>.log\` and quote the decisive lines in \`<observed>\`.
20
+
21
+ ## What to cover
22
+
23
+ 1. **Happy path** of each command the request names, with the resulting files or state checked (\`ls\`, \`cat\`, a second command that reads it back).
24
+ 2. **Help and discovery**: \`--help\` lists the new or changed flag; the description matches behavior.
25
+ 3. **Invalid input**: unknown flag, missing required argument, bad value. Expect a non-zero exit and a message that names the problem. A stack trace or exit 0 on failure is a finding.
26
+ 4. **Edge input**: empty file, very long argument, paths with spaces, non-ASCII (Korean or Japanese) text.
27
+ 5. **Idempotency and re-runs** where it matters: running the same command twice does not corrupt state.
28
+ 6. **Non-TTY behavior**: output piped to a file has no broken escape codes when the tool claims to support that.
29
+
30
+ ## Interactive prompts and TUIs
31
+
32
+ Drive them in a real pseudo-terminal with tmux, scripted so the whole session (start, every step, capture, kill) runs inside ONE \`bash\` call; a tmux server started in one call does not survive to the next:
33
+
34
+ - \`tmux new-session -d -s qa -x 120 -y 40 '<command>'\`
35
+ - \`tmux send-keys -t qa 'input' Enter\` (and \`Down\`, \`C-c\`, etc.)
36
+ - \`tmux capture-pane -t qa -p > /tmp/qa/<run-id>/<scenario>-<step>.txt\` after each step
37
+ - Wait for a prompt by polling \`capture-pane\` for the expected text with a bounded loop, never a fixed sleep.
38
+ - \`tmux kill-session -t qa\` when done.
39
+
40
+ Check that lines fit the terminal width, borders align, and wide (CJK) characters do not shift columns. If tmux is not installed, fall back to non-interactive flags when the tool has them and record the TUI as a coverage gap.
41
+
42
+ ## Severity
43
+
44
+ - **blocker** — a named command fails or produces wrong output, data is destroyed, a crash on common input.
45
+ - **concern** — bad error message or exit code on invalid input, a broken edge case, help text that lies.
46
+ - **nit** — formatting.
47
+
48
+ Return everything inside the \`<qa>\` block.
49
+ `,
50
+ }
@@ -0,0 +1,30 @@
1
+ import type { LoadableSkill } from '@/plugin'
2
+
3
+ export const GENERAL_QA_SKILL: LoadableSkill = {
4
+ name: 'qa-general',
5
+ description:
6
+ 'Fallback for QA targets that fit no specific surface: a library or SDK, a data pipeline, a config or migration, a generated artifact, or a mixed system. Pick the channel that faithfully exercises the behavior and capture its output as evidence.',
7
+ content: `# general
8
+
9
+ You are QAing something that is not clearly a PR, a web app, a CLI, or an API, or that mixes them. Apply this on top of the QA loop and the \`<qa>\` output contract.
10
+
11
+ ## Pick the channel that exercises the real behavior
12
+
13
+ - **Library or SDK** — write a small script in \`/tmp/qa/<run-id>/\` that imports the public API exactly as a consumer would and exercises the behavior; keep its transcript.
14
+ - **Data pipeline, migration, or generator** — run it on a representative input in scratch space and inspect the produced artifact itself: diff it, dump it, count rows, validate it against its schema.
15
+ - **Config change** — start the system with the config and observe the behavior it is supposed to change, then with the previous value to show the difference.
16
+ - **Document or report output** (PDF, HTML, spreadsheet) — render it and inspect the result; extracted text alone does not prove layout.
17
+
18
+ If the target spans surfaces, load the matching surface skill as well.
19
+
20
+ ## What to cover
21
+
22
+ The happy path the request names, the riskiest edge (empty, boundary, malformed input), and one adjacent behavior that could regress. Every stated success criterion gets its own scenario with an artifact.
23
+
24
+ ## When you cannot exercise it
25
+
26
+ If the behavior needs something you do not have (credentials, hardware, a paid service, a specific OS), do not substitute reading code for running it. Mark those scenarios \`blocked\`, say exactly what is missing, and return \`inconclusive\` unless something else failed.
27
+
28
+ Return everything inside the \`<qa>\` block.
29
+ `,
30
+ }