niceeval 0.5.4 → 0.6.1

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 (337) hide show
  1. package/INDEX.md +55 -0
  2. package/dist/agents/types.d.ts +321 -0
  3. package/dist/context/types.d.ts +335 -0
  4. package/dist/i18n/core.d.ts +11 -0
  5. package/dist/i18n/en.d.ts +181 -0
  6. package/dist/i18n/index.d.ts +5 -0
  7. package/dist/i18n/zh-CN.d.ts +183 -0
  8. package/dist/o11y/otlp/parse.d.ts +3 -0
  9. package/dist/o11y/otlp/receiver.d.ts +21 -0
  10. package/dist/o11y/otlp/turn-otel.d.ts +43 -0
  11. package/dist/o11y/types.d.ts +190 -0
  12. package/dist/report/aggregate.d.ts +57 -0
  13. package/dist/report/aggregate.js +212 -0
  14. package/dist/report/built-ins/experiment-comparison.d.ts +1 -0
  15. package/dist/report/built-ins/experiment-comparison.js +13 -0
  16. package/dist/report/built-ins/index.d.ts +1 -0
  17. package/dist/report/built-ins/index.js +2 -0
  18. package/dist/report/components.d.ts +180 -0
  19. package/dist/report/components.js +132 -0
  20. package/dist/report/compute.d.ts +106 -0
  21. package/dist/report/compute.js +600 -0
  22. package/dist/report/flag.d.ts +6 -0
  23. package/dist/report/flag.js +8 -0
  24. package/dist/report/format.d.ts +26 -0
  25. package/dist/report/format.js +109 -0
  26. package/dist/report/index.d.ts +19 -0
  27. package/dist/report/index.js +30 -0
  28. package/dist/report/load.d.ts +12 -0
  29. package/dist/report/load.js +56 -0
  30. package/dist/report/locale.d.ts +94 -0
  31. package/dist/report/locale.js +183 -0
  32. package/dist/report/metrics.d.ts +21 -0
  33. package/dist/report/metrics.js +115 -0
  34. package/dist/report/primitives.d.ts +64 -0
  35. package/dist/report/primitives.js +113 -0
  36. package/dist/report/react/AttemptList.d.ts +23 -0
  37. package/dist/report/react/AttemptList.js +19 -0
  38. package/dist/report/react/DeltaTable.d.ts +8 -0
  39. package/dist/report/react/DeltaTable.js +24 -0
  40. package/dist/report/react/EvalList.d.ts +10 -0
  41. package/dist/report/react/EvalList.js +0 -0
  42. package/dist/report/react/ExperimentList.d.ts +11 -0
  43. package/dist/report/react/ExperimentList.js +67 -0
  44. package/dist/report/react/GroupSummary.d.ts +8 -0
  45. package/dist/report/react/GroupSummary.js +8 -0
  46. package/dist/report/react/MetricBars.d.ts +10 -0
  47. package/dist/report/react/MetricBars.js +48 -0
  48. package/dist/report/react/MetricLine.d.ts +9 -0
  49. package/dist/report/react/MetricLine.js +68 -0
  50. package/dist/report/react/MetricMatrix.d.ts +10 -0
  51. package/dist/report/react/MetricMatrix.js +23 -0
  52. package/dist/report/react/MetricScatter.d.ts +9 -0
  53. package/dist/report/react/MetricScatter.js +96 -0
  54. package/dist/report/react/MetricTable.d.ts +11 -0
  55. package/dist/report/react/MetricTable.js +26 -0
  56. package/dist/report/react/RunOverview.d.ts +8 -0
  57. package/dist/report/react/RunOverview.js +12 -0
  58. package/dist/report/react/Scoreboard.d.ts +8 -0
  59. package/dist/report/react/Scoreboard.js +26 -0
  60. package/dist/report/react/cell.d.ts +9 -0
  61. package/dist/report/react/cell.js +9 -0
  62. package/dist/report/react/chart-math.d.ts +14 -0
  63. package/dist/report/react/chart-math.js +68 -0
  64. package/dist/report/react/colors.d.ts +14 -0
  65. package/dist/report/react/colors.js +45 -0
  66. package/dist/report/react/fixtures.d.ts +19 -0
  67. package/dist/report/react/fixtures.js +398 -0
  68. package/dist/report/react/format.d.ts +3 -0
  69. package/dist/report/react/format.js +7 -0
  70. package/dist/report/react/index.d.ts +6 -0
  71. package/dist/report/react/index.js +14 -0
  72. package/dist/report/report.d.ts +24 -0
  73. package/dist/report/report.js +49 -0
  74. package/dist/report/text/faces.d.ts +29 -0
  75. package/dist/report/text/faces.js +541 -0
  76. package/dist/report/text/layout.d.ts +75 -0
  77. package/dist/report/text/layout.js +173 -0
  78. package/dist/report/text/plot.d.ts +37 -0
  79. package/dist/report/text/plot.js +81 -0
  80. package/dist/report/text/table.d.ts +8 -0
  81. package/dist/report/text/table.js +86 -0
  82. package/dist/report/tree.d.ts +99 -0
  83. package/dist/report/tree.js +228 -0
  84. package/dist/report/types.d.ts +408 -0
  85. package/dist/report/types.js +4 -0
  86. package/dist/report/web.d.ts +15 -0
  87. package/dist/report/web.js +37 -0
  88. package/dist/results/locator.d.ts +75 -0
  89. package/dist/results/locator.js +0 -0
  90. package/dist/results/select.d.ts +45 -0
  91. package/dist/results/select.js +230 -0
  92. package/dist/results/source-hash.d.ts +9 -0
  93. package/dist/results/types.d.ts +204 -0
  94. package/dist/runner/eval-source.d.ts +22 -0
  95. package/dist/runner/fingerprint.d.ts +23 -0
  96. package/dist/runner/types.d.ts +332 -0
  97. package/dist/sandbox/docker-stream.d.ts +24 -0
  98. package/dist/sandbox/docker.d.ts +84 -0
  99. package/dist/sandbox/e2b.d.ts +31 -0
  100. package/dist/sandbox/errors.d.ts +4 -0
  101. package/dist/sandbox/local-files.d.ts +6 -0
  102. package/dist/sandbox/paths.d.ts +4 -0
  103. package/dist/sandbox/registry.d.ts +13 -0
  104. package/dist/sandbox/resolve.d.ts +40 -0
  105. package/dist/sandbox/retry.d.ts +10 -0
  106. package/dist/sandbox/shell.d.ts +11 -0
  107. package/dist/sandbox/source-files.d.ts +13 -0
  108. package/dist/sandbox/types.d.ts +203 -0
  109. package/dist/sandbox/vercel.d.ts +37 -0
  110. package/dist/scoring/types.d.ts +70 -0
  111. package/dist/shared/aggregate.d.ts +29 -0
  112. package/dist/shared/aggregate.js +55 -0
  113. package/dist/shared/types.d.ts +27 -0
  114. package/dist/shared/verdict.d.ts +27 -0
  115. package/dist/shared/verdict.js +34 -0
  116. package/dist/tty-line.d.ts +6 -0
  117. package/dist/types.d.ts +7 -0
  118. package/dist/util.d.ts +22 -0
  119. package/docs-site/zh/concepts/adapter.mdx +7 -7
  120. package/docs-site/zh/concepts/assert.mdx +19 -18
  121. package/docs-site/zh/concepts/evals.mdx +12 -11
  122. package/docs-site/zh/concepts/experiment.mdx +2 -2
  123. package/docs-site/zh/concepts/judge.mdx +9 -9
  124. package/docs-site/zh/concepts/overview.mdx +3 -3
  125. package/docs-site/zh/concepts/tier.mdx +6 -6
  126. package/docs-site/zh/example/claude-code-codex-plugin.mdx +9 -5
  127. package/docs-site/zh/example/claude-code-codex-skill.mdx +12 -7
  128. package/docs-site/zh/guides/agent-feedback-loop.mdx +237 -78
  129. package/docs-site/zh/guides/connect-otel.mdx +3 -3
  130. package/docs-site/zh/guides/connect-your-agent.mdx +2 -2
  131. package/docs-site/zh/guides/custom-reports.mdx +196 -36
  132. package/docs-site/zh/guides/experiments.mdx +7 -1
  133. package/docs-site/zh/guides/fixtures.mdx +4 -2
  134. package/docs-site/zh/guides/official-adapters.mdx +31 -9
  135. package/docs-site/zh/guides/publish-report.mdx +4 -4
  136. package/docs-site/zh/guides/report-components.mdx +194 -55
  137. package/docs-site/zh/guides/reporters.mdx +2 -2
  138. package/docs-site/zh/guides/results-data.mdx +61 -34
  139. package/docs-site/zh/guides/runner.mdx +2 -2
  140. package/docs-site/zh/guides/sandbox-agent.mdx +3 -2
  141. package/docs-site/zh/guides/{sandbox-backends.mdx → sandbox-providers.mdx} +12 -12
  142. package/docs-site/zh/guides/viewing-results.mdx +200 -79
  143. package/docs-site/zh/guides/write-experiment.mdx +9 -1
  144. package/docs-site/zh/guides/write-send.mdx +4 -3
  145. package/docs-site/zh/index.mdx +4 -3
  146. package/docs-site/zh/introduction.mdx +2 -1
  147. package/docs-site/zh/quickstart.mdx +5 -3
  148. package/docs-site/zh/reference/builtin-agents.mdx +38 -13
  149. package/docs-site/zh/reference/capabilities.mdx +1 -1
  150. package/docs-site/zh/reference/cli.mdx +36 -13
  151. package/docs-site/zh/reference/define-agent.mdx +15 -13
  152. package/docs-site/zh/reference/define-config.mdx +1 -1
  153. package/docs-site/zh/reference/define-eval.mdx +3 -3
  154. package/docs-site/zh/reference/events.mdx +9 -3
  155. package/docs-site/zh/reference/expect.mdx +13 -8
  156. package/package.json +10 -6
  157. package/src/agents/bub.ts +129 -42
  158. package/src/agents/claude-code.test.ts +152 -0
  159. package/src/agents/claude-code.ts +127 -13
  160. package/src/agents/codex.test.ts +192 -0
  161. package/src/agents/codex.ts +151 -12
  162. package/src/agents/index.ts +9 -3
  163. package/src/agents/manifest.ts +41 -0
  164. package/src/agents/plugin-config.test.ts +94 -0
  165. package/src/agents/shared.ts +33 -2
  166. package/src/agents/skills.test.ts +214 -0
  167. package/src/agents/skills.ts +304 -0
  168. package/src/agents/streaming.ts +5 -5
  169. package/src/agents/types.ts +59 -3
  170. package/src/cli.ts +74 -47
  171. package/src/context/context.test.ts +20 -2
  172. package/src/context/control-flow.ts +1 -1
  173. package/src/context/session.test.ts +92 -1
  174. package/src/context/session.ts +23 -1
  175. package/src/context/types.ts +4 -4
  176. package/src/define.test.ts +8 -8
  177. package/src/define.ts +11 -11
  178. package/src/expect/index.ts +5 -5
  179. package/src/i18n/en.ts +33 -15
  180. package/src/i18n/zh-CN.ts +37 -21
  181. package/src/index.ts +1 -1
  182. package/src/o11y/execution-tree.test.ts +452 -0
  183. package/src/o11y/execution-tree.ts +367 -0
  184. package/src/o11y/otlp/receiver.ts +1 -1
  185. package/src/o11y/otlp/select.ts +7 -3
  186. package/src/o11y/parsers/claude-code.test.ts +175 -0
  187. package/src/o11y/parsers/claude-code.ts +32 -0
  188. package/src/o11y/types.ts +6 -1
  189. package/src/report/aggregate.ts +26 -20
  190. package/src/report/built-in-user-parity.test.tsx +640 -0
  191. package/src/report/built-ins/experiment-comparison.tsx +19 -0
  192. package/src/report/built-ins/index.ts +2 -0
  193. package/src/report/components.tsx +225 -50
  194. package/src/report/compute.ts +335 -54
  195. package/src/report/dual-render.test.tsx +1129 -0
  196. package/src/report/flag.ts +1 -1
  197. package/src/report/format.ts +35 -2
  198. package/src/report/index.ts +47 -14
  199. package/src/report/load.ts +1 -1
  200. package/src/report/locale.ts +229 -0
  201. package/src/report/metrics.ts +39 -15
  202. package/src/report/primitives.tsx +100 -2
  203. package/src/report/react/AttemptList.tsx +113 -0
  204. package/src/report/react/DeltaTable.tsx +9 -6
  205. package/src/report/react/EvalList.tsx +0 -0
  206. package/src/report/react/ExperimentList.tsx +220 -0
  207. package/src/report/react/GroupSummary.tsx +66 -0
  208. package/src/report/react/MetricBars.tsx +24 -13
  209. package/src/report/react/MetricLine.tsx +21 -21
  210. package/src/report/react/MetricMatrix.tsx +8 -4
  211. package/src/report/react/MetricScatter.tsx +126 -96
  212. package/src/report/react/MetricTable.tsx +115 -15
  213. package/src/report/react/RunOverview.tsx +48 -26
  214. package/src/report/react/Scoreboard.tsx +15 -7
  215. package/src/report/react/cell.tsx +18 -9
  216. package/src/{view/app/lib/chart.ts → report/react/chart-math.ts} +2 -6
  217. package/src/report/react/colors.ts +24 -15
  218. package/src/report/react/enhance.js +168 -0
  219. package/src/report/react/fixtures.ts +169 -34
  220. package/src/report/react/format.ts +1 -1
  221. package/src/report/react/index.tsx +21 -6
  222. package/src/report/react/render.test.tsx +180 -31
  223. package/src/report/react/styles.css +432 -213
  224. package/src/report/report.test.ts +398 -94
  225. package/src/report/report.ts +29 -11
  226. package/src/report/text/faces.ts +344 -96
  227. package/src/report/text/layout.ts +76 -11
  228. package/src/report/text/table.ts +98 -0
  229. package/src/report/tree.ts +103 -20
  230. package/src/report/types.ts +184 -25
  231. package/src/report/web.ts +46 -19
  232. package/src/results/annotated-source.test.ts +146 -0
  233. package/src/results/annotated-source.ts +128 -0
  234. package/src/results/attempt-evidence.test.ts +158 -0
  235. package/src/results/attempt-evidence.ts +158 -0
  236. package/src/results/attempt-source.ts +48 -0
  237. package/src/results/copy.ts +154 -139
  238. package/src/results/format.ts +54 -47
  239. package/src/results/host-equivalence.test.ts +527 -0
  240. package/src/results/index.ts +56 -14
  241. package/src/results/locator.test.ts +198 -0
  242. package/src/results/locator.ts +0 -0
  243. package/src/results/open.ts +392 -175
  244. package/src/results/results.test.ts +943 -494
  245. package/src/results/select.ts +129 -23
  246. package/src/results/skipped-notice.ts +0 -0
  247. package/src/results/source-hash.ts +28 -0
  248. package/src/results/types.ts +98 -62
  249. package/src/results/writer.ts +304 -176
  250. package/src/runner/attempt.test.ts +170 -0
  251. package/src/runner/attempt.ts +48 -13
  252. package/src/runner/discover.test.ts +65 -0
  253. package/src/runner/discover.ts +6 -2
  254. package/src/runner/eval-source.test.ts +121 -0
  255. package/src/runner/eval-source.ts +45 -0
  256. package/src/runner/fingerprint.ts +54 -1
  257. package/src/runner/remote-sandbox.ts +1 -1
  258. package/src/runner/report.test.ts +6 -8
  259. package/src/runner/report.ts +5 -4
  260. package/src/runner/reporters/artifacts.ts +27 -41
  261. package/src/runner/reporters/braintrust.test.ts +3 -3
  262. package/src/runner/reporters/braintrust.ts +1 -1
  263. package/src/runner/reporters/console.ts +6 -6
  264. package/src/runner/reporters/json.ts +4 -4
  265. package/src/runner/reporters/live.test.ts +56 -0
  266. package/src/runner/reporters/live.ts +36 -10
  267. package/src/runner/reporters/quiet.test.ts +66 -0
  268. package/src/runner/reporters/quiet.ts +49 -0
  269. package/src/runner/reporters/shared.ts +5 -5
  270. package/src/runner/reporters/table.ts +17 -17
  271. package/src/runner/run.ts +61 -102
  272. package/src/runner/types.ts +43 -32
  273. package/src/sandbox/checkpoint.ts +1 -1
  274. package/src/sandbox/docker.ts +15 -2
  275. package/src/sandbox/e2b.ts +11 -5
  276. package/src/sandbox/errors.ts +13 -0
  277. package/src/sandbox/index.ts +3 -3
  278. package/src/sandbox/local-files.ts +1 -1
  279. package/src/sandbox/registry.ts +6 -5
  280. package/src/sandbox/resolve.ts +38 -23
  281. package/src/sandbox/retry.test.ts +68 -0
  282. package/src/sandbox/retry.ts +44 -0
  283. package/src/sandbox/types.ts +24 -24
  284. package/src/sandbox/vercel.ts +17 -3
  285. package/src/scoring/collector.ts +3 -3
  286. package/src/scoring/judge.ts +2 -2
  287. package/src/scoring/scoped.ts +13 -1
  288. package/src/scoring/types.ts +7 -7
  289. package/src/scoring/verdict.ts +4 -4
  290. package/src/shared/aggregate.ts +23 -3
  291. package/src/shared/types.ts +2 -2
  292. package/src/shared/{outcome.ts → verdict.ts} +15 -15
  293. package/src/show/compose.ts +12 -132
  294. package/src/show/index.ts +119 -95
  295. package/src/show/render.ts +462 -159
  296. package/src/show/show.test.ts +471 -253
  297. package/src/source-loc.ts +1 -1
  298. package/src/tty-line.ts +32 -0
  299. package/src/types.ts +2 -2
  300. package/src/util.test.ts +21 -1
  301. package/src/util.ts +5 -1
  302. package/src/view/app/App.tsx +33 -198
  303. package/src/view/app/components/AttemptModal.tsx +6 -6
  304. package/src/view/app/components/CodeView.tsx +2 -2
  305. package/src/view/app/components/CopyControls.tsx +32 -36
  306. package/src/view/app/components/SkippedRunsBanner.tsx +140 -0
  307. package/src/view/app/components/ui/badge.tsx +1 -1
  308. package/src/view/app/components/ui/dialog.tsx +2 -2
  309. package/src/view/app/i18n.ts +40 -31
  310. package/src/view/app/lib/artifact-url.ts +1 -1
  311. package/src/view/app/lib/attempt-route.test.ts +37 -47
  312. package/src/view/app/lib/attempt-route.ts +26 -26
  313. package/src/view/app/lib/rows.ts +9 -112
  314. package/src/view/app/lib/verdict.ts +25 -0
  315. package/src/view/app/main.tsx +9 -12
  316. package/src/view/app/pages/RunsPage.tsx +4 -4
  317. package/src/view/app/pages/TracesPage.tsx +3 -3
  318. package/src/view/app/shared.ts +0 -1
  319. package/src/view/app/types.ts +6 -51
  320. package/src/view/artifact-serving.test.ts +136 -0
  321. package/src/view/client-dist/app.css +1 -1
  322. package/src/view/client-dist/app.js +21 -21
  323. package/src/view/data.test.ts +214 -143
  324. package/src/view/data.ts +166 -120
  325. package/src/view/index.ts +23 -11
  326. package/src/view/server.ts +52 -17
  327. package/src/view/shared/types.ts +22 -31
  328. package/src/view/styles.css +117 -327
  329. package/src/view/view-report.test.ts +146 -98
  330. package/src/report/default-report.tsx +0 -222
  331. package/src/report/dual-face.test.tsx +0 -527
  332. package/src/report/react/CaseList.tsx +0 -70
  333. package/src/view/app/components/CostScoreChart.tsx +0 -147
  334. package/src/view/app/components/ExperimentTable.tsx +0 -284
  335. package/src/view/app/components/GroupSelector.tsx +0 -61
  336. package/src/view/app/components/primitives.tsx +0 -43
  337. package/src/view/app/lib/outcome.ts +0 -82
@@ -151,73 +151,111 @@ h1 {
151
151
  .report-slot {
152
152
  margin: 0 auto 36px;
153
153
  }
154
- .summary {
155
- display: grid;
156
- grid-template-columns: repeat(4, minmax(0, 1fr));
157
- border: 1px solid var(--line);
154
+ /* schemaVersion 不匹配等读不了的 run:报告槽不解析它们,这里占位提示。同一原因(且文案
155
+ 完全相同)的条目按 SkippedRunsBanner 分组折叠,不逐条重复整句原因 + 命令——那是原实现
156
+ 把首屏挤满、警示色又淡到看不出来的根因。 */
157
+ .incompatible-banner {
158
+ border: 1px solid color-mix(in oklch, var(--warn), var(--line) 35%);
159
+ border-left: 3px solid var(--warn);
158
160
  border-radius: 8px;
159
- overflow: hidden;
161
+ background: color-mix(in oklch, var(--warn), var(--panel) 85%);
162
+ padding: 12px 16px 14px;
160
163
  margin: 0 auto 36px;
161
- background: linear-gradient(180deg, color-mix(in oklch, var(--panel), var(--panel-2) 18%), var(--panel));
162
- max-width: 100%;
164
+ font-size: 13px;
163
165
  }
164
- .metric {
165
- padding: 14px 20px 15px;
166
- border-right: 1px solid var(--line);
167
- min-width: 0;
166
+ .ib-head {
167
+ display: flex;
168
+ align-items: center;
169
+ gap: 8px;
170
+ }
171
+ .ib-head b { font-weight: 640; }
172
+ .ib-icon {
173
+ width: 16px;
174
+ height: 16px;
175
+ flex-shrink: 0;
176
+ color: var(--warn);
177
+ stroke-width: 2;
178
+ }
179
+ .ib-groups {
168
180
  display: grid;
169
- grid-template-columns: 1fr auto;
170
- align-items: baseline;
171
- column-gap: 12px;
181
+ gap: 10px;
182
+ margin-top: 10px;
172
183
  }
173
- .metric:last-child { border-right: 0; }
174
- .metric .label {
175
- color: var(--soft);
176
- font-size: 12px;
177
- text-transform: uppercase;
178
- letter-spacing: 0.08em;
184
+ .ib-group-head {
185
+ display: flex;
186
+ align-items: baseline;
187
+ gap: 8px;
179
188
  }
180
- .metric .value {
181
- margin-top: 0;
182
- font-size: clamp(19px, 2vw, 24px);
183
- line-height: 1;
184
- font-weight: 680;
189
+ .ib-group-count {
190
+ display: inline-flex;
191
+ align-items: center;
192
+ justify-content: center;
193
+ min-width: 20px;
194
+ height: 20px;
195
+ padding: 0 6px;
196
+ flex-shrink: 0;
197
+ border-radius: 999px;
198
+ background: color-mix(in oklch, var(--warn), var(--panel) 72%);
199
+ color: var(--text);
200
+ font-size: 11px;
201
+ font-weight: 640;
185
202
  font-variant-numeric: tabular-nums;
186
- white-space: nowrap;
187
203
  }
188
- /* schemaVersion 不匹配的 run:榜单不解析它们,这里占位并给出 npx 查看命令。 */
189
- .incompatible-banner {
190
- border: 1px solid color-mix(in oklch, var(--warn), var(--line) 45%);
191
- border-left: 3px solid var(--warn);
192
- border-radius: 8px;
193
- background: color-mix(in oklch, var(--warn), var(--panel) 92%);
194
- padding: 12px 16px;
195
- margin: 0 auto 36px;
196
- font-size: 13px;
197
- }
198
- .incompatible-banner ul {
199
- margin: 8px 0 0;
204
+ .ib-group-desc { color: var(--soft); }
205
+ .ib-group ul {
206
+ margin: 5px 0 0;
200
207
  padding: 0;
201
208
  list-style: none;
202
209
  display: grid;
203
- gap: 6px;
210
+ gap: 3px;
204
211
  }
205
- .incompatible-banner li {
212
+ .ib-group li {
206
213
  display: flex;
207
- flex-wrap: wrap;
208
- align-items: baseline;
209
- gap: 4px 12px;
214
+ align-items: center;
215
+ gap: 8px;
216
+ min-height: 22px;
210
217
  }
211
- .incompatible-banner .ib-dir { font-weight: 600; }
212
- .incompatible-banner .ib-meta { color: var(--soft); }
213
- .incompatible-banner code {
218
+ .ib-dir {
214
219
  font-family: var(--mono, ui-monospace, monospace);
215
- background: color-mix(in oklch, var(--panel-2), var(--panel) 40%);
220
+ font-weight: 560;
221
+ }
222
+ .ib-more {
223
+ color: var(--soft);
224
+ font-size: 12px;
225
+ }
226
+ .ib-copy {
227
+ display: inline-flex;
228
+ align-items: center;
229
+ justify-content: center;
230
+ width: 22px;
231
+ height: 22px;
232
+ flex-shrink: 0;
216
233
  border: 1px solid var(--line);
217
234
  border-radius: 5px;
218
- padding: 1px 7px;
219
- user-select: all;
235
+ background: var(--panel);
236
+ color: var(--muted);
237
+ cursor: pointer;
238
+ transition: color 0.12s, border-color 0.12s;
239
+ }
240
+ .ib-copy svg { width: 12px; height: 12px; stroke: currentColor; fill: none; stroke-width: 2; }
241
+ .ib-copy:hover { color: var(--text); border-color: var(--line-strong); }
242
+ .ib-copy.is-copied { color: var(--good); border-color: color-mix(in oklch, var(--good), var(--line) 40%); }
243
+ .ib-toggle {
244
+ display: inline-flex;
245
+ align-items: center;
246
+ gap: 4px;
247
+ margin-top: 10px;
248
+ border: 0;
249
+ background: none;
250
+ color: var(--muted);
251
+ font: inherit;
252
+ font-size: 12px;
253
+ cursor: pointer;
254
+ padding: 0;
220
255
  }
256
+ .ib-toggle svg { width: 12px; height: 12px; transition: transform 0.12s; }
257
+ .ib-toggle svg.is-open { transform: rotate(90deg); }
258
+ .ib-toggle:hover { color: var(--text); }
221
259
 
222
260
  .section-head {
223
261
  display: flex;
@@ -276,87 +314,7 @@ th {
276
314
  top: 0;
277
315
  z-index: 2;
278
316
  }
279
- th button {
280
- color: inherit;
281
- background: transparent;
282
- border: 0;
283
- padding: 0;
284
- cursor: pointer;
285
- font: inherit;
286
- height: 100%;
287
- }
288
- th button::after {
289
- content: "↕";
290
- margin-left: 6px;
291
- color: var(--soft);
292
- font-size: 11px;
293
- }
294
- th button[data-sorted="asc"]::after { content: "↑"; color: var(--text); }
295
- th button[data-sorted="desc"]::after { content: "↓"; color: var(--text); }
296
- tbody tr.main-row { cursor: pointer; }
297
- tbody tr.main-row:hover { background: color-mix(in oklch, var(--panel-2), var(--panel) 34%); }
298
- tbody tr.is-open { background: var(--panel-2); }
299
317
  tbody tr:last-child td { border-bottom: 0; }
300
- .group-selector {
301
- display: flex;
302
- gap: 12px;
303
- flex-wrap: wrap;
304
- margin-bottom: 20px;
305
- }
306
- .group-card {
307
- flex: 1;
308
- min-width: 140px;
309
- border: 1px solid var(--line);
310
- border-radius: 8px;
311
- padding: 16px 20px;
312
- cursor: pointer;
313
- background: var(--panel);
314
- outline: none;
315
- transition: border-color 120ms, background 120ms;
316
- }
317
- .group-card:hover { border-color: var(--line-strong); background: var(--panel-2); }
318
- .group-card.is-selected { border-color: var(--text); background: var(--panel-2); }
319
- .group-card-name { font-size: 13px; font-weight: 680; color: var(--muted); margin-bottom: 8px; text-transform: uppercase; letter-spacing: 0.05em; }
320
- .group-card-rate { font-size: 30px; font-weight: 760; font-variant-numeric: tabular-nums; line-height: 1; margin-bottom: 6px; }
321
- .group-card-meta { font-size: 12px; color: var(--muted); }
322
- .group-card-time { font-size: 11px; color: var(--soft); margin-top: 2px; }
323
- .cost-score-chart {
324
- border: 1px solid var(--line);
325
- border-radius: 8px;
326
- padding: 16px 20px 10px;
327
- margin-bottom: 20px;
328
- background: var(--panel);
329
- }
330
- .csc-head { font-size: 13px; font-weight: 680; color: var(--muted); margin-bottom: 8px; text-transform: uppercase; letter-spacing: 0.05em; }
331
- /* viewBox 640×300 等比缩放,不设上限会在宽屏被拉成巨图;tooltip 用百分比定位,上限必须设在 wrap 上。 */
332
- .csc-svg-wrap { position: relative; width: 100%; max-width: 760px; }
333
- .csc-svg { width: 100%; height: auto; display: block; overflow: visible; }
334
- .csc-grid line { stroke: var(--line); stroke-width: 1; }
335
- .csc-axis text { fill: var(--muted); font-size: 10px; }
336
- .csc-point { cursor: pointer; }
337
- .csc-point .csc-hit { fill: transparent; }
338
- .csc-point .csc-dot { stroke: var(--panel); stroke-width: 2; }
339
- .csc-point text { fill: var(--text); font-size: 11px; }
340
- .csc-leader { stroke: var(--line-strong); stroke-width: 1; }
341
- .csc-point:hover .csc-dot,
342
- .csc-point:focus-visible .csc-dot { r: 6.5; }
343
- .csc-point:focus-visible { outline: none; }
344
- .csc-tooltip {
345
- position: absolute;
346
- transform: translate(-50%, -100%);
347
- margin-top: -12px;
348
- background: var(--panel);
349
- border: 1px solid var(--line-strong);
350
- border-radius: 6px;
351
- padding: 8px 10px;
352
- font-size: 12px;
353
- white-space: nowrap;
354
- pointer-events: none;
355
- box-shadow: 0 4px 16px color-mix(in oklch, var(--text), transparent 88%);
356
- z-index: 5;
357
- }
358
- .csc-tooltip b { font-size: 13px; }
359
- .csc-tooltip .csc-tooltip-meta { color: var(--muted); margin-top: 2px; }
360
318
  .section-sub-head {
361
319
  display: flex;
362
320
  align-items: center;
@@ -370,28 +328,6 @@ tbody tr:last-child td { border-bottom: 0; }
370
328
  font-weight: 680;
371
329
  color: var(--text);
372
330
  }
373
- .chev {
374
- display: inline-block;
375
- width: 9px;
376
- height: 9px;
377
- border-right: 1.5px solid var(--muted);
378
- border-bottom: 1.5px solid var(--muted);
379
- transform: rotate(-45deg);
380
- margin-right: 12px;
381
- transition: transform 160ms ease;
382
- }
383
- .is-open .chev { transform: rotate(45deg); }
384
- .chev-icon {
385
- display: inline-block;
386
- width: 14px;
387
- height: 14px;
388
- flex-shrink: 0;
389
- margin-right: 9px;
390
- color: var(--muted);
391
- vertical-align: -2px;
392
- transition: transform 160ms ease;
393
- }
394
- .is-open .chev-icon { transform: rotate(90deg); }
395
331
  .name { font-weight: 610; }
396
332
  .sub { color: var(--muted); font-size: 12px; margin-top: 3px; }
397
333
  .num {
@@ -403,167 +339,12 @@ tbody tr:last-child td { border-bottom: 0; }
403
339
  .bad { color: var(--bad); }
404
340
  .warn { color: var(--warn); }
405
341
  .infra-err { color: var(--warn); }
406
- .pill {
407
- display: inline-flex;
408
- align-items: center;
409
- height: 22px;
410
- border: 1px solid var(--line);
411
- border-radius: 999px;
412
- padding: 0 8px;
413
- color: var(--muted);
414
- font-size: 12px;
415
- white-space: nowrap;
416
- }
417
- .detail-cell {
418
- padding: 0;
419
- background: color-mix(in oklch, var(--panel-2), var(--panel) 24%);
420
- }
421
- .detail {
422
- padding: 20px 30px 24px;
423
- border-bottom: 1px solid var(--line);
424
- }
425
- .config-strip {
426
- display: flex;
427
- flex-wrap: wrap;
428
- gap: 8px;
429
- margin: 0 0 18px;
430
- }
431
- .config-chip {
432
- display: inline-flex;
433
- align-items: center;
434
- gap: 6px;
435
- min-height: 26px;
436
- border: 1px solid color-mix(in oklch, var(--line), transparent 16%);
437
- border-radius: 999px;
438
- padding: 4px 9px;
439
- color: var(--muted);
440
- font-size: 12px;
441
- line-height: 1.2;
442
- }
443
- .config-chip b {
444
- color: var(--text);
445
- font-weight: 560;
446
- }
447
- .detail-kpis {
448
- display: grid;
449
- grid-template-columns: repeat(7, minmax(0, 1fr));
450
- border-bottom: 1px solid var(--line);
451
- margin: -2px 0 18px;
452
- }
453
- .detail-kpi {
454
- padding: 0 18px 16px 0;
455
- min-width: 0;
456
- }
457
- .detail-kpi span {
458
- display: block;
459
- color: var(--soft);
460
- font-size: 11px;
461
- text-transform: uppercase;
462
- letter-spacing: 0.07em;
463
- }
464
- .detail-kpi b {
465
- display: block;
466
- margin-top: 7px;
467
- font-size: 18px;
468
- font-weight: 680;
469
- font-variant-numeric: tabular-nums;
470
- }
471
- .detail h3 {
472
- margin: 0 0 14px;
473
- font-size: 13px;
474
- color: var(--muted);
475
- font-weight: 580;
476
- text-transform: uppercase;
477
- letter-spacing: 0.06em;
478
- }
479
- .eval-list {
480
- display: grid;
481
- max-height: 430px;
482
- overflow: auto;
483
- padding-right: 4px;
484
- }
485
- .eval-grid-head,
486
- .eval-item {
487
- display: grid;
488
- grid-template-columns: 76px minmax(220px, 1.1fr) minmax(260px, 1fr) 84px 86px 82px 54px;
489
- column-gap: 16px;
490
- align-items: baseline;
491
- padding: 11px 0;
492
- border-bottom: 1px solid color-mix(in oklch, var(--line), transparent 35%);
493
- font-size: 13px;
494
- }
495
- .eval-grid-head {
496
- position: sticky;
497
- top: 0;
498
- z-index: 1;
499
- background: color-mix(in oklch, var(--panel-2), var(--panel) 24%);
500
- color: var(--soft);
501
- font-size: 11px;
502
- text-transform: uppercase;
503
- letter-spacing: 0.06em;
504
- }
505
342
  .eval-id {
506
343
  display: flex;
507
344
  align-items: center;
508
345
  overflow-wrap: anywhere;
509
346
  font-weight: 560;
510
347
  }
511
- .eval-id:not(:has(.chev-icon)) {
512
- padding-left: 23px; /* 14px 图标 + 9px 间距,单轮 eval 没有展开箭头,靠 padding 补位对齐 */
513
- }
514
- .assertions-cell {
515
- min-width: 0;
516
- display: flex;
517
- align-items: flex-start;
518
- gap: 8px;
519
- overflow-wrap: anywhere;
520
- line-height: 1.25;
521
- }
522
- .assertions {
523
- min-width: 0;
524
- flex: 1;
525
- color: color-mix(in oklch, var(--muted), var(--text) 6%);
526
- display: -webkit-box;
527
- -webkit-box-orient: vertical;
528
- -webkit-line-clamp: 2;
529
- line-clamp: 2;
530
- overflow: hidden;
531
- }
532
- .assertions-link { cursor: pointer; }
533
- .assertions-link:hover { color: var(--text); text-decoration: underline; text-underline-offset: 2px; }
534
- .copy-reason {
535
- width: 26px;
536
- height: 26px;
537
- display: inline-grid;
538
- place-items: center;
539
- flex: 0 0 auto;
540
- background: var(--panel);
541
- border: 1px solid var(--line);
542
- border-radius: 6px;
543
- cursor: pointer;
544
- color: var(--muted);
545
- font-family: inherit;
546
- opacity: 0;
547
- transition: opacity 0.1s, color 0.1s, border-color 0.1s, background 0.1s;
548
- }
549
- .copy-reason svg {
550
- width: 14px;
551
- height: 14px;
552
- fill: none;
553
- stroke: currentColor;
554
- stroke-width: 2;
555
- stroke-linecap: round;
556
- stroke-linejoin: round;
557
- }
558
- .eval-item:hover .copy-reason,
559
- .copy-reason:focus-visible,
560
- .copy-reason.is-copied { opacity: 1; }
561
- .copy-reason:hover { color: var(--text); border-color: var(--line-strong); }
562
- .copy-reason.is-copied {
563
- color: var(--good);
564
- border-color: color-mix(in oklch, var(--good), var(--line) 40%);
565
- background: color-mix(in oklch, var(--good), transparent 88%);
566
- }
567
348
  .copy-all-errors {
568
349
  height: 36px;
569
350
  display: inline-flex;
@@ -652,7 +433,7 @@ tbody tr:last-child td { border-bottom: 0; }
652
433
  border-bottom: none;
653
434
  }
654
435
  /* Modal */
655
- /* modal chrome(backdrop / panel / header / title / close / body / outcome chip)→
436
+ /* modal chrome(backdrop / panel / header / title / close / body / verdict chip)→
656
437
  迁到 components/ui/dialog.tsx + components/ui/badge.tsx + AttemptModal 的 Tailwind utilities。 */
657
438
 
658
439
  /* No-source modal body: same visual language as the code view (checks + transcript). */
@@ -942,10 +723,6 @@ a.brand:hover { opacity: 0.75; }
942
723
  main { width: min(100vw - 28px, 1120px); padding-top: 42px; }
943
724
  .hero { text-align: left; margin-left: 0; }
944
725
  .meta { justify-content: flex-start; }
945
- .summary { grid-template-columns: repeat(2, minmax(0, 1fr)); }
946
- .metric { grid-template-columns: 1fr; row-gap: 8px; }
947
- .metric:nth-child(2) { border-right: 0; }
948
- .metric:nth-child(-n+2) { border-bottom: 1px solid var(--line); }
949
726
  .section-head { align-items: stretch; flex-direction: column; }
950
727
  .search { width: 100%; min-width: 0; }
951
728
  .detail { padding: 18px 16px 20px; }
@@ -955,7 +732,7 @@ a.brand:hover { opacity: 0.75; }
955
732
  .eval-item { grid-template-columns: 72px minmax(180px, 1fr) minmax(220px, 1fr) 76px 78px 72px 50px; }
956
733
  }
957
734
 
958
- /* ───────────────────────── Modal: error + code view (outcome chip → components/ui/badge.tsx) ───────────────────────── */
735
+ /* ───────────────────────── Modal: error + code view (verdict chip → components/ui/badge.tsx) ───────────────────────── */
959
736
  .modal-error {
960
737
  font-family: ui-monospace, SFMono-Regular, Menlo, monospace;
961
738
  font-size: 12px;
@@ -975,7 +752,8 @@ a.brand:hover { opacity: 0.75; }
975
752
  border: 1px solid var(--line);
976
753
  border-radius: 10px;
977
754
  overflow: hidden;
978
- background: var(--panel-2);
755
+ /* review 是独立阅读面;源码区必须完全不透底,不能让报告纹理穿进来。 */
756
+ background: var(--panel);
979
757
  }
980
758
  .code-file-head {
981
759
  font-family: ui-monospace, SFMono-Regular, Menlo, monospace;
@@ -989,13 +767,25 @@ a.brand:hover { opacity: 0.75; }
989
767
  font-family: ui-monospace, SFMono-Regular, Menlo, monospace;
990
768
  font-size: 12.5px;
991
769
  line-height: 1.7;
992
- }
770
+ /* 整块横向滚动(github-diff 式),不是逐行各滚各的——见下面 .code-line 的
771
+ width: max-content:一行的真实宽度撑开这个滚动容器,滚动条对齐着整个代码块,
772
+ 而不是消失在每行自己那条不可见的 scrollbar 里(旧写法读长 prompt 行直接读丢)。 */
773
+ overflow-x: auto;
774
+ }
775
+ .code-lines::-webkit-scrollbar { height: 8px; }
776
+ .code-lines::-webkit-scrollbar-track { background: transparent; }
777
+ .code-lines::-webkit-scrollbar-thumb { background: var(--line-strong); border-radius: 4px; }
778
+ .code-lines::-webkit-scrollbar-thumb:hover { background: var(--soft); }
993
779
 
994
780
  .code-line {
995
781
  display: grid;
996
782
  grid-template-columns: 46px 22px minmax(0, 1fr) auto;
997
783
  align-items: center;
998
784
  border-left: 3px solid transparent;
785
+ /* 行宽跟着内容走(短行仍撑满可见区域,长行把 .code-lines 撑宽触发横向滚动),
786
+ 配合 .code-lines 的 overflow-x 实现整块同步滚动。 */
787
+ width: max-content;
788
+ min-width: 100%;
999
789
  }
1000
790
  .code-line.line-clickable { cursor: pointer; }
1001
791
  .code-line.line-clickable:hover { background: color-mix(in oklch, var(--panel), var(--line) 45%); }
@@ -1011,31 +801,32 @@ a.brand:hover { opacity: 0.75; }
1011
801
  .code-line.line-send.line-clickable:hover { background: color-mix(in oklch, var(--blue), var(--panel) 82%); }
1012
802
  .code-line.is-open { background: color-mix(in oklch, var(--panel), var(--line) 34%); }
1013
803
 
1014
- /* github-diff 式整行红绿:行底色 + 更深的行号格 + 实色左缘(过=绿/加,不过=红/减)。 */
804
+ /* github-diff 式整行红绿:明确色块 + 更深 gutter + 实色左缘。
805
+ 状态色占比必须足够高,暗色模式不能退化成只有一个很淡的图标。 */
1015
806
  .code-line.line-pass {
1016
- background: color-mix(in oklch, var(--good), var(--panel) 74%);
807
+ background: color-mix(in oklch, var(--good), var(--panel) 58%);
1017
808
  border-left-color: var(--good);
1018
809
  }
1019
810
  .code-line.line-pass .ln {
1020
- background: color-mix(in oklch, var(--good), var(--panel) 52%);
1021
- color: color-mix(in oklch, var(--good), var(--text) 45%);
811
+ background: color-mix(in oklch, var(--good), var(--panel) 34%);
812
+ color: var(--text);
1022
813
  }
1023
814
  .code-line.line-fail {
1024
- background: color-mix(in oklch, var(--bad), var(--panel) 74%);
815
+ background: color-mix(in oklch, var(--bad), var(--panel) 58%);
1025
816
  border-left-color: var(--bad);
1026
817
  }
1027
818
  .code-line.line-fail .ln {
1028
- background: color-mix(in oklch, var(--bad), var(--panel) 52%);
1029
- color: color-mix(in oklch, var(--bad), var(--text) 45%);
819
+ background: color-mix(in oklch, var(--bad), var(--panel) 34%);
820
+ color: var(--text);
1030
821
  }
1031
- /* soft 断言没过阈值:不影响 outcome,视觉上跟真正的 gate 失败(红)区分开,用 warn(黄)。 */
822
+ /* soft 断言没过阈值:不影响 verdict,视觉上跟真正的 gate 失败(红)区分开,用 warn(黄)。 */
1032
823
  .code-line.line-warn {
1033
- background: color-mix(in oklch, var(--warn), var(--panel) 74%);
824
+ background: color-mix(in oklch, var(--warn), var(--panel) 58%);
1034
825
  border-left-color: var(--warn);
1035
826
  }
1036
827
  .code-line.line-warn .ln {
1037
- background: color-mix(in oklch, var(--warn), var(--panel) 52%);
1038
- color: color-mix(in oklch, var(--warn), var(--text) 45%);
828
+ background: color-mix(in oklch, var(--warn), var(--panel) 34%);
829
+ color: var(--text);
1039
830
  }
1040
831
  .code-line.line-pass.line-clickable:hover { background: color-mix(in oklch, var(--good), var(--panel) 64%); }
1041
832
  .code-line.line-fail.line-clickable:hover { background: color-mix(in oklch, var(--bad), var(--panel) 64%); }
@@ -1060,8 +851,7 @@ a.brand:hover { opacity: 0.75; }
1060
851
  .gstat.warn { color: var(--warn); }
1061
852
  .gsend { width: 13px; height: 13px; flex-shrink: 0; color: var(--blue); }
1062
853
 
1063
- .ctext { white-space: pre; overflow-x: auto; color: var(--text); padding-left: 8px; }
1064
- .ctext::-webkit-scrollbar { height: 0; }
854
+ .ctext { white-space: pre; color: var(--text); padding-left: 8px; }
1065
855
  .lbadges { padding-right: 12px; }
1066
856
 
1067
857
  .lbadges { display: inline-flex; gap: 6px; align-items: center; white-space: nowrap; justify-self: end; }