niceeval 0.17.1 → 0.19.0

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 (692) hide show
  1. package/INDEX.md +5 -2
  2. package/dist/adapter-usage.cjs.map +1 -1
  3. package/dist/adapter-usage.d.cts +1 -1
  4. package/dist/adapter-usage.d.mts +1 -1
  5. package/dist/adapter-usage.d.ts +1 -1
  6. package/dist/adapter.cjs +8 -3
  7. package/dist/adapter.cjs.map +1 -1
  8. package/dist/adapter.d.cts +26 -15
  9. package/dist/adapter.d.mts +26 -15
  10. package/dist/adapter.d.ts +26 -15
  11. package/dist/agents/{effect-runtime.cjs → agent-effects.cjs} +12 -18
  12. package/dist/agents/agent-effects.cjs.map +1 -0
  13. package/dist/agents/agent-effects.d.cts +23 -0
  14. package/dist/agents/agent-effects.d.mts +23 -0
  15. package/dist/agents/agent-effects.d.ts +23 -0
  16. package/dist/agents/agent-effects.js +16 -0
  17. package/dist/agents/agent-effects.js.map +1 -0
  18. package/dist/agents/ai-sdk.cjs +1 -1
  19. package/dist/agents/ai-sdk.cjs.map +1 -1
  20. package/dist/agents/bub.cjs +31 -18
  21. package/dist/agents/bub.cjs.map +1 -1
  22. package/dist/agents/bub.d.cts +9 -12
  23. package/dist/agents/bub.d.mts +9 -12
  24. package/dist/agents/bub.d.ts +9 -12
  25. package/dist/agents/claude-code-native.cjs +2 -2
  26. package/dist/agents/claude-code-native.cjs.map +1 -1
  27. package/dist/agents/claude-code.cjs +17 -11
  28. package/dist/agents/claude-code.cjs.map +1 -1
  29. package/dist/agents/claude-code.d.cts +11 -11
  30. package/dist/agents/claude-code.d.mts +11 -11
  31. package/dist/agents/claude-code.d.ts +11 -11
  32. package/dist/agents/codex-app-server.cjs +2 -2
  33. package/dist/agents/codex-app-server.cjs.map +1 -1
  34. package/dist/agents/codex.cjs +18 -12
  35. package/dist/agents/codex.cjs.map +1 -1
  36. package/dist/agents/codex.d.cts +12 -12
  37. package/dist/agents/codex.d.mts +12 -12
  38. package/dist/agents/codex.d.ts +12 -12
  39. package/dist/agents/deepseek-harness.cjs +2 -2
  40. package/dist/agents/deepseek-harness.cjs.map +1 -1
  41. package/dist/agents/hermes.cjs +3 -3
  42. package/dist/agents/hermes.cjs.map +1 -1
  43. package/dist/agents/index.cjs +3 -3
  44. package/dist/agents/index.cjs.map +1 -1
  45. package/dist/agents/index.d.cts +3 -3
  46. package/dist/agents/index.d.mts +3 -3
  47. package/dist/agents/index.d.ts +3 -3
  48. package/dist/agents/index.js +2 -2
  49. package/dist/agents/index.mjs +2 -2
  50. package/dist/agents/mcp.cjs.map +1 -1
  51. package/dist/agents/mcp.d.cts +2 -2
  52. package/dist/agents/mcp.d.mts +2 -2
  53. package/dist/agents/mcp.d.ts +2 -2
  54. package/dist/agents/omp.cjs +2 -2
  55. package/dist/agents/omp.cjs.map +1 -1
  56. package/dist/agents/openai-compat.cjs +2 -2
  57. package/dist/agents/openai-compat.cjs.map +1 -1
  58. package/dist/agents/openclaw.cjs +3 -3
  59. package/dist/agents/openclaw.cjs.map +1 -1
  60. package/dist/agents/opencode.cjs +16 -4
  61. package/dist/agents/opencode.cjs.map +1 -1
  62. package/dist/agents/plugin-extensions.cjs +71 -0
  63. package/dist/agents/plugin-extensions.cjs.map +1 -0
  64. package/dist/agents/plugin-extensions.d.cts +26 -0
  65. package/dist/agents/plugin-extensions.d.mts +26 -0
  66. package/dist/agents/plugin-extensions.d.ts +26 -0
  67. package/dist/agents/plugin-extensions.js +8 -0
  68. package/dist/agents/plugin-extensions.js.map +1 -0
  69. package/dist/agents/post-setup.cjs +14 -3
  70. package/dist/agents/post-setup.cjs.map +1 -1
  71. package/dist/agents/post-setup.d.cts +4 -2
  72. package/dist/agents/post-setup.d.mts +4 -2
  73. package/dist/agents/post-setup.d.ts +4 -2
  74. package/dist/agents/post-setup.js +8 -6
  75. package/dist/agents/sdk-streams.cjs +3 -3
  76. package/dist/agents/sdk-streams.cjs.map +1 -1
  77. package/dist/agents/{session-slot.cjs → session-key.cjs} +4 -4
  78. package/dist/agents/session-key.cjs.map +1 -0
  79. package/dist/agents/session-key.d.cts +12 -0
  80. package/dist/agents/session-key.d.mts +12 -0
  81. package/dist/agents/session-key.d.ts +12 -0
  82. package/dist/agents/session-key.js +6 -0
  83. package/dist/agents/session-key.js.map +1 -0
  84. package/dist/agents/shared.cjs +2 -2
  85. package/dist/agents/shared.cjs.map +1 -1
  86. package/dist/agents/shared.d.cts +4 -4
  87. package/dist/agents/shared.d.mts +4 -4
  88. package/dist/agents/shared.d.ts +4 -4
  89. package/dist/agents/skills.cjs.map +1 -1
  90. package/dist/agents/skills.d.cts +3 -3
  91. package/dist/agents/skills.d.mts +3 -3
  92. package/dist/agents/skills.d.ts +3 -3
  93. package/dist/agents/types.cjs.map +1 -1
  94. package/dist/agents/types.d.cts +55 -54
  95. package/dist/agents/types.d.mts +55 -54
  96. package/dist/agents/types.d.ts +55 -54
  97. package/dist/agents/ui-message-stream.cjs +4 -4
  98. package/dist/agents/ui-message-stream.cjs.map +1 -1
  99. package/dist/agents/unrecognized-events.cjs +32 -0
  100. package/dist/agents/unrecognized-events.cjs.map +1 -0
  101. package/dist/agents/unrecognized-events.d.cts +8 -0
  102. package/dist/agents/unrecognized-events.d.mts +8 -0
  103. package/dist/agents/unrecognized-events.d.ts +8 -0
  104. package/dist/agents/unrecognized-events.js +6 -0
  105. package/dist/agents/unrecognized-events.js.map +1 -0
  106. package/dist/assertions/api.cjs.map +1 -1
  107. package/dist/assertions/api.d.cts +3 -0
  108. package/dist/assertions/api.d.mts +3 -0
  109. package/dist/assertions/api.d.ts +3 -0
  110. package/dist/assertions/collection.cjs +1 -1
  111. package/dist/assertions/collection.cjs.map +1 -1
  112. package/dist/assertions/judge.cjs +1 -1
  113. package/dist/assertions/judge.cjs.map +1 -1
  114. package/dist/assertions/match.cjs +52 -21
  115. package/dist/assertions/match.cjs.map +1 -1
  116. package/dist/assertions/match.d.cts +16 -5
  117. package/dist/assertions/match.d.mts +16 -5
  118. package/dist/assertions/match.d.ts +16 -5
  119. package/dist/assertions/match.js +96 -94
  120. package/dist/assertions/measurements.cjs +61 -0
  121. package/dist/assertions/measurements.cjs.map +1 -0
  122. package/dist/assertions/measurements.d.cts +24 -0
  123. package/dist/assertions/measurements.d.mts +24 -0
  124. package/dist/assertions/measurements.d.ts +24 -0
  125. package/dist/assertions/measurements.js +8 -0
  126. package/dist/assertions/measurements.js.map +1 -0
  127. package/dist/assertions/numeric.cjs +2 -0
  128. package/dist/assertions/numeric.cjs.map +1 -1
  129. package/dist/assertions/numeric.d.cts +1 -1
  130. package/dist/assertions/numeric.d.mts +1 -1
  131. package/dist/assertions/numeric.d.ts +1 -1
  132. package/dist/assertions/quantity.cjs +42 -0
  133. package/dist/assertions/quantity.cjs.map +1 -0
  134. package/dist/assertions/quantity.d.cts +18 -0
  135. package/dist/assertions/quantity.d.mts +18 -0
  136. package/dist/assertions/quantity.d.ts +18 -0
  137. package/dist/assertions/quantity.js +22 -0
  138. package/dist/assertions/quantity.js.map +1 -0
  139. package/dist/assertions/record/attachment.cjs +2 -0
  140. package/dist/assertions/record/attachment.cjs.map +1 -1
  141. package/dist/assertions/record/codec.cjs +5 -1
  142. package/dist/assertions/record/codec.cjs.map +1 -1
  143. package/dist/assertions/record/model.cjs.map +1 -1
  144. package/dist/assertions/record/model.d.cts +3 -1
  145. package/dist/assertions/record/model.d.mts +3 -1
  146. package/dist/assertions/record/model.d.ts +3 -1
  147. package/dist/assertions/runtime.cjs +11 -6
  148. package/dist/assertions/runtime.cjs.map +1 -1
  149. package/dist/cli/program.cjs +3 -0
  150. package/dist/cli/program.cjs.map +1 -1
  151. package/dist/context/assert-first.cjs +6 -4
  152. package/dist/context/assert-first.cjs.map +1 -1
  153. package/dist/context/assert-first.d.cts +9 -3
  154. package/dist/context/assert-first.d.mts +9 -3
  155. package/dist/context/assert-first.d.ts +9 -3
  156. package/dist/context/notes.cjs +47 -0
  157. package/dist/context/notes.cjs.map +1 -0
  158. package/dist/context/notes.d.cts +10 -0
  159. package/dist/context/notes.d.mts +10 -0
  160. package/dist/context/notes.d.ts +10 -0
  161. package/dist/context/notes.js +6 -0
  162. package/dist/context/notes.js.map +1 -0
  163. package/dist/context/send-failures.cjs +1 -7
  164. package/dist/context/send-failures.cjs.map +1 -1
  165. package/dist/context/send-failures.d.cts +2 -3
  166. package/dist/context/send-failures.d.mts +2 -3
  167. package/dist/context/send-failures.d.ts +2 -3
  168. package/dist/context/send-retry.cjs +0 -1
  169. package/dist/context/send-retry.cjs.map +1 -1
  170. package/dist/context/send-retry.d.cts +1 -2
  171. package/dist/context/send-retry.d.mts +1 -2
  172. package/dist/context/send-retry.d.ts +1 -2
  173. package/dist/context/session.cjs +10 -17
  174. package/dist/context/session.cjs.map +1 -1
  175. package/dist/context/session.d.cts +4 -10
  176. package/dist/context/session.d.mts +4 -10
  177. package/dist/context/session.d.ts +4 -10
  178. package/dist/define.cjs +89 -64
  179. package/dist/define.cjs.map +1 -1
  180. package/dist/define.d.cts +3 -3
  181. package/dist/define.d.mts +3 -3
  182. package/dist/define.d.ts +3 -3
  183. package/dist/eval/cli/contribution.cjs +1 -1
  184. package/dist/eval/cli/contribution.cjs.map +1 -1
  185. package/dist/eval/host/index.cjs +1 -1
  186. package/dist/eval/host/index.cjs.map +1 -1
  187. package/dist/eval/record/evaluation.cjs +1 -1
  188. package/dist/eval/record/evaluation.cjs.map +1 -1
  189. package/dist/eval/record/membership-provenance.d.cts +3 -3
  190. package/dist/eval/record/membership-provenance.d.mts +3 -3
  191. package/dist/eval/record/membership-provenance.d.ts +3 -3
  192. package/dist/eval/record/score.cjs +2 -2
  193. package/dist/eval/record/score.cjs.map +1 -1
  194. package/dist/eval/record/sealed-assertion.cjs +1 -1
  195. package/dist/eval/record/sealed-assertion.cjs.map +1 -1
  196. package/dist/eval/record/verdict.cjs +2 -2
  197. package/dist/eval/record/verdict.cjs.map +1 -1
  198. package/dist/expect/index.cjs +13 -1
  199. package/dist/expect/index.cjs.map +1 -1
  200. package/dist/expect/index.d.cts +5 -1
  201. package/dist/expect/index.d.mts +5 -1
  202. package/dist/expect/index.d.ts +5 -1
  203. package/dist/expect/index.js +92 -72
  204. package/dist/expect/index.mjs +92 -72
  205. package/dist/experiment/host/cli/contribution.cjs +52 -39
  206. package/dist/experiment/host/cli/contribution.cjs.map +1 -1
  207. package/dist/experiment/host/cli/contribution.d.cts +18 -4
  208. package/dist/experiment/host/cli/contribution.d.mts +18 -4
  209. package/dist/experiment/host/cli/contribution.d.ts +18 -4
  210. package/dist/experiment/host/cli/contribution.js +2 -2
  211. package/dist/experiment/host/cli/debug-protocol.cjs +2 -2
  212. package/dist/experiment/host/cli/debug-protocol.cjs.map +1 -1
  213. package/dist/experiment/host/cli/debug-protocol.d.cts +2 -2
  214. package/dist/experiment/host/cli/debug-protocol.d.mts +2 -2
  215. package/dist/experiment/host/cli/debug-protocol.d.ts +2 -2
  216. package/dist/experiment/host/cli/output-protocol.cjs +1 -1
  217. package/dist/experiment/host/cli/output-protocol.cjs.map +1 -1
  218. package/dist/experiment/host/cli/output-protocol.d.cts +2 -2
  219. package/dist/experiment/host/cli/output-protocol.d.mts +2 -2
  220. package/dist/experiment/host/cli/output-protocol.d.ts +2 -2
  221. package/dist/experiment/host/current.cjs +2 -2
  222. package/dist/experiment/host/current.cjs.map +1 -1
  223. package/dist/experiment/host/index.cjs +1 -1
  224. package/dist/experiment/host/index.cjs.map +1 -1
  225. package/dist/experiment/host/operations.cjs +24 -23
  226. package/dist/experiment/host/operations.cjs.map +1 -1
  227. package/dist/experiment/host/runtime.cjs +11 -10
  228. package/dist/experiment/host/runtime.cjs.map +1 -1
  229. package/dist/experiment/host/runtime.d.cts +3 -3
  230. package/dist/experiment/host/runtime.d.mts +3 -3
  231. package/dist/experiment/host/runtime.d.ts +3 -3
  232. package/dist/experiment/host/teardown.cjs +4 -3
  233. package/dist/experiment/host/teardown.cjs.map +1 -1
  234. package/dist/experiment/host/teardown.d.cts +3 -3
  235. package/dist/experiment/host/teardown.d.mts +3 -3
  236. package/dist/experiment/host/teardown.d.ts +3 -3
  237. package/dist/experiment/host/types.cjs.map +1 -1
  238. package/dist/experiment/host/types.d.cts +19 -19
  239. package/dist/experiment/host/types.d.mts +19 -19
  240. package/dist/experiment/host/types.d.ts +19 -19
  241. package/dist/index.cjs +3 -19
  242. package/dist/index.cjs.map +1 -1
  243. package/dist/index.d.cts +4 -8
  244. package/dist/index.d.mts +4 -8
  245. package/dist/index.d.ts +4 -8
  246. package/dist/index.js +12 -34
  247. package/dist/index.mjs +12 -34
  248. package/dist/inspection/assertions.cjs +1 -1
  249. package/dist/inspection/assertions.cjs.map +1 -1
  250. package/dist/inspection/cli/contribution.cjs +8 -7
  251. package/dist/inspection/cli/contribution.cjs.map +1 -1
  252. package/dist/inspection/facts.cjs +4 -4
  253. package/dist/inspection/facts.cjs.map +1 -1
  254. package/dist/inspection/index.cjs +1 -5
  255. package/dist/inspection/index.cjs.map +1 -1
  256. package/dist/inspection/index.d.cts +1 -2
  257. package/dist/inspection/index.d.mts +1 -2
  258. package/dist/inspection/index.d.ts +1 -2
  259. package/dist/inspection/index.js +20 -26
  260. package/dist/inspection/lifecycle.cjs +7 -0
  261. package/dist/inspection/lifecycle.cjs.map +1 -0
  262. package/dist/inspection/lifecycle.d.cts +4 -0
  263. package/dist/inspection/lifecycle.d.mts +4 -0
  264. package/dist/inspection/lifecycle.d.ts +4 -0
  265. package/dist/inspection/lifecycle.js +6 -0
  266. package/dist/inspection/lifecycle.js.map +1 -0
  267. package/dist/inspection/notes.cjs +40 -0
  268. package/dist/inspection/notes.cjs.map +1 -0
  269. package/dist/inspection/notes.d.cts +12 -0
  270. package/dist/inspection/notes.d.mts +12 -0
  271. package/dist/inspection/notes.d.ts +12 -0
  272. package/dist/inspection/notes.js +8 -0
  273. package/dist/inspection/notes.js.map +1 -0
  274. package/dist/inspection/overview.cjs +12 -15
  275. package/dist/inspection/overview.cjs.map +1 -1
  276. package/dist/inspection/overview.d.cts +8 -6
  277. package/dist/inspection/overview.d.mts +8 -6
  278. package/dist/inspection/overview.d.ts +8 -6
  279. package/dist/inspection/planned-attempt.cjs +45 -0
  280. package/dist/inspection/planned-attempt.cjs.map +1 -0
  281. package/dist/inspection/planned-attempt.d.cts +109 -0
  282. package/dist/inspection/planned-attempt.d.mts +109 -0
  283. package/dist/inspection/planned-attempt.d.ts +109 -0
  284. package/dist/inspection/planned-attempt.js +24 -0
  285. package/dist/inspection/planned-attempt.js.map +1 -0
  286. package/dist/inspection/project-result.cjs +5 -5
  287. package/dist/inspection/project-result.cjs.map +1 -1
  288. package/dist/inspection/project-result.d.cts +58 -57
  289. package/dist/inspection/project-result.d.mts +58 -57
  290. package/dist/inspection/project-result.d.ts +58 -57
  291. package/dist/inspection/project-result.js +4 -4
  292. package/dist/inspection/project.cjs +3 -3
  293. package/dist/inspection/project.cjs.map +1 -1
  294. package/dist/inspection/protocol-values.cjs +1 -1
  295. package/dist/inspection/protocol-values.cjs.map +1 -1
  296. package/dist/inspection/protocol-values.d.cts +1 -1
  297. package/dist/inspection/protocol-values.d.mts +1 -1
  298. package/dist/inspection/protocol-values.d.ts +1 -1
  299. package/dist/inspection/protocol.cjs +8 -8
  300. package/dist/inspection/protocol.cjs.map +1 -1
  301. package/dist/inspection/protocol.d.cts +237 -219
  302. package/dist/inspection/protocol.d.mts +237 -219
  303. package/dist/inspection/protocol.d.ts +237 -219
  304. package/dist/inspection/public.cjs +1 -5
  305. package/dist/inspection/public.cjs.map +1 -1
  306. package/dist/inspection/public.d.cts +1 -2
  307. package/dist/inspection/public.d.mts +1 -2
  308. package/dist/inspection/public.d.ts +1 -2
  309. package/dist/inspection/public.js +14 -20
  310. package/dist/inspection/public.mjs +166 -129
  311. package/dist/inspection/query.cjs +1 -1
  312. package/dist/inspection/query.cjs.map +1 -1
  313. package/dist/inspection/results.cjs +61 -52
  314. package/dist/inspection/results.cjs.map +1 -1
  315. package/dist/inspection/results.d.cts +157 -140
  316. package/dist/inspection/results.d.mts +157 -140
  317. package/dist/inspection/results.d.ts +157 -140
  318. package/dist/inspection/select.cjs +33 -28
  319. package/dist/inspection/select.cjs.map +1 -1
  320. package/dist/inspection/source.cjs +1 -1
  321. package/dist/inspection/source.cjs.map +1 -1
  322. package/dist/inspection/sources.cjs +2 -2
  323. package/dist/inspection/sources.cjs.map +1 -1
  324. package/dist/inspection/trace.cjs +9 -8
  325. package/dist/inspection/trace.cjs.map +1 -1
  326. package/dist/inspection/trace.d.cts +4 -4
  327. package/dist/inspection/trace.d.mts +4 -4
  328. package/dist/inspection/trace.d.ts +4 -4
  329. package/dist/loaders/index.cjs +7 -279
  330. package/dist/loaders/index.cjs.map +1 -1
  331. package/dist/loaders/index.d.cts +2 -95
  332. package/dist/loaders/index.d.mts +2 -95
  333. package/dist/loaders/index.d.ts +2 -95
  334. package/dist/loaders/index.js +10 -12
  335. package/dist/loaders/index.mjs +10 -12
  336. package/dist/loaders/runtime.cjs +283 -0
  337. package/dist/loaders/runtime.cjs.map +1 -0
  338. package/dist/loaders/runtime.d.cts +95 -0
  339. package/dist/loaders/runtime.d.mts +95 -0
  340. package/dist/loaders/runtime.d.ts +95 -0
  341. package/dist/loaders/runtime.js +16 -0
  342. package/dist/loaders/runtime.js.map +1 -0
  343. package/dist/migrations/judge-provider.md +6 -6
  344. package/dist/{model-slots.cjs → model-roles.cjs} +26 -7
  345. package/dist/model-roles.cjs.map +1 -0
  346. package/dist/model-roles.d.cts +26 -0
  347. package/dist/model-roles.d.mts +26 -0
  348. package/dist/model-roles.d.ts +26 -0
  349. package/dist/model-roles.js +14 -0
  350. package/dist/model-roles.js.map +1 -0
  351. package/dist/o11y/adapter-usage-projection.cjs +9 -8
  352. package/dist/o11y/adapter-usage-projection.cjs.map +1 -1
  353. package/dist/o11y/adapter-usage-projection.d.cts +23 -15
  354. package/dist/o11y/adapter-usage-projection.d.mts +23 -15
  355. package/dist/o11y/adapter-usage-projection.d.ts +23 -15
  356. package/dist/o11y/cache-read.cjs +14 -14
  357. package/dist/o11y/cache-read.cjs.map +1 -1
  358. package/dist/o11y/cache-read.d.cts +5 -5
  359. package/dist/o11y/cache-read.d.mts +5 -5
  360. package/dist/o11y/cache-read.d.ts +5 -5
  361. package/dist/o11y/derive.cjs.map +1 -1
  362. package/dist/o11y/derive.d.cts +2 -2
  363. package/dist/o11y/derive.d.mts +2 -2
  364. package/dist/o11y/derive.d.ts +2 -2
  365. package/dist/o11y/eval-usage.cjs +9 -9
  366. package/dist/o11y/eval-usage.cjs.map +1 -1
  367. package/dist/o11y/eval-usage.d.cts +7 -7
  368. package/dist/o11y/eval-usage.d.mts +7 -7
  369. package/dist/o11y/eval-usage.d.ts +7 -7
  370. package/dist/o11y/parsers/bub.cjs +57 -81
  371. package/dist/o11y/parsers/bub.cjs.map +1 -1
  372. package/dist/o11y/parsers/index.cjs.map +1 -1
  373. package/dist/o11y/parsers/index.d.cts +2 -0
  374. package/dist/o11y/parsers/index.d.mts +2 -0
  375. package/dist/o11y/parsers/index.d.ts +2 -0
  376. package/dist/o11y/parsers/opencode.cjs +129 -217
  377. package/dist/o11y/parsers/opencode.cjs.map +1 -1
  378. package/dist/o11y/parsers/opencode.d.cts +2 -5
  379. package/dist/o11y/parsers/opencode.d.mts +2 -5
  380. package/dist/o11y/parsers/opencode.d.ts +2 -5
  381. package/dist/o11y/types.cjs.map +1 -1
  382. package/dist/o11y/types.d.cts +1 -1
  383. package/dist/o11y/types.d.mts +1 -1
  384. package/dist/o11y/types.d.ts +1 -1
  385. package/dist/plugin/contracts.cjs +170 -120
  386. package/dist/plugin/contracts.cjs.map +1 -1
  387. package/dist/plugin/contracts.d.cts +102 -90
  388. package/dist/plugin/contracts.d.mts +102 -90
  389. package/dist/plugin/contracts.d.ts +102 -90
  390. package/dist/plugin/contracts.js +16 -12
  391. package/dist/plugin/index.cjs.map +1 -1
  392. package/dist/plugin/index.d.cts +1 -1
  393. package/dist/plugin/index.d.mts +1 -1
  394. package/dist/plugin/index.d.ts +1 -1
  395. package/dist/plugin/link.cjs +69 -64
  396. package/dist/plugin/link.cjs.map +1 -1
  397. package/dist/plugin/link.d.cts +5 -3
  398. package/dist/plugin/link.d.mts +5 -3
  399. package/dist/plugin/link.d.ts +5 -3
  400. package/dist/plugin/link.js +12 -8
  401. package/dist/record/codec/seal-manifest.d.cts +3 -3
  402. package/dist/record/codec/seal-manifest.d.mts +3 -3
  403. package/dist/record/codec/seal-manifest.d.ts +3 -3
  404. package/dist/record/family/attempt-note/definition.cjs +12 -0
  405. package/dist/record/family/attempt-note/definition.cjs.map +1 -0
  406. package/dist/record/family/attempt-note/definition.d.cts +6 -0
  407. package/dist/record/family/attempt-note/definition.d.mts +6 -0
  408. package/dist/record/family/attempt-note/definition.d.ts +6 -0
  409. package/dist/record/family/attempt-note/definition.js +6 -0
  410. package/dist/record/family/attempt-note/definition.js.map +1 -0
  411. package/dist/record/family/attempt-note/persistence.cjs +10 -0
  412. package/dist/record/family/attempt-note/persistence.cjs.map +1 -0
  413. package/dist/record/family/attempt-note/persistence.d.cts +6 -0
  414. package/dist/record/family/attempt-note/persistence.d.mts +6 -0
  415. package/dist/record/family/attempt-note/persistence.d.ts +6 -0
  416. package/dist/record/family/attempt-note/persistence.js +6 -0
  417. package/dist/record/family/attempt-note/persistence.js.map +1 -0
  418. package/dist/record/family/attempt-note/schema.cjs +20 -0
  419. package/dist/record/family/attempt-note/schema.cjs.map +1 -0
  420. package/dist/record/family/attempt-note/schema.d.cts +21 -0
  421. package/dist/record/family/attempt-note/schema.d.mts +21 -0
  422. package/dist/record/family/attempt-note/schema.d.ts +21 -0
  423. package/dist/record/family/attempt-note/schema.js +10 -0
  424. package/dist/record/family/attempt-note/schema.js.map +1 -0
  425. package/dist/record/family/catalog.cjs +32 -29
  426. package/dist/record/family/catalog.cjs.map +1 -1
  427. package/dist/record/family/catalog.d.cts +7 -2
  428. package/dist/record/family/catalog.d.mts +7 -2
  429. package/dist/record/family/catalog.d.ts +7 -2
  430. package/dist/record/family/current.cjs +48 -45
  431. package/dist/record/family/current.cjs.map +1 -1
  432. package/dist/record/family/current.d.cts +15 -0
  433. package/dist/record/family/current.d.mts +15 -0
  434. package/dist/record/family/current.d.ts +15 -0
  435. package/dist/record/model/index.js +66 -64
  436. package/dist/record/model/run-context.cjs +9 -3
  437. package/dist/record/model/run-context.cjs.map +1 -1
  438. package/dist/record/model/run-context.d.cts +9 -2
  439. package/dist/record/model/run-context.d.mts +9 -2
  440. package/dist/record/model/run-context.d.ts +9 -2
  441. package/dist/record/model/run-context.js +8 -6
  442. package/dist/record/model/seal-manifest.d.cts +1 -1
  443. package/dist/record/model/seal-manifest.d.mts +1 -1
  444. package/dist/record/model/seal-manifest.d.ts +1 -1
  445. package/dist/run/cli/contribution.cjs +6 -32
  446. package/dist/run/cli/contribution.cjs.map +1 -1
  447. package/dist/run/host/runtime.cjs +1 -1
  448. package/dist/run/host/runtime.cjs.map +1 -1
  449. package/dist/run/storage/sqlite.cjs +7 -7
  450. package/dist/run/storage/sqlite.cjs.map +1 -1
  451. package/dist/runner/adapter-usage.cjs +5 -3
  452. package/dist/runner/adapter-usage.cjs.map +1 -1
  453. package/dist/runner/adapter-usage.d.cts +2 -2
  454. package/dist/runner/adapter-usage.d.mts +2 -2
  455. package/dist/runner/adapter-usage.d.ts +2 -2
  456. package/dist/runner/adoption.cjs +10 -12
  457. package/dist/runner/adoption.cjs.map +1 -1
  458. package/dist/runner/attempt-notes.cjs +8 -0
  459. package/dist/runner/attempt-notes.cjs.map +1 -0
  460. package/dist/runner/attempt-notes.d.cts +4 -0
  461. package/dist/runner/attempt-notes.d.mts +4 -0
  462. package/dist/runner/attempt-notes.d.ts +4 -0
  463. package/dist/runner/attempt-notes.js +8 -0
  464. package/dist/runner/attempt-notes.js.map +1 -0
  465. package/dist/runner/attempt.cjs +61 -73
  466. package/dist/runner/attempt.cjs.map +1 -1
  467. package/dist/runner/attempt.d.cts +3 -3
  468. package/dist/runner/attempt.d.mts +3 -3
  469. package/dist/runner/attempt.d.ts +3 -3
  470. package/dist/runner/command-plan.cjs +21 -26
  471. package/dist/runner/command-plan.cjs.map +1 -1
  472. package/dist/runner/command-plan.d.cts +4 -4
  473. package/dist/runner/command-plan.d.mts +4 -4
  474. package/dist/runner/command-plan.d.ts +4 -4
  475. package/dist/runner/command-plan.js +4 -4
  476. package/dist/runner/config-identity.cjs.map +1 -1
  477. package/dist/runner/config-identity.d.cts +2 -2
  478. package/dist/runner/config-identity.d.mts +2 -2
  479. package/dist/runner/config-identity.d.ts +2 -2
  480. package/dist/runner/discover.cjs +6 -6
  481. package/dist/runner/discover.cjs.map +1 -1
  482. package/dist/runner/eval-selection.cjs +15 -1
  483. package/dist/runner/eval-selection.cjs.map +1 -1
  484. package/dist/runner/eval-selection.d.cts +4 -0
  485. package/dist/runner/eval-selection.d.mts +4 -0
  486. package/dist/runner/eval-selection.d.ts +4 -0
  487. package/dist/runner/feedback/human.cjs +1 -1
  488. package/dist/runner/feedback/human.cjs.map +1 -1
  489. package/dist/runner/feedback/json.cjs +4 -2
  490. package/dist/runner/feedback/json.cjs.map +1 -1
  491. package/dist/runner/feedback/json.d.cts +1 -1
  492. package/dist/runner/feedback/json.d.mts +1 -1
  493. package/dist/runner/feedback/json.d.ts +1 -1
  494. package/dist/runner/feedback/sink.cjs +1 -1
  495. package/dist/runner/feedback/sink.cjs.map +1 -1
  496. package/dist/runner/feedback/sink.d.cts +1 -1
  497. package/dist/runner/feedback/sink.d.mts +1 -1
  498. package/dist/runner/feedback/sink.d.ts +1 -1
  499. package/dist/runner/fingerprint.cjs +1 -1
  500. package/dist/runner/fingerprint.cjs.map +1 -1
  501. package/dist/runner/lock.cjs +1 -3
  502. package/dist/runner/lock.cjs.map +1 -1
  503. package/dist/runner/lock.d.cts +0 -2
  504. package/dist/runner/lock.d.mts +0 -2
  505. package/dist/runner/lock.d.ts +0 -2
  506. package/dist/runner/lock.js +14 -16
  507. package/dist/runner/project-target.cjs.map +1 -1
  508. package/dist/runner/project-target.d.cts +2 -2
  509. package/dist/runner/project-target.d.mts +2 -2
  510. package/dist/runner/project-target.d.ts +2 -2
  511. package/dist/runner/record/attachments.cjs +2 -2
  512. package/dist/runner/record/attachments.cjs.map +1 -1
  513. package/dist/runner/record/planning.cjs +8 -2
  514. package/dist/runner/record/planning.cjs.map +1 -1
  515. package/dist/runner/record.cjs +2 -0
  516. package/dist/runner/record.cjs.map +1 -1
  517. package/dist/runner/report.cjs +3 -3
  518. package/dist/runner/report.cjs.map +1 -1
  519. package/dist/runner/report.d.cts +1 -1
  520. package/dist/runner/report.d.mts +1 -1
  521. package/dist/runner/report.d.ts +1 -1
  522. package/dist/runner/reuse-plan.cjs +2 -2
  523. package/dist/runner/reuse-plan.cjs.map +1 -1
  524. package/dist/runner/run.cjs +10 -45
  525. package/dist/runner/run.cjs.map +1 -1
  526. package/dist/runner/sandbox-pool.cjs +1 -2
  527. package/dist/runner/sandbox-pool.cjs.map +1 -1
  528. package/dist/runner/setup-prefix-preparation.cjs +1 -1
  529. package/dist/runner/setup-prefix-preparation.cjs.map +1 -1
  530. package/dist/runner/teardown-registry.cjs +2 -2
  531. package/dist/runner/teardown-registry.cjs.map +1 -1
  532. package/dist/runner/teardown-registry.d.cts +1 -1
  533. package/dist/runner/teardown-registry.d.mts +1 -1
  534. package/dist/runner/teardown-registry.d.ts +1 -1
  535. package/dist/runner/types.cjs +4 -4
  536. package/dist/runner/types.cjs.map +1 -1
  537. package/dist/runner/types.d.cts +64 -92
  538. package/dist/runner/types.d.mts +64 -92
  539. package/dist/runner/types.d.ts +64 -92
  540. package/dist/runner/types.js +2 -2
  541. package/dist/sandbox/action.cjs +1 -1
  542. package/dist/sandbox/action.cjs.map +1 -1
  543. package/dist/sandbox/cli/contribution.cjs +17 -14
  544. package/dist/sandbox/cli/contribution.cjs.map +1 -1
  545. package/dist/sandbox/cli/contribution.d.cts +1 -1
  546. package/dist/sandbox/cli/contribution.d.mts +1 -1
  547. package/dist/sandbox/cli/contribution.d.ts +1 -1
  548. package/dist/sandbox/commands.cjs +1 -1
  549. package/dist/sandbox/commands.cjs.map +1 -1
  550. package/dist/sandbox/commands.d.cts +1 -1
  551. package/dist/sandbox/commands.d.mts +1 -1
  552. package/dist/sandbox/commands.d.ts +1 -1
  553. package/dist/sandbox/compose.cjs +2 -2
  554. package/dist/sandbox/compose.cjs.map +1 -1
  555. package/dist/sandbox/docker-profile/registry.cjs +5 -10
  556. package/dist/sandbox/docker-profile/registry.cjs.map +1 -1
  557. package/dist/sandbox/docker-profile/registry.d.cts +0 -4
  558. package/dist/sandbox/docker-profile/registry.d.mts +0 -4
  559. package/dist/sandbox/docker-profile/registry.d.ts +0 -4
  560. package/dist/sandbox/docker-profile/registry.js +6 -14
  561. package/dist/sandbox/docker-profile/runtime.cjs +1 -1
  562. package/dist/sandbox/docker-profile/runtime.cjs.map +1 -1
  563. package/dist/sandbox/docker-profile/schema.cjs +12 -16
  564. package/dist/sandbox/docker-profile/schema.cjs.map +1 -1
  565. package/dist/sandbox/docker-profile/schema.d.cts +2 -6
  566. package/dist/sandbox/docker-profile/schema.d.mts +2 -6
  567. package/dist/sandbox/docker-profile/schema.d.ts +2 -6
  568. package/dist/sandbox/docker-profile/schema.js +24 -32
  569. package/dist/sandbox/incus.cjs +2 -2
  570. package/dist/sandbox/incus.cjs.map +1 -1
  571. package/dist/sandbox/index.cjs +4 -36
  572. package/dist/sandbox/index.cjs.map +1 -1
  573. package/dist/sandbox/index.d.cts +4 -8
  574. package/dist/sandbox/index.d.mts +4 -8
  575. package/dist/sandbox/index.d.ts +4 -8
  576. package/dist/sandbox/index.js +116 -174
  577. package/dist/sandbox/index.mjs +116 -174
  578. package/dist/sandbox/layer.cjs +3 -3
  579. package/dist/sandbox/layer.cjs.map +1 -1
  580. package/dist/sandbox/link.cjs +5 -0
  581. package/dist/sandbox/link.cjs.map +1 -1
  582. package/dist/sandbox/link.d.cts +3 -0
  583. package/dist/sandbox/link.d.mts +3 -0
  584. package/dist/sandbox/link.d.ts +3 -0
  585. package/dist/sandbox/runtime.cjs +9 -4
  586. package/dist/sandbox/runtime.cjs.map +1 -1
  587. package/dist/shared/failure-class.cjs +5 -5
  588. package/dist/shared/failure-class.cjs.map +1 -1
  589. package/dist/shared/failure-class.d.cts +4 -10
  590. package/dist/shared/failure-class.d.mts +4 -10
  591. package/dist/shared/failure-class.d.ts +4 -10
  592. package/dist/shared/types.cjs.map +1 -1
  593. package/dist/shared/types.d.cts +1 -1
  594. package/dist/shared/types.d.mts +1 -1
  595. package/dist/shared/types.d.ts +1 -1
  596. package/dist/show/contribution.cjs +17 -2
  597. package/dist/show/contribution.cjs.map +1 -1
  598. package/dist/show/contribution.d.cts +14 -0
  599. package/dist/show/contribution.d.mts +14 -0
  600. package/dist/show/contribution.d.ts +14 -0
  601. package/dist/show/model.cjs +8 -5
  602. package/dist/show/model.cjs.map +1 -1
  603. package/dist/show/model.d.cts +11 -8
  604. package/dist/show/model.d.mts +11 -8
  605. package/dist/show/model.d.ts +11 -8
  606. package/dist/show/render.cjs +34 -22
  607. package/dist/show/render.cjs.map +1 -1
  608. package/dist/sources/projector.cjs +1 -1
  609. package/dist/sources/projector.cjs.map +1 -1
  610. package/dist/types.cjs +1 -1
  611. package/dist/types.cjs.map +1 -1
  612. package/dist/types.d.cts +1 -1
  613. package/dist/types.d.mts +1 -1
  614. package/dist/types.d.ts +1 -1
  615. package/dist/types.js +2 -2
  616. package/dist/view/app-dist/.vite/manifest.json +2 -2
  617. package/dist/view/app-dist/assets/{index-B3ijZRph.js → index-CxrOUvHC.js} +17 -17
  618. package/dist/view/app-dist/assets/{quality-cost-scatter-B--DfJ9s.js → quality-cost-scatter-B6Aiss7x.js} +1 -1
  619. package/dist/view/app-dist/index.html +1 -1
  620. package/dist/view/cli/contribution.cjs +1 -1
  621. package/dist/view/cli/contribution.cjs.map +1 -1
  622. package/dist/view/function-runtime.mjs +318 -121
  623. package/dist/view/http-protocol.d.cts +1 -1
  624. package/dist/view/http-protocol.d.mts +1 -1
  625. package/dist/view/http-protocol.d.ts +1 -1
  626. package/dist/view/inspection-host.cjs +2 -2
  627. package/dist/view/inspection-host.cjs.map +1 -1
  628. package/dist/view/inspection-host.d.cts +349 -331
  629. package/dist/view/inspection-host.d.mts +349 -331
  630. package/dist/view/inspection-host.d.ts +349 -331
  631. package/docs-site/zh/examples/integrations/ai-sdk-v7.mdx +9 -9
  632. package/docs-site/zh/examples/integrations/claude-sdk.mdx +8 -8
  633. package/docs-site/zh/examples/integrations/codex-sdk.mdx +4 -4
  634. package/docs-site/zh/examples/integrations/langgraph.mdx +6 -6
  635. package/docs-site/zh/examples/integrations/pi-sdk.mdx +8 -8
  636. package/docs-site/zh/explanation/adapter.mdx +6 -6
  637. package/docs-site/zh/explanation/assert.mdx +7 -7
  638. package/docs-site/zh/explanation/drive.mdx +4 -4
  639. package/docs-site/zh/explanation/hitl.mdx +2 -2
  640. package/docs-site/zh/explanation/judge.mdx +3 -3
  641. package/docs-site/zh/index.mdx +18 -0
  642. package/docs-site/zh/introduction.mdx +4 -3
  643. package/docs-site/zh/reference/builtin-agents.mdx +46 -49
  644. package/docs-site/zh/reference/cli.mdx +16 -14
  645. package/docs-site/zh/reference/define-agent.mdx +19 -19
  646. package/docs-site/zh/reference/define-config.mdx +7 -15
  647. package/docs-site/zh/reference/define-eval.mdx +20 -14
  648. package/docs-site/zh/reference/define-experiment.mdx +18 -39
  649. package/docs-site/zh/reference/events.mdx +1 -1
  650. package/docs-site/zh/reference/expect.mdx +21 -0
  651. package/docs-site/zh/reference/official-adapters.mdx +12 -5
  652. package/docs-site/zh/troubleshooting/debugging.mdx +2 -2
  653. package/docs-site/zh/troubleshooting/recover-after-kill.mdx +1 -1
  654. package/docs-site/zh/tutorials/agent-onboarding.mdx +2 -2
  655. package/docs-site/zh/tutorials/authoring.mdx +4 -3
  656. package/docs-site/zh/tutorials/compare-runs.mdx +1 -1
  657. package/docs-site/zh/tutorials/concurrency.mdx +8 -8
  658. package/docs-site/zh/tutorials/configuration.mdx +5 -5
  659. package/docs-site/zh/tutorials/control-run-cost.mdx +49 -9
  660. package/docs-site/zh/tutorials/eval-groups.mdx +1 -1
  661. package/docs-site/zh/tutorials/evaluation-kinds.mdx +5 -5
  662. package/docs-site/zh/tutorials/experiment-services.mdx +32 -32
  663. package/docs-site/zh/tutorials/handle-execution-failures.mdx +2 -28
  664. package/docs-site/zh/tutorials/plugins.mdx +49 -143
  665. package/docs-site/zh/tutorials/sandbox-agent.mdx +1 -1
  666. package/docs-site/zh/tutorials/sandbox-reuse.mdx +7 -7
  667. package/docs-site/zh/tutorials/upgrade-0-18.mdx +64 -0
  668. package/docs-site/zh/tutorials/upgrade-0-19.mdx +121 -0
  669. package/docs-site/zh/tutorials/upgrade-0-20.mdx +12 -0
  670. package/docs-site/zh/tutorials/verify-judge.mdx +4 -3
  671. package/docs-site/zh/tutorials/viewing-results.mdx +2 -2
  672. package/docs-site/zh/tutorials/write-experiment.mdx +1 -1
  673. package/docs-site/zh/tutorials/write-send.mdx +13 -13
  674. package/package.json +1 -1
  675. package/dist/agents/effect-runtime.cjs.map +0 -1
  676. package/dist/agents/effect-runtime.d.cts +0 -29
  677. package/dist/agents/effect-runtime.d.mts +0 -29
  678. package/dist/agents/effect-runtime.d.ts +0 -29
  679. package/dist/agents/effect-runtime.js +0 -16
  680. package/dist/agents/effect-runtime.js.map +0 -1
  681. package/dist/agents/session-slot.cjs.map +0 -1
  682. package/dist/agents/session-slot.d.cts +0 -12
  683. package/dist/agents/session-slot.d.mts +0 -12
  684. package/dist/agents/session-slot.d.ts +0 -12
  685. package/dist/agents/session-slot.js +0 -6
  686. package/dist/agents/session-slot.js.map +0 -1
  687. package/dist/model-slots.cjs.map +0 -1
  688. package/dist/model-slots.d.cts +0 -19
  689. package/dist/model-slots.d.mts +0 -19
  690. package/dist/model-slots.d.ts +0 -19
  691. package/dist/model-slots.js +0 -8
  692. package/dist/model-slots.js.map +0 -1
@@ -71,7 +71,7 @@ codex-sdk/
71
71
  <div className="gd-file">
72
72
  <div className="gd-head"><span className="gd-name">{"niceeval.config.ts"}</span><span className="gd-stats"><span className="gd-plus">{"+11"}</span></span></div>
73
73
  <div className="gd-body">
74
- <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineConfig } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { OpenAIProvider } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/judge\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineConfig"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" name: { "}</span><span className="gdt2">{"\"zh-CN\""}</span><span className="gdt0">{": "}</span><span className="gdt2">{"\"Codex SDK 示例\""}</span><span className="gdt0">{", en: "}</span><span className="gdt2">{"\"Codex SDK example\""}</span><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" judgeRuntime: "}</span><span className="gdt5">{"OpenAIProvider"}</span><span className="gdt0">{"({ model: "}</span><span className="gdt2">{"\"gpt-5.4\""}</span><span className="gdt0">{" }),"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" timeoutMs: "}</span><span className="gdt1">{"180_000"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt6">{"// 每个 attempt 都要经一个真实子进程(server.ts)+ Codex CLI 子进程(真的在工作目录里"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt6">{"// 跑命令/改文件)+ 网络调用,比其它示例都重,别开太高并发。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" maxConcurrency: "}</span><span className="gdt1">{"2"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
74
+ <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineConfig } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { OpenAIProvider } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/judge\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineConfig"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" name: { "}</span><span className="gdt2">{"\"zh-CN\""}</span><span className="gdt0">{": "}</span><span className="gdt2">{"\"Codex SDK 示例\""}</span><span className="gdt0">{", en: "}</span><span className="gdt2">{"\"Codex SDK example\""}</span><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" judgeProvider: "}</span><span className="gdt5">{"OpenAIProvider"}</span><span className="gdt0">{"({ model: "}</span><span className="gdt2">{"\"gpt-5.4\""}</span><span className="gdt0">{" }),"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" timeoutMs: "}</span><span className="gdt1">{"180_000"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt6">{"// 每个 attempt 都要经一个真实子进程(server.ts)+ Codex CLI 子进程(真的在工作目录里"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt6">{"// 跑命令/改文件)+ 网络调用,比其它示例都重,别开太高并发。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" maxConcurrency: "}</span><span className="gdt1">{"2"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
75
75
  </div>
76
76
  </div>
77
77
 
@@ -85,7 +85,7 @@ codex-sdk/
85
85
  <div className="gd-file">
86
86
  <div className="gd-head"><span className="gd-name">{"evals/basic-qa.eval.ts"}</span><span className="gd-stats"><span className="gd-plus">{"+23"}</span></span></div>
87
87
  <div className="gd-body">
88
- <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineEval } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { includes } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/expect\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 这条 eval 验证 agent 能正常问答,顺带冒烟 usage 有没有从 turn.completed 的"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// usage(input_tokens/output_tokens/cached_input_tokens)正确映射进 Turn.usage。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// Codex 是自主编码 agent,即使是纯问答也可能顺手探索一下工作目录——这里不强断言"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// usedNoTools(),只看回答本身和有没有失败的动作。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineEval"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"测试 agent 能正常问答\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"async"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"test"}</span><span className="gdt0">{"("}</span><span className="gdt7">{"t"}</span><span className="gdt0">{") {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"12"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"const"}</span><span className="gdt0">{" "}</span><span className="gdt1">{"turn"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"="}</span><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"send"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"1+1 等于几?用一句话回答就好,不用跑命令也不用建文件。\""}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"13"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" turn."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"()."}</span><span className="gdt5">{"orStop"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"14"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"15"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"group"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"正常收发\""}</span><span className="gdt0">{", () "}</span><span className="gdt4">{"=>"}</span><span className="gdt0">{" {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"16"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"17"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"noFailedActions"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"18"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"check"}</span><span className="gdt0">{"(turn.message, "}</span><span className="gdt5">{"includes"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"2\""}</span><span className="gdt0">{"));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"19"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" });"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"20"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"21"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"maxTokens"}</span><span className="gdt0">{"("}</span><span className="gdt1">{"40_000"}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"22"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"23"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
88
+ <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineEval } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { includes } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/expect\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 这条 eval 验证 agent 能正常问答,顺带冒烟 usage 有没有从 turn.completed 的"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// usage(input_tokens/output_tokens/cached_input_tokens)正确映射进 Turn.usage。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// Codex 是自主编码 agent,即使是纯问答也可能顺手探索一下工作目录——这里不强断言"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// usedNoTools(),只看回答本身和有没有失败的动作。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineEval"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"测试 agent 能正常问答\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"async"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"test"}</span><span className="gdt0">{"("}</span><span className="gdt7">{"t"}</span><span className="gdt0">{") {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"12"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"const"}</span><span className="gdt0">{" "}</span><span className="gdt1">{"turn"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"="}</span><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"send"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"1+1 等于几?用一句话回答就好,不用跑命令也不用建文件。\""}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"13"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" turn."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"()."}</span><span className="gdt5">{"orStop"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"14"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"15"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"group"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"正常收发\""}</span><span className="gdt0">{", () "}</span><span className="gdt4">{"=>"}</span><span className="gdt0">{" {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"16"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"17"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"noFailedActions"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"18"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"check"}</span><span className="gdt0">{"(turn.reply, "}</span><span className="gdt5">{"includes"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"2\""}</span><span className="gdt0">{"));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"19"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" });"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"20"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"21"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"maxTokens"}</span><span className="gdt0">{"("}</span><span className="gdt1">{"40_000"}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"22"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"23"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
89
89
  </div>
90
90
  </div>
91
91
 
@@ -99,7 +99,7 @@ codex-sdk/
99
99
  <div className="gd-file">
100
100
  <div className="gd-head"><span className="gd-name">{"evals/run-command.eval.ts"}</span><span className="gd-stats"><span className="gd-plus">{"+21"}</span></span></div>
101
101
  <div className="gd-body">
102
- <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineEval } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { includes, toolMatch } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/expect\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 这条 eval 验证 agent 会真的跑 shell 命令(而不是凭空回答)。工具事件由官方转换器"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// `createCodexThreadEventStream` 从 ThreadEvent 的 `command_execution` item 映射(started 发"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// called,completed 按 exit_code 落 result)。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineEval"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"测试 agent 能在工作目录里跑一个真实 shell 命令\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"async"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"test"}</span><span className="gdt0">{"("}</span><span className="gdt7">{"t"}</span><span className="gdt0">{") {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"const"}</span><span className="gdt0">{" "}</span><span className="gdt1">{"turn"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"="}</span><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"send"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"在当前工作目录跑 `echo niceeval-run-command-926`,把命令的输出告诉我。\""}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"12"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" turn."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"()."}</span><span className="gdt5">{"orStop"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"13"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"14"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"group"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"调用了 shell 且没有失败的动作\""}</span><span className="gdt0">{", () "}</span><span className="gdt4">{"=>"}</span><span className="gdt0">{" {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"15"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"calledTool"}</span><span className="gdt0">{"("}</span><span className="gdt5">{"toolMatch"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"command_execution\""}</span><span className="gdt0">{", { status: "}</span><span className="gdt2">{"\"completed\""}</span><span className="gdt0">{" }));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"16"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"noFailedActions"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"17"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" });"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"18"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"19"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"check"}</span><span className="gdt0">{"(turn.message, "}</span><span className="gdt5">{"includes"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"niceeval-run-command-926\""}</span><span className="gdt0">{"));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"20"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"21"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
102
+ <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineEval } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { includes, toolMatch } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval/expect\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 这条 eval 验证 agent 会真的跑 shell 命令(而不是凭空回答)。工具事件由官方转换器"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// `createCodexThreadEventStream` 从 ThreadEvent 的 `command_execution` item 映射(started 发"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// called,completed 按 exit_code 落 result)。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineEval"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"测试 agent 能在工作目录里跑一个真实 shell 命令\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"10"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"async"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"test"}</span><span className="gdt0">{"("}</span><span className="gdt7">{"t"}</span><span className="gdt0">{") {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"11"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"const"}</span><span className="gdt0">{" "}</span><span className="gdt1">{"turn"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"="}</span><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"send"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"在当前工作目录跑 `echo niceeval-run-command-926`,把命令的输出告诉我。\""}</span><span className="gdt0">{");"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"12"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" turn."}</span><span className="gdt5">{"succeeded"}</span><span className="gdt0">{"()."}</span><span className="gdt5">{"orStop"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"13"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"14"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" "}</span><span className="gdt4">{"await"}</span><span className="gdt0">{" t."}</span><span className="gdt5">{"group"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"调用了 shell 且没有失败的动作\""}</span><span className="gdt0">{", () "}</span><span className="gdt4">{"=>"}</span><span className="gdt0">{" {"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"15"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"calledTool"}</span><span className="gdt0">{"("}</span><span className="gdt5">{"toolMatch"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"command_execution\""}</span><span className="gdt0">{", { status: "}</span><span className="gdt2">{"\"completed\""}</span><span className="gdt0">{" }));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"16"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"noFailedActions"}</span><span className="gdt0">{"();"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"17"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" });"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"18"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"19"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" t."}</span><span className="gdt5">{"check"}</span><span className="gdt0">{"(turn.reply, "}</span><span className="gdt5">{"includes"}</span><span className="gdt0">{"("}</span><span className="gdt2">{"\"niceeval-run-command-926\""}</span><span className="gdt0">{"));"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"20"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" },"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"21"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
103
103
  </div>
104
104
  </div>
105
105
 
@@ -113,6 +113,6 @@ codex-sdk/
113
113
  <div className="gd-file">
114
114
  <div className="gd-head"><span className="gd-name">{"experiments/codex-sdk.ts"}</span><span className="gd-stats"><span className="gd-plus">{"+9"}</span></span></div>
115
115
  <div className="gd-body">
116
- <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineExperiment } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" agent "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"../agents/codex-sdk.ts\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 单配置基线:这个示例只有一个 agent、不比较模型,用 .env 里的默认模型(gpt-5.4)。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineExperiment"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"codex-sdk:真实 Codex SDK 后端\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" agent,"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" attempts: "}</span><span className="gdt1">{"1"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
116
+ <table className="gd-table"><tbody><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"1"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" { defineExperiment } "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"niceeval\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"2"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"import"}</span><span className="gdt0">{" agent "}</span><span className="gdt4">{"from"}</span><span className="gdt0">{" "}</span><span className="gdt2">{"\"../agents/codex-sdk.ts\""}</span><span className="gdt0">{";"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"3"}</td><td className="gd-sign">{"+"}</td><td className="gd-code">{" "}</td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"4"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt6">{"// 单配置基线:这个示例只有一个 agent、不比较模型,用 .env 里的默认模型(gpt-5.4)。"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"5"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt4">{"export"}</span><span className="gdt0">{" "}</span><span className="gdt4">{"default"}</span><span className="gdt0">{" "}</span><span className="gdt5">{"defineExperiment"}</span><span className="gdt0">{"({"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"6"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" description: "}</span><span className="gdt2">{"\"codex-sdk:真实 Codex SDK 后端\""}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"7"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" adapter: agent,"}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"8"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{" attempts: "}</span><span className="gdt1">{"1"}</span><span className="gdt0">{","}</span></td></tr><tr className="gd-add"><td className="gd-ln"></td><td className="gd-ln">{"9"}</td><td className="gd-sign">{"+"}</td><td className="gd-code"><span className="gdt0">{"});"}</span></td></tr></tbody></table>
117
117
  </div>
118
118
  </div>