@iblai/iblai-js 2.76.0 → 2.77.0

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 (207) hide show
  1. package/dist/data-layer/playwright/evaluation-tab-helpers.d.ts +181 -68
  2. package/dist/playwright/index.cjs +414 -164
  3. package/dist/playwright/index.cjs.map +1 -1
  4. package/dist/playwright/index.d.ts +181 -69
  5. package/dist/playwright/index.esm.js +407 -165
  6. package/dist/playwright/index.esm.js.map +1 -1
  7. package/dist/playwright/playwright/evaluation-tab-helpers.d.ts +181 -68
  8. package/dist/security/playwright/evaluation-tab-helpers.d.ts +181 -68
  9. package/dist/web-containers/playwright/evaluation-tab-helpers.d.ts +181 -68
  10. package/dist/web-containers/source/activity-empty-state.js +2 -2
  11. package/dist/web-containers/source/activity-empty-state.js.map +1 -1
  12. package/dist/web-containers/source/activity-tab/activity-empty-state.js +2 -2
  13. package/dist/web-containers/source/activity-tab/activity-empty-state.js.map +1 -1
  14. package/dist/web-containers/source/add-items-modal.js +26 -91
  15. package/dist/web-containers/source/add-items-modal.js.map +1 -1
  16. package/dist/web-containers/source/benchmark-combobox.js +2 -2
  17. package/dist/web-containers/source/benchmark-combobox.js.map +1 -1
  18. package/dist/web-containers/source/benchmark-items-modal.js +28 -7
  19. package/dist/web-containers/source/benchmark-items-modal.js.map +1 -1
  20. package/dist/web-containers/source/benchmarks/add-items-modal.js +26 -91
  21. package/dist/web-containers/source/benchmarks/add-items-modal.js.map +1 -1
  22. package/dist/web-containers/source/benchmarks/benchmark-items-modal.js +28 -7
  23. package/dist/web-containers/source/benchmarks/benchmark-items-modal.js.map +1 -1
  24. package/dist/web-containers/source/benchmarks/create-benchmark-modal.js +166 -20
  25. package/dist/web-containers/source/benchmarks/create-benchmark-modal.js.map +1 -1
  26. package/dist/web-containers/source/benchmarks/delete-item-confirm.js +10 -9
  27. package/dist/web-containers/source/benchmarks/delete-item-confirm.js.map +1 -1
  28. package/dist/web-containers/source/benchmarks/evals-ui.js +39 -0
  29. package/dist/web-containers/source/benchmarks/evals-ui.js.map +1 -0
  30. package/dist/web-containers/source/benchmarks/index.js +30 -11
  31. package/dist/web-containers/source/benchmarks/index.js.map +1 -1
  32. package/dist/web-containers/source/benchmarks/qa-form.js +233 -0
  33. package/dist/web-containers/source/benchmarks/qa-form.js.map +1 -0
  34. package/dist/web-containers/source/components/activity-tab/activity-empty-state.js +2 -2
  35. package/dist/web-containers/source/components/activity-tab/activity-empty-state.js.map +1 -1
  36. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js +2 -2
  37. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js.map +1 -1
  38. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js +6 -6
  39. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js.map +1 -1
  40. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js +7 -6
  41. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js.map +1 -1
  42. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js +51 -43
  43. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js.map +1 -1
  44. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/index.js +68 -37
  45. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/index.js.map +1 -1
  46. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/labels.js +68 -46
  47. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/labels.js.map +1 -1
  48. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js +13 -6
  49. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js.map +1 -1
  50. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js +16 -7
  51. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js.map +1 -1
  52. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/utils.js +14 -1
  53. package/dist/web-containers/source/components/modals/edit-mentor-modal/tabs/evaluation-tab/utils.js.map +1 -1
  54. package/dist/web-containers/source/components/platform-sidebar/index.js +2 -0
  55. package/dist/web-containers/source/components/platform-sidebar/index.js.map +1 -1
  56. package/dist/web-containers/source/components/profile/benchmarks/add-items-modal.js +26 -91
  57. package/dist/web-containers/source/components/profile/benchmarks/add-items-modal.js.map +1 -1
  58. package/dist/web-containers/source/components/profile/benchmarks/benchmark-items-modal.js +28 -7
  59. package/dist/web-containers/source/components/profile/benchmarks/benchmark-items-modal.js.map +1 -1
  60. package/dist/web-containers/source/components/profile/benchmarks/create-benchmark-modal.js +166 -20
  61. package/dist/web-containers/source/components/profile/benchmarks/create-benchmark-modal.js.map +1 -1
  62. package/dist/web-containers/source/components/profile/benchmarks/delete-item-confirm.js +10 -9
  63. package/dist/web-containers/source/components/profile/benchmarks/delete-item-confirm.js.map +1 -1
  64. package/dist/web-containers/source/components/profile/benchmarks/evals-ui.js +39 -0
  65. package/dist/web-containers/source/components/profile/benchmarks/evals-ui.js.map +1 -0
  66. package/dist/web-containers/source/components/profile/benchmarks/index.js +30 -11
  67. package/dist/web-containers/source/components/profile/benchmarks/index.js.map +1 -1
  68. package/dist/web-containers/source/components/profile/benchmarks/qa-form.js +233 -0
  69. package/dist/web-containers/source/components/profile/benchmarks/qa-form.js.map +1 -0
  70. package/dist/web-containers/source/components/rich-text-editor.js +24 -3
  71. package/dist/web-containers/source/components/rich-text-editor.js.map +1 -1
  72. package/dist/web-containers/source/create-benchmark-modal.js +166 -20
  73. package/dist/web-containers/source/create-benchmark-modal.js.map +1 -1
  74. package/dist/web-containers/source/delete-experiment-modal.js +6 -6
  75. package/dist/web-containers/source/delete-experiment-modal.js.map +1 -1
  76. package/dist/web-containers/source/delete-item-confirm.js +10 -9
  77. package/dist/web-containers/source/delete-item-confirm.js.map +1 -1
  78. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js +2 -2
  79. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js.map +1 -1
  80. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js +6 -6
  81. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js.map +1 -1
  82. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js +7 -6
  83. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js.map +1 -1
  84. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js +51 -43
  85. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js.map +1 -1
  86. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/index.js +68 -37
  87. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/index.js.map +1 -1
  88. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/labels.js +68 -46
  89. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/labels.js.map +1 -1
  90. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js +13 -6
  91. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js.map +1 -1
  92. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js +16 -7
  93. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js.map +1 -1
  94. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/utils.js +14 -1
  95. package/dist/web-containers/source/edit-mentor-modal/tabs/evaluation-tab/utils.js.map +1 -1
  96. package/dist/web-containers/source/en.json.js +100 -74
  97. package/dist/web-containers/source/en.json.js.map +1 -1
  98. package/dist/web-containers/source/es.json.js +81 -55
  99. package/dist/web-containers/source/es.json.js.map +1 -1
  100. package/dist/web-containers/source/evals-ui.js +39 -0
  101. package/dist/web-containers/source/evals-ui.js.map +1 -0
  102. package/dist/web-containers/source/evaluate-llm-picker.js +7 -6
  103. package/dist/web-containers/source/evaluate-llm-picker.js.map +1 -1
  104. package/dist/web-containers/source/evaluation-tab/benchmark-combobox.js +2 -2
  105. package/dist/web-containers/source/evaluation-tab/benchmark-combobox.js.map +1 -1
  106. package/dist/web-containers/source/evaluation-tab/delete-experiment-modal.js +6 -6
  107. package/dist/web-containers/source/evaluation-tab/delete-experiment-modal.js.map +1 -1
  108. package/dist/web-containers/source/evaluation-tab/evaluate-llm-picker.js +7 -6
  109. package/dist/web-containers/source/evaluation-tab/evaluate-llm-picker.js.map +1 -1
  110. package/dist/web-containers/source/evaluation-tab/experiment-detail-modal.js +51 -43
  111. package/dist/web-containers/source/evaluation-tab/experiment-detail-modal.js.map +1 -1
  112. package/dist/web-containers/source/evaluation-tab/index.js +68 -37
  113. package/dist/web-containers/source/evaluation-tab/index.js.map +1 -1
  114. package/dist/web-containers/source/evaluation-tab/labels.js +68 -46
  115. package/dist/web-containers/source/evaluation-tab/labels.js.map +1 -1
  116. package/dist/web-containers/source/evaluation-tab/llm-judge-modal.js +13 -6
  117. package/dist/web-containers/source/evaluation-tab/llm-judge-modal.js.map +1 -1
  118. package/dist/web-containers/source/evaluation-tab/start-experiment-modal.js +16 -7
  119. package/dist/web-containers/source/evaluation-tab/start-experiment-modal.js.map +1 -1
  120. package/dist/web-containers/source/evaluation-tab/utils.js +14 -1
  121. package/dist/web-containers/source/evaluation-tab/utils.js.map +1 -1
  122. package/dist/web-containers/source/experiment-detail-modal.js +51 -43
  123. package/dist/web-containers/source/experiment-detail-modal.js.map +1 -1
  124. package/dist/web-containers/source/fr.json.js +82 -56
  125. package/dist/web-containers/source/fr.json.js.map +1 -1
  126. package/dist/web-containers/source/i18n/messages/en.json.js +100 -74
  127. package/dist/web-containers/source/i18n/messages/en.json.js.map +1 -1
  128. package/dist/web-containers/source/i18n/messages/es.json.js +81 -55
  129. package/dist/web-containers/source/i18n/messages/es.json.js.map +1 -1
  130. package/dist/web-containers/source/i18n/messages/fr.json.js +82 -56
  131. package/dist/web-containers/source/i18n/messages/fr.json.js.map +1 -1
  132. package/dist/web-containers/source/i18n/messages/zh.json.js +71 -45
  133. package/dist/web-containers/source/i18n/messages/zh.json.js.map +1 -1
  134. package/dist/web-containers/source/llm-judge-modal.js +13 -6
  135. package/dist/web-containers/source/llm-judge-modal.js.map +1 -1
  136. package/dist/web-containers/source/messages/en.json.js +100 -74
  137. package/dist/web-containers/source/messages/en.json.js.map +1 -1
  138. package/dist/web-containers/source/messages/es.json.js +81 -55
  139. package/dist/web-containers/source/messages/es.json.js.map +1 -1
  140. package/dist/web-containers/source/messages/fr.json.js +82 -56
  141. package/dist/web-containers/source/messages/fr.json.js.map +1 -1
  142. package/dist/web-containers/source/messages/zh.json.js +71 -45
  143. package/dist/web-containers/source/messages/zh.json.js.map +1 -1
  144. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js +2 -2
  145. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/benchmark-combobox.js.map +1 -1
  146. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js +6 -6
  147. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/delete-experiment-modal.js.map +1 -1
  148. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js +7 -6
  149. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/evaluate-llm-picker.js.map +1 -1
  150. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js +51 -43
  151. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/experiment-detail-modal.js.map +1 -1
  152. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/index.js +68 -37
  153. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/index.js.map +1 -1
  154. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/labels.js +68 -46
  155. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/labels.js.map +1 -1
  156. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js +13 -6
  157. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/llm-judge-modal.js.map +1 -1
  158. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js +16 -7
  159. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/start-experiment-modal.js.map +1 -1
  160. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/utils.js +14 -1
  161. package/dist/web-containers/source/modals/edit-mentor-modal/tabs/evaluation-tab/utils.js.map +1 -1
  162. package/dist/web-containers/source/platform-sidebar/index.js +2 -0
  163. package/dist/web-containers/source/platform-sidebar/index.js.map +1 -1
  164. package/dist/web-containers/source/profile/benchmarks/add-items-modal.js +26 -91
  165. package/dist/web-containers/source/profile/benchmarks/add-items-modal.js.map +1 -1
  166. package/dist/web-containers/source/profile/benchmarks/benchmark-items-modal.js +28 -7
  167. package/dist/web-containers/source/profile/benchmarks/benchmark-items-modal.js.map +1 -1
  168. package/dist/web-containers/source/profile/benchmarks/create-benchmark-modal.js +166 -20
  169. package/dist/web-containers/source/profile/benchmarks/create-benchmark-modal.js.map +1 -1
  170. package/dist/web-containers/source/profile/benchmarks/delete-item-confirm.js +10 -9
  171. package/dist/web-containers/source/profile/benchmarks/delete-item-confirm.js.map +1 -1
  172. package/dist/web-containers/source/profile/benchmarks/evals-ui.js +39 -0
  173. package/dist/web-containers/source/profile/benchmarks/evals-ui.js.map +1 -0
  174. package/dist/web-containers/source/profile/benchmarks/index.js +30 -11
  175. package/dist/web-containers/source/profile/benchmarks/index.js.map +1 -1
  176. package/dist/web-containers/source/profile/benchmarks/qa-form.js +233 -0
  177. package/dist/web-containers/source/profile/benchmarks/qa-form.js.map +1 -0
  178. package/dist/web-containers/source/qa-form.js +233 -0
  179. package/dist/web-containers/source/qa-form.js.map +1 -0
  180. package/dist/web-containers/source/rich-text-editor.js +24 -3
  181. package/dist/web-containers/source/rich-text-editor.js.map +1 -1
  182. package/dist/web-containers/source/start-experiment-modal.js +16 -7
  183. package/dist/web-containers/source/start-experiment-modal.js.map +1 -1
  184. package/dist/web-containers/source/tabs/evaluation-tab/benchmark-combobox.js +2 -2
  185. package/dist/web-containers/source/tabs/evaluation-tab/benchmark-combobox.js.map +1 -1
  186. package/dist/web-containers/source/tabs/evaluation-tab/delete-experiment-modal.js +6 -6
  187. package/dist/web-containers/source/tabs/evaluation-tab/delete-experiment-modal.js.map +1 -1
  188. package/dist/web-containers/source/tabs/evaluation-tab/evaluate-llm-picker.js +7 -6
  189. package/dist/web-containers/source/tabs/evaluation-tab/evaluate-llm-picker.js.map +1 -1
  190. package/dist/web-containers/source/tabs/evaluation-tab/experiment-detail-modal.js +51 -43
  191. package/dist/web-containers/source/tabs/evaluation-tab/experiment-detail-modal.js.map +1 -1
  192. package/dist/web-containers/source/tabs/evaluation-tab/index.js +68 -37
  193. package/dist/web-containers/source/tabs/evaluation-tab/index.js.map +1 -1
  194. package/dist/web-containers/source/tabs/evaluation-tab/labels.js +68 -46
  195. package/dist/web-containers/source/tabs/evaluation-tab/labels.js.map +1 -1
  196. package/dist/web-containers/source/tabs/evaluation-tab/llm-judge-modal.js +13 -6
  197. package/dist/web-containers/source/tabs/evaluation-tab/llm-judge-modal.js.map +1 -1
  198. package/dist/web-containers/source/tabs/evaluation-tab/start-experiment-modal.js +16 -7
  199. package/dist/web-containers/source/tabs/evaluation-tab/start-experiment-modal.js.map +1 -1
  200. package/dist/web-containers/source/tabs/evaluation-tab/utils.js +14 -1
  201. package/dist/web-containers/source/tabs/evaluation-tab/utils.js.map +1 -1
  202. package/dist/web-containers/source/tsconfig.build.tsbuildinfo +1 -1
  203. package/dist/web-containers/source/web-containers/tsconfig.build.tsbuildinfo +1 -1
  204. package/dist/web-containers/source/zh.json.js +71 -45
  205. package/dist/web-containers/source/zh.json.js.map +1 -1
  206. package/dist/web-utils/playwright/evaluation-tab-helpers.d.ts +181 -68
  207. package/package.json +4 -4
@@ -6829,12 +6829,18 @@ async function expectLogDetailsStatus(scope, status) {
6829
6829
  * `aria-labelledby` win the accessible-name computation — and then resolve
6830
6830
  * every in-dialog control from that scoped locator. This matters here more
6831
6831
  * than usual:
6832
- * - the toolbar "New Evaluation" / "Manage benchmarks" buttons share their
6833
- * text with the dialog titles they open;
6834
- * - the LLM-judge dialog's title AND submit button both read "Evaluate";
6835
- * - dialogs nest three deep (Manage benchmarks → benchmark items → add
6836
- * Q&A), and the add-Q&A dialog has Manual / CSV Upload sub-tabs whose
6837
- * footers each have their own Cancel button.
6832
+ * - "New Evaluation", "Manage Benchmarks", "New Benchmark" and "New Review"
6833
+ * are each both a button and the title of the dialog it opens;
6834
+ * - "Add Q&A" is a button on the Evals tab and in a benchmark's Q&A dialog,
6835
+ * and starts the title of the dialog both open;
6836
+ * - dialogs nest three deep (Manage Benchmarks → benchmark Q&A → Add Q&A),
6837
+ * and the New Benchmark and Add Q&A dialogs hold the same Manual / CSV
6838
+ * Upload form;
6839
+ * - the New Benchmark dialog is two steps with their own tabs (Details,
6840
+ * Q&A), so its Q&A step is a tabpanel around the form's own tabpanels.
6841
+ *
6842
+ * Every form label has an info button named "More info about <label>", so a
6843
+ * field is looked up by its label with `exact: true`.
6838
6844
  *
6839
6845
  * The helpers assume the Edit Mentor dialog is already open. Call
6840
6846
  * `switchToEvaluationTab(page)` first; from then on, every helper accepts
@@ -6858,10 +6864,18 @@ const EVALS_LABELS = {
6858
6864
  actions: {
6859
6865
  newExperiment: 'New Evaluation',
6860
6866
  manageBenchmarks: 'Manage Benchmarks',
6861
- viewResults: 'View results',
6862
- evaluate: 'New review',
6867
+ /** The no-benchmarks empty state's button; opens the New Benchmark dialog. */
6868
+ newBenchmark: 'New Benchmark',
6869
+ /** The tab's own button for a benchmark with no Q&A; opens the Add Q&A dialog. */
6870
+ addQa: 'Add Q&A',
6871
+ viewResults: 'View Results',
6872
+ evaluate: 'New Review',
6863
6873
  exportCsv: 'Export CSV',
6864
- checkStatus: 'Check status',
6874
+ checkStatus: 'Check Status',
6875
+ /** Accessible description of View Results / New Review / Export CSV until the evaluation completes. */
6876
+ notReadyHint: 'Available once the evaluation completes',
6877
+ /** Accessible description of the same three items on an evaluation that failed. */
6878
+ failedHint: 'Not available because the evaluation failed',
6865
6879
  delete: 'Delete',
6866
6880
  },
6867
6881
  checkStatusToast: {
@@ -6869,35 +6883,67 @@ const EVALS_LABELS = {
6869
6883
  completed: 'Evaluation completed',
6870
6884
  failed: 'Evaluation failed',
6871
6885
  },
6872
- emptyBenchmarks: 'No benchmarks yet. Use "Manage Benchmarks" above to create one.',
6886
+ emptyBenchmarks: 'Your organization has no benchmarks yet. Create one to evaluate this agent.',
6887
+ /** `data-testid` on the no-benchmarks empty state (sentence + "New Benchmark"). */
6888
+ noBenchmarksTestId: 'evaluation-no-benchmarks',
6889
+ /** Shown for a benchmark with no Q&A, and as the hover text of the disabled "New Evaluation". */
6890
+ emptyQa: 'This benchmark has no Q&A yet. Add at least one question to run an evaluation.',
6891
+ /** `data-testid` on the empty state of a benchmark with no Q&A and no evaluations. */
6892
+ noQaEmptyTestId: 'evaluation-no-qa-empty',
6893
+ /** `data-testid` on the line above the runs table of a benchmark with evaluations but no Q&A. */
6894
+ noQaNoticeTestId: 'evaluation-no-qa-notice',
6895
+ /** Accessible name of every dialog's own close (X) button. */
6896
+ dialogClose: 'Close',
6873
6897
  startDialog: {
6874
6898
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6875
6899
  dialogName: 'start-evaluation',
6876
- experimentNameLabel: 'Evaluation name',
6900
+ experimentNameLabel: 'Evaluation Name',
6877
6901
  cancel: 'Cancel',
6878
6902
  start: 'Start Evaluation',
6903
+ /** Toast after a start; `{name}` is the evaluation's name. */
6904
+ successToast: 'Evaluation started: {name}',
6879
6905
  },
6880
6906
  manageBenchmarksDialog: {
6881
6907
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6882
6908
  dialogName: 'manage-benchmarks',
6909
+ title: 'Manage Benchmarks',
6883
6910
  searchAriaLabel: 'Search benchmarks',
6884
6911
  newBenchmark: 'New Benchmark',
6885
6912
  empty: 'No benchmarks found',
6913
+ emptyNone: 'Your organization has no benchmarks yet.',
6886
6914
  },
6887
6915
  createBenchmarkDialog: {
6888
6916
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6889
6917
  dialogName: 'create-benchmark',
6918
+ /** Step tab (`role="tab"`) holding Name and Description. */
6919
+ detailsTab: 'Details',
6920
+ /** Step tab holding the Manual / CSV Upload form (labels under `addItemsDialog`); disabled until Name has text. */
6921
+ qaTab: 'Q&A',
6890
6922
  nameLabel: 'Name',
6923
+ /** Under Name (`role="alert"`) when "Next" or "Create" refuses a name that could never be opened again. */
6924
+ nameInvalid: 'Name cannot contain a slash (/) or be made only of dots.',
6925
+ /** Under Name (`role="alert"`) when "Create" finds a benchmark that already has this name; the dialog goes back to Details. */
6926
+ nameTaken: 'A benchmark with this name already exists. Enter a different name.',
6891
6927
  descriptionLabel: 'Description',
6928
+ /** `data-testid` on the note at the top of the Q&A step. */
6929
+ qaHintTestId: 'benchmark-qa-hint',
6930
+ qaHint: 'You can skip this and add Q&A later. A benchmark needs at least one question before you can run an evaluation.',
6892
6931
  cancel: 'Cancel',
6932
+ /** The Details step's button; disabled until Name has text. */
6933
+ next: 'Next',
6934
+ /** The Q&A step's button; works with or without Q&A. */
6893
6935
  create: 'Create',
6936
+ /** Toast for a benchmark created without Q&A. With Q&A it reads "Benchmark created with N Q&A pairs". */
6937
+ successToastNoQa: 'Benchmark created. Add Q&A before you run an evaluation.',
6894
6938
  },
6895
6939
  benchmarkItemsDialog: {
6896
6940
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6897
6941
  dialogName: 'benchmark-items',
6942
+ /** Column headers of the Q&A table. Below 640px the ANSWER column is hidden and each answer sits under its question. */
6943
+ columns: ['QUESTION', 'ANSWER', 'ACTIONS'],
6898
6944
  addItems: 'Add Q&A',
6899
6945
  deleteItemAriaLabel: 'Delete Q&A',
6900
- empty: 'No Q&A in this benchmark yet',
6946
+ empty: 'This benchmark has no Q&A yet.',
6901
6947
  },
6902
6948
  addItemsDialog: {
6903
6949
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
@@ -6906,10 +6952,12 @@ const EVALS_LABELS = {
6906
6952
  csvTab: 'CSV Upload',
6907
6953
  questionLabel: 'Question',
6908
6954
  answerLabel: 'Answer',
6909
- addRow: 'Add row',
6910
- removeRowAriaLabel: 'Remove row',
6955
+ addRow: 'Add Row',
6956
+ removeRowAriaLabel: 'Remove Row',
6911
6957
  add: 'Add',
6912
- csvFileLabel: 'CSV file',
6958
+ csvFileLabel: 'CSV File',
6959
+ /** Under CSV File (`role="alert"`) for a file whose first row has no `input` column; the submit button is disabled. */
6960
+ csvNoInputColumn: 'The first row of this file needs a column named "input". Download the sample CSV to see the format.',
6913
6961
  upload: 'Upload',
6914
6962
  cancel: 'Cancel',
6915
6963
  },
@@ -6923,14 +6971,14 @@ const EVALS_LABELS = {
6923
6971
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6924
6972
  dialogName: 'evaluation-detail',
6925
6973
  refreshAriaLabel: 'Refresh',
6926
- newReview: 'New review',
6927
- waitingHint: 'Waiting for the evaluation to produce traces. This page auto-refreshes.',
6974
+ newReview: 'New Review',
6975
+ waitingHint: 'Waiting for the first responses. This page refreshes by itself.',
6928
6976
  noScoresHint: 'No scores yet',
6929
- scoreNameLabel: 'Score name',
6977
+ scoreNameLabel: 'Score Name',
6930
6978
  scoreValueLabel: 'Value',
6931
6979
  scoreCommentLabel: 'Comment',
6932
6980
  addScore: 'Add',
6933
- removeScoreAriaLabel: 'Remove score',
6981
+ removeScoreAriaLabel: 'Delete Score',
6934
6982
  reviewsHeading: 'Reviews',
6935
6983
  reviewDetailsAriaLabel: 'Toggle review details',
6936
6984
  reviewStatus: {
@@ -6943,12 +6991,13 @@ const EVALS_LABELS = {
6943
6991
  judgeDialog: {
6944
6992
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6945
6993
  dialogName: 'llm-judge',
6994
+ title: 'New Review',
6946
6995
  criteriaLabel: 'Criteria',
6947
- scoreNameLabel: 'Score name',
6948
- maxConcurrencyLabel: 'Max concurrency',
6996
+ scoreNameLabel: 'Score Name',
6997
+ maxConcurrencyLabel: 'Max Concurrency',
6949
6998
  selectorAriaLabel: 'LLM Model Selector',
6950
6999
  cancel: 'Cancel',
6951
- submit: 'Evaluate',
7000
+ submit: 'Start Review',
6952
7001
  },
6953
7002
  llmPickerDialog: {
6954
7003
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
@@ -6960,6 +7009,8 @@ const EVALS_LABELS = {
6960
7009
  deleteDialog: {
6961
7010
  /** `aria-label` on the dialog's `DialogContent` — scopes in-dialog locators. */
6962
7011
  dialogName: 'delete-evaluation',
7012
+ /** The dialog's question; `{runName}` is the evaluation's name. */
7013
+ confirmation: 'Are you sure you want to delete the evaluation "{runName}"? This action cannot be undone.',
6963
7014
  cancel: 'Cancel',
6964
7015
  confirm: 'Delete',
6965
7016
  },
@@ -6981,6 +7032,15 @@ function escapeRegExp(value) {
6981
7032
  function dialogByName(scope, dialogName) {
6982
7033
  return asPage$3(scope).locator(`[role="dialog"][aria-label="${dialogName}"]`);
6983
7034
  }
7035
+ /**
7036
+ * A toast showing `text`. A string must be the toast's whole text, case
7037
+ * included. Scoped to the toast itself, because the same words can sit
7038
+ * elsewhere on the page (a disabled row action's hidden reason ends in
7039
+ * "the evaluation failed").
7040
+ */
7041
+ function getToast(scope, text) {
7042
+ return asPage$3(scope).locator('[data-sonner-toast]').getByText(text, { exact: true });
7043
+ }
6984
7044
  // ── Tab navigation ─────────────────────────────────────────────────────
6985
7045
  /**
6986
7046
  * Check whether the Evals tab is rendered in the Edit Mentor dialog.
@@ -6999,14 +7059,14 @@ async function isEvaluationTabVisible(page) {
6999
7059
  }
7000
7060
  /**
7001
7061
  * Switch to the Evals tab. Assumes the Edit Mentor dialog is open. Waits
7002
- * for the toolbar "Manage benchmarks" button (unique to this pane) before
7062
+ * for the toolbar "Manage Benchmarks" button (unique to this pane) before
7003
7063
  * returning, so callers can rely on the panel being interactive.
7004
7064
  */
7005
7065
  async function switchToEvaluationTab(page) {
7006
7066
  const tab = page.getByRole('tab', { name: EVALS_LABELS.tabName, exact: true });
7007
7067
  await expect(tab).toBeVisible({ timeout: 10000 });
7008
7068
  await tab.click();
7009
- await expect(page.getByRole('button', { name: EVALS_LABELS.actions.manageBenchmarks })).toBeVisible({ timeout: 10000 });
7069
+ await expect(page.getByRole('button', { name: EVALS_LABELS.actions.manageBenchmarks, exact: true })).toBeVisible({ timeout: 10000 });
7010
7070
  logger.info('Switched to Evals tab');
7011
7071
  }
7012
7072
  // ── Toolbar: benchmark picker + action buttons ─────────────────────────
@@ -7022,7 +7082,7 @@ function getBenchmarkComboboxDropdown(scope) {
7022
7082
  function getNewEvaluationButton(scope) {
7023
7083
  return scope.getByRole('button', { name: EVALS_LABELS.actions.newExperiment, exact: true });
7024
7084
  }
7025
- /** Locator for the toolbar "Manage benchmarks" button (opens the benchmarks dialog). */
7085
+ /** Locator for the toolbar "Manage Benchmarks" button (opens the benchmarks dialog). */
7026
7086
  function getManageBenchmarksButton(scope) {
7027
7087
  return scope.getByRole('button', { name: EVALS_LABELS.actions.manageBenchmarks, exact: true });
7028
7088
  }
@@ -7053,9 +7113,53 @@ async function selectBenchmark(scope, benchmarkName, opts = {}) {
7053
7113
  await expectSelectedBenchmark(scope, benchmarkName);
7054
7114
  logger.info(`Selected benchmark: ${benchmarkName}`);
7055
7115
  }
7056
- /** Assert the "no benchmarks yet" empty notice is shown in the tab body. */
7116
+ /** Locator for the empty state an organization with no benchmarks sees (sentence + "New Benchmark"). */
7117
+ function getNoBenchmarksEmptyState(scope) {
7118
+ return scope.getByTestId(EVALS_LABELS.noBenchmarksTestId);
7119
+ }
7120
+ /** Assert the no-benchmarks empty state says so and offers "New Benchmark". */
7057
7121
  async function expectNoBenchmarksNotice(scope) {
7058
- await expect(scope.getByText(EVALS_LABELS.emptyBenchmarks)).toBeVisible({ timeout: 10000 });
7122
+ const emptyState = getNoBenchmarksEmptyState(scope);
7123
+ await expect(emptyState).toContainText(EVALS_LABELS.emptyBenchmarks, { timeout: 10000 });
7124
+ await expect(emptyState.getByRole('button', { name: EVALS_LABELS.actions.newBenchmark, exact: true })).toBeVisible();
7125
+ }
7126
+ /**
7127
+ * Locator for what the tab shows when the selected benchmark has no Q&A: the
7128
+ * empty state when it has no evaluations either, or the line above the runs
7129
+ * table when it has. Both hold the same sentence and the same "Add Q&A" button.
7130
+ */
7131
+ function getNoQaPrompt(scope) {
7132
+ return scope
7133
+ .getByTestId(EVALS_LABELS.noQaEmptyTestId)
7134
+ .or(scope.getByTestId(EVALS_LABELS.noQaNoticeTestId));
7135
+ }
7136
+ /**
7137
+ * Assert the selected benchmark cannot be evaluated for lack of Q&A: the tab
7138
+ * says why and offers "Add Q&A", and "New Evaluation" is disabled with the
7139
+ * same sentence as its hover text.
7140
+ */
7141
+ async function expectBenchmarkNeedsQa(scope) {
7142
+ const prompt = getNoQaPrompt(scope);
7143
+ await expect(prompt).toContainText(EVALS_LABELS.emptyQa, { timeout: 15000 });
7144
+ await expect(prompt.getByRole('button', { name: EVALS_LABELS.actions.addQa, exact: true })).toBeVisible();
7145
+ const newEvaluation = getNewEvaluationButton(scope);
7146
+ await expect(newEvaluation).toBeDisabled();
7147
+ // A disabled button ignores the pointer, so the hover text hangs on the span wrapped around it.
7148
+ await newEvaluation.locator('xpath=..').hover();
7149
+ await expect(asPage$3(scope).getByRole('tooltip', { name: EVALS_LABELS.emptyQa, exact: true })).toBeVisible({ timeout: 5000 });
7150
+ }
7151
+ /**
7152
+ * Click the tab's own "Add Q&A" button (shown only while the selected
7153
+ * benchmark has no Q&A) and wait for the Add Q&A dialog. Returns the dialog;
7154
+ * continue with `addQaPairsManually` or `uploadQaCsv`.
7155
+ */
7156
+ async function openAddQaFromEvalsTab(scope) {
7157
+ await getNoQaPrompt(scope)
7158
+ .getByRole('button', { name: EVALS_LABELS.actions.addQa, exact: true })
7159
+ .click();
7160
+ const dialog = getAddItemsDialog(scope);
7161
+ await expect(dialog.getByRole('tab', { name: EVALS_LABELS.addItemsDialog.manualTab, exact: true })).toBeVisible({ timeout: 10000 });
7162
+ return dialog;
7059
7163
  }
7060
7164
  // ── Runs table ─────────────────────────────────────────────────────────
7061
7165
  /**
@@ -7074,11 +7178,9 @@ async function expectRunInTable(scope, runName) {
7074
7178
  async function expectRunNotInTable(scope, runName) {
7075
7179
  await expect(getRunRow(scope, runName)).toBeHidden({ timeout: 15000 });
7076
7180
  }
7077
- /** Assert the status badge on a run's row (badge text is upper-cased by the UI). */
7181
+ /** Assert the status on a run's row, read from the `data-status` of its status icon. */
7078
7182
  async function expectRunStatus(scope, runName, status) {
7079
- await expect(getRunRow(scope, runName).getByText(status, { exact: true })).toBeVisible({
7080
- timeout: 15000,
7081
- });
7183
+ await expect(getRunRow(scope, runName).locator(`[data-status="${status.toLowerCase()}"]`)).toBeVisible({ timeout: 15000 });
7082
7184
  }
7083
7185
  /** Assert the empty-state row shown when the selected benchmark has no runs for this agent. */
7084
7186
  async function expectRunsTableEmpty(scope, benchmarkName) {
@@ -7098,12 +7200,15 @@ async function openRunActionsMenu(scope, runName) {
7098
7200
  return menu;
7099
7201
  }
7100
7202
  /**
7101
- * Row action: "View results" — opens the evaluation detail dialog.
7203
+ * Row action: "View Results" — opens the evaluation detail dialog.
7102
7204
  * Only enabled once the run is COMPLETED. Returns the scoped detail dialog.
7103
7205
  */
7104
7206
  async function openRunResults(scope, runName) {
7105
7207
  const menu = await openRunActionsMenu(scope, runName);
7106
- const item = menu.getByRole('menuitem', { name: EVALS_LABELS.actions.viewResults });
7208
+ const item = menu.getByRole('menuitem', {
7209
+ name: EVALS_LABELS.actions.viewResults,
7210
+ exact: true,
7211
+ });
7107
7212
  await expect(item).toBeEnabled({ timeout: 5000 });
7108
7213
  await item.click();
7109
7214
  const dialog = getEvaluationDetailDialog(scope);
@@ -7112,12 +7217,12 @@ async function openRunResults(scope, runName) {
7112
7217
  return dialog;
7113
7218
  }
7114
7219
  /**
7115
- * Row action: "New review" — opens the LLM-judge dialog for the run.
7220
+ * Row action: "New Review" — opens the New Review (LLM-judge) dialog for the run.
7116
7221
  * Only enabled once the run is COMPLETED. Returns the scoped judge dialog.
7117
7222
  */
7118
7223
  async function openNewReviewForRun(scope, runName) {
7119
7224
  const menu = await openRunActionsMenu(scope, runName);
7120
- const item = menu.getByRole('menuitem', { name: EVALS_LABELS.actions.evaluate });
7225
+ const item = menu.getByRole('menuitem', { name: EVALS_LABELS.actions.evaluate, exact: true });
7121
7226
  await expect(item).toBeEnabled({ timeout: 5000 });
7122
7227
  await item.click();
7123
7228
  const dialog = getLlmJudgeDialog(scope);
@@ -7125,25 +7230,27 @@ async function openNewReviewForRun(scope, runName) {
7125
7230
  return dialog;
7126
7231
  }
7127
7232
  /**
7128
- * Row action: "Check status" — triggers a fresh status fetch and waits for
7233
+ * Row action: "Check Status" — triggers a fresh status fetch and waits for
7129
7234
  * the outcome toast. Returns which toast appeared. The menu stays open by
7130
7235
  * design (the item prevents default), so we close it with Escape.
7236
+ *
7237
+ * Safe to call again right away: it first waits for an earlier check's toast
7238
+ * to go, since that one can read the same as the next.
7131
7239
  */
7132
7240
  async function checkRunStatus(scope, runName) {
7133
7241
  const page = asPage$3(scope);
7134
- const menu = await openRunActionsMenu(scope, runName);
7135
- await menu.getByRole('menuitem', { name: EVALS_LABELS.actions.checkStatus }).click();
7136
7242
  const toasts = EVALS_LABELS.checkStatusToast;
7137
- const toast = page
7138
- .getByText(new RegExp(`${toasts.pending}|${toasts.completed}|${toasts.failed}`))
7139
- .first();
7243
+ const toast = getToast(scope, new RegExp(`^(${toasts.pending}|${toasts.completed}|${toasts.failed})$`));
7244
+ await expect(toast).toHaveCount(0, { timeout: 15000 });
7245
+ const menu = await openRunActionsMenu(scope, runName);
7246
+ await menu.getByRole('menuitem', { name: EVALS_LABELS.actions.checkStatus, exact: true }).click();
7140
7247
  await expect(toast).toBeVisible({ timeout: 15000 });
7141
7248
  const text = (await toast.innerText()).trim();
7142
7249
  await page.keyboard.press('Escape');
7143
7250
  await expect(menu).toBeHidden({ timeout: 5000 });
7144
- if (text.includes(toasts.completed))
7251
+ if (text === toasts.completed)
7145
7252
  return 'completed';
7146
- if (text.includes(toasts.failed))
7253
+ if (text === toasts.failed)
7147
7254
  return 'failed';
7148
7255
  return 'pending';
7149
7256
  }
@@ -7154,7 +7261,7 @@ async function checkRunStatus(scope, runName) {
7154
7261
  async function exportRunCsv(scope, runName) {
7155
7262
  const page = asPage$3(scope);
7156
7263
  const menu = await openRunActionsMenu(scope, runName);
7157
- const item = menu.getByRole('menuitem', { name: EVALS_LABELS.actions.exportCsv });
7264
+ const item = menu.getByRole('menuitem', { name: EVALS_LABELS.actions.exportCsv, exact: true });
7158
7265
  await expect(item).toBeEnabled({ timeout: 5000 });
7159
7266
  const downloadPromise = page.waitForEvent('download', { timeout: 30000 });
7160
7267
  await item.click();
@@ -7171,33 +7278,43 @@ function getStartEvaluationDialog(scope) {
7171
7278
  async function openStartEvaluationDialog(scope) {
7172
7279
  await getNewEvaluationButton(scope).click();
7173
7280
  const dialog = getStartEvaluationDialog(scope);
7174
- await expect(dialog.getByLabel(EVALS_LABELS.startDialog.experimentNameLabel)).toBeVisible({
7175
- timeout: 10000,
7176
- });
7281
+ await expect(dialog.getByLabel(EVALS_LABELS.startDialog.experimentNameLabel, { exact: true })).toBeVisible({ timeout: 10000 });
7177
7282
  return dialog;
7178
7283
  }
7179
7284
  /**
7180
7285
  * End-to-end: open the "New Evaluation" dialog, optionally name the run,
7181
- * submit, and wait for the dialog to close. When `name` is omitted the
7182
- * backend auto-generates one.
7286
+ * submit, and wait for the dialog to close and for the toast that names the
7287
+ * started evaluation. When `name` is omitted the backend auto-generates one.
7183
7288
  */
7184
7289
  async function startEvaluation(scope, opts = {}) {
7290
+ var _a;
7185
7291
  const dialog = await openStartEvaluationDialog(scope);
7186
7292
  if (opts.name !== undefined) {
7187
- await dialog.getByLabel(EVALS_LABELS.startDialog.experimentNameLabel).fill(opts.name);
7293
+ await dialog
7294
+ .getByLabel(EVALS_LABELS.startDialog.experimentNameLabel, { exact: true })
7295
+ .fill(opts.name);
7188
7296
  }
7189
7297
  // Scoped to the dialog: "Start Evaluation" is unique today, but the toolbar
7190
7298
  // "New Evaluation" trigger sits right behind the overlay — never query the page.
7191
- const submit = dialog.getByRole('button', { name: EVALS_LABELS.startDialog.start });
7299
+ const submit = dialog.getByRole('button', {
7300
+ name: EVALS_LABELS.startDialog.start,
7301
+ exact: true,
7302
+ });
7192
7303
  await expect(submit).toBeEnabled({ timeout: 5000 });
7193
7304
  await submit.click();
7194
7305
  await expect(dialog).toBeHidden({ timeout: 15000 });
7306
+ // Checked here, right after the dialog closes: the toast is gone within seconds.
7307
+ const started = EVALS_LABELS.startDialog.successToast;
7308
+ const typedName = (_a = opts.name) === null || _a === void 0 ? void 0 : _a.trim();
7309
+ await expect(getToast(scope, typedName
7310
+ ? started.replace('{name}', () => typedName)
7311
+ : new RegExp(`^${escapeRegExp(started.replace('{name}', ''))}`))).toBeVisible({ timeout: 10000 });
7195
7312
  logger.info(`Started evaluation${opts.name ? `: ${opts.name}` : ''}`);
7196
7313
  }
7197
7314
  /** Cancel out of the start dialog and wait for it to close. */
7198
7315
  async function cancelStartEvaluation(scope) {
7199
7316
  const dialog = getStartEvaluationDialog(scope);
7200
- await dialog.getByRole('button', { name: EVALS_LABELS.startDialog.cancel }).click();
7317
+ await dialog.getByRole('button', { name: EVALS_LABELS.startDialog.cancel, exact: true }).click();
7201
7318
  await expect(dialog).toBeHidden({ timeout: 5000 });
7202
7319
  }
7203
7320
  // ── Delete evaluation dialog ───────────────────────────────────────────
@@ -7206,33 +7323,41 @@ function getDeleteEvaluationDialog(scope) {
7206
7323
  return dialogByName(scope, EVALS_LABELS.deleteDialog.dialogName);
7207
7324
  }
7208
7325
  /**
7209
- * Delete a run from its row actions menu and confirm in the follow-up
7210
- * dialog. The confirm button ("Delete") shares its name with the menu item
7211
- * that opened the dialog, so the confirm click MUST stay dialog-scoped.
7326
+ * Click "Delete" in a run's row actions menu and wait for the confirm dialog
7327
+ * to ask about that run by name. Returns the dialog.
7212
7328
  */
7213
- async function deleteEvaluation(scope, runName) {
7329
+ async function openDeleteEvaluationDialog(scope, runName) {
7214
7330
  const menu = await openRunActionsMenu(scope, runName);
7215
- await menu.getByRole('menuitem', { name: EVALS_LABELS.actions.delete }).click();
7331
+ await menu.getByRole('menuitem', { name: EVALS_LABELS.actions.delete, exact: true }).click();
7216
7332
  const dialog = getDeleteEvaluationDialog(scope);
7217
- const confirm = dialog.getByRole('button', { name: EVALS_LABELS.deleteDialog.confirm });
7218
- await expect(confirm).toBeVisible({ timeout: 5000 });
7219
- await confirm.click();
7333
+ await expect(dialog).toContainText(EVALS_LABELS.deleteDialog.confirmation.replace('{runName}', () => runName), { timeout: 5000 });
7334
+ return dialog;
7335
+ }
7336
+ /**
7337
+ * Delete a run from its row actions menu and confirm in the follow-up
7338
+ * dialog, which names the run. The confirm button ("Delete") shares its name
7339
+ * with the menu item that opened the dialog, so the confirm click MUST stay
7340
+ * dialog-scoped.
7341
+ */
7342
+ async function deleteEvaluation(scope, runName) {
7343
+ const dialog = await openDeleteEvaluationDialog(scope, runName);
7344
+ await dialog
7345
+ .getByRole('button', { name: EVALS_LABELS.deleteDialog.confirm, exact: true })
7346
+ .click();
7220
7347
  await expect(dialog).toBeHidden({ timeout: 15000 });
7221
7348
  logger.info(`Deleted evaluation: ${runName}`);
7222
7349
  }
7223
7350
  /** Open the delete confirm from the row menu, then cancel out of it. */
7224
7351
  async function cancelDeleteEvaluation(scope, runName) {
7225
- const menu = await openRunActionsMenu(scope, runName);
7226
- await menu.getByRole('menuitem', { name: EVALS_LABELS.actions.delete }).click();
7227
- const dialog = getDeleteEvaluationDialog(scope);
7228
- await dialog.getByRole('button', { name: EVALS_LABELS.deleteDialog.cancel }).click();
7352
+ const dialog = await openDeleteEvaluationDialog(scope, runName);
7353
+ await dialog.getByRole('button', { name: EVALS_LABELS.deleteDialog.cancel, exact: true }).click();
7229
7354
  await expect(dialog).toBeHidden({ timeout: 5000 });
7230
7355
  }
7231
- // ── LLM-judge ("Evaluate") dialog ──────────────────────────────────────
7356
+ // ── New Review (LLM-judge) dialog ──────────────────────────────────────
7232
7357
  /**
7233
- * Locator for the LLM-judge dialog. Scoped by aria-label — its title AND its
7234
- * submit button both read "Evaluate", and it can be stacked on top of the
7235
- * evaluation-detail dialog.
7358
+ * Locator for the New Review (LLM-judge) dialog. Scoped by aria-label — its
7359
+ * title is also the name of the row action and of the detail-dialog button
7360
+ * that open it, and it can be stacked on top of the evaluation-detail dialog.
7236
7361
  */
7237
7362
  function getLlmJudgeDialog(scope) {
7238
7363
  return dialogByName(scope, EVALS_LABELS.judgeDialog.dialogName);
@@ -7274,15 +7399,18 @@ async function selectJudgeLlm(scope, providerName, modelName) {
7274
7399
  async function submitLlmJudge(scope, opts) {
7275
7400
  const dialog = getLlmJudgeDialog(scope);
7276
7401
  await expect(dialog).toBeVisible({ timeout: 10000 });
7277
- await dialog.getByLabel(EVALS_LABELS.judgeDialog.criteriaLabel).fill(opts.criteria);
7278
- await dialog.getByLabel(EVALS_LABELS.judgeDialog.scoreNameLabel).fill(opts.scoreName);
7402
+ await dialog
7403
+ .getByLabel(EVALS_LABELS.judgeDialog.criteriaLabel, { exact: true })
7404
+ .fill(opts.criteria);
7405
+ await dialog
7406
+ .getByLabel(EVALS_LABELS.judgeDialog.scoreNameLabel, { exact: true })
7407
+ .fill(opts.scoreName);
7279
7408
  if (opts.maxConcurrency !== undefined) {
7280
7409
  await dialog
7281
- .getByLabel(EVALS_LABELS.judgeDialog.maxConcurrencyLabel)
7410
+ .getByLabel(EVALS_LABELS.judgeDialog.maxConcurrencyLabel, { exact: true })
7282
7411
  .fill(String(opts.maxConcurrency));
7283
7412
  }
7284
7413
  await selectJudgeLlm(scope, opts.provider, opts.model);
7285
- // Dialog-scoped on purpose: the dialog title is also "Evaluate".
7286
7414
  const submit = dialog.getByRole('button', { name: EVALS_LABELS.judgeDialog.submit, exact: true });
7287
7415
  await expect(submit).toBeEnabled({ timeout: 5000 });
7288
7416
  await submit.click();
@@ -7292,7 +7420,7 @@ async function submitLlmJudge(scope, opts) {
7292
7420
  /** Cancel out of the judge dialog and wait for it to close. */
7293
7421
  async function cancelLlmJudge(scope) {
7294
7422
  const dialog = getLlmJudgeDialog(scope);
7295
- await dialog.getByRole('button', { name: EVALS_LABELS.judgeDialog.cancel }).click();
7423
+ await dialog.getByRole('button', { name: EVALS_LABELS.judgeDialog.cancel, exact: true }).click();
7296
7424
  await expect(dialog).toBeHidden({ timeout: 5000 });
7297
7425
  }
7298
7426
  // ── Evaluation detail dialog ───────────────────────────────────────────
@@ -7310,7 +7438,7 @@ async function refreshEvaluationDetail(scope) {
7310
7438
  await refresh.click();
7311
7439
  await expect(refresh).toBeEnabled({ timeout: 15000 });
7312
7440
  }
7313
- /** Open the LLM-judge dialog from the detail dialog's "New review" CTA. Returns the judge dialog. */
7441
+ /** Open the New Review dialog from the detail dialog's "New Review" button. Returns the judge dialog. */
7314
7442
  async function openNewReviewFromDetail(scope) {
7315
7443
  const dialog = getEvaluationDetailDialog(scope);
7316
7444
  const cta = dialog.getByRole('button', {
@@ -7323,7 +7451,7 @@ async function openNewReviewFromDetail(scope) {
7323
7451
  await expect(judgeDialog).toBeVisible({ timeout: 10000 });
7324
7452
  return judgeDialog;
7325
7453
  }
7326
- /** Assert the detail dialog is still waiting for traces (auto-refreshing hint). */
7454
+ /** Assert the detail dialog is still waiting for the first responses (it refreshes by itself). */
7327
7455
  async function expectDetailWaitingForTraces(scope) {
7328
7456
  await expect(getEvaluationDetailDialog(scope).getByText(EVALS_LABELS.detailDialog.waitingHint)).toBeVisible({ timeout: 10000 });
7329
7457
  }
@@ -7341,7 +7469,9 @@ function getTraceRow(scope, questionText) {
7341
7469
  async function expandTrace(scope, questionText) {
7342
7470
  await getTraceRow(scope, questionText).click();
7343
7471
  // The manual-score form only exists on the expanded trace.
7344
- await expect(getEvaluationDetailDialog(scope).getByLabel(EVALS_LABELS.detailDialog.scoreNameLabel)).toBeVisible({ timeout: 10000 });
7472
+ await expect(getEvaluationDetailDialog(scope).getByLabel(EVALS_LABELS.detailDialog.scoreNameLabel, {
7473
+ exact: true,
7474
+ })).toBeVisible({ timeout: 10000 });
7345
7475
  }
7346
7476
  /**
7347
7477
  * Add a manual score to the currently expanded trace. `value` must be
@@ -7349,10 +7479,16 @@ async function expandTrace(scope, questionText) {
7349
7479
  */
7350
7480
  async function addManualScore(scope, opts) {
7351
7481
  const dialog = getEvaluationDetailDialog(scope);
7352
- await dialog.getByLabel(EVALS_LABELS.detailDialog.scoreNameLabel).fill(opts.name);
7353
- await dialog.getByLabel(EVALS_LABELS.detailDialog.scoreValueLabel).fill(String(opts.value));
7482
+ await dialog
7483
+ .getByLabel(EVALS_LABELS.detailDialog.scoreNameLabel, { exact: true })
7484
+ .fill(opts.name);
7485
+ await dialog
7486
+ .getByLabel(EVALS_LABELS.detailDialog.scoreValueLabel, { exact: true })
7487
+ .fill(String(opts.value));
7354
7488
  if (opts.comment !== undefined) {
7355
- await dialog.getByLabel(EVALS_LABELS.detailDialog.scoreCommentLabel).fill(opts.comment);
7489
+ await dialog
7490
+ .getByLabel(EVALS_LABELS.detailDialog.scoreCommentLabel, { exact: true })
7491
+ .fill(opts.comment);
7356
7492
  }
7357
7493
  await dialog
7358
7494
  .getByRole('button', { name: EVALS_LABELS.detailDialog.addScore, exact: true })
@@ -7366,7 +7502,7 @@ async function expectTraceScore(scope, scoreName) {
7366
7502
  }
7367
7503
  /**
7368
7504
  * Remove a score chip from the expanded trace. The chip's trash button
7369
- * carries `aria-label="Remove score"`; we scope through the chip matched by
7505
+ * carries `aria-label="Delete Score"`; we scope through the chip matched by
7370
7506
  * score name so multiple chips can't cross-match.
7371
7507
  */
7372
7508
  async function removeTraceScore(scope, scoreName) {
@@ -7412,22 +7548,33 @@ async function expandReview(scope, scoreName) {
7412
7548
  await row.click();
7413
7549
  await expect(row).toHaveAttribute('aria-expanded', 'true', { timeout: 5000 });
7414
7550
  }
7415
- /** Close the detail dialog via Escape and wait for it to be gone. */
7551
+ /** Close the detail dialog with its Close button and wait for it to be gone. */
7416
7552
  async function closeEvaluationDetailDialog(scope) {
7417
- const dialog = getEvaluationDetailDialog(scope);
7418
- await asPage$3(scope).keyboard.press('Escape');
7553
+ await closeDialogByButton(getEvaluationDetailDialog(scope));
7554
+ }
7555
+ // ── Manage Benchmarks dialog (embedded tenant Benchmarks) ──────────────
7556
+ /**
7557
+ * Close a dialog with its own Close (X) button. Not Escape: closing a child
7558
+ * dialog hands focus back to the icon button that opened it, which opens that
7559
+ * button's tooltip, and the next Escape then closes only the tooltip. The
7560
+ * detail dialog can also drop an Escape while it refreshes itself.
7561
+ */
7562
+ async function closeDialogByButton(dialog) {
7563
+ await dialog.getByRole('button', { name: EVALS_LABELS.dialogClose, exact: true }).click();
7419
7564
  await expect(dialog).toBeHidden({ timeout: 10000 });
7420
7565
  }
7421
- // ── Manage benchmarks dialog (embedded tenant Benchmarks) ──────────────
7422
- /** Locator for the "Manage benchmarks" dialog. Scoped by aria-label — its title collides with the toolbar button that opens it. */
7566
+ /** Locator for the "Manage Benchmarks" dialog. Scoped by aria-label — its title collides with the toolbar button that opens it. */
7423
7567
  function getManageBenchmarksDialog(scope) {
7424
7568
  return dialogByName(scope, EVALS_LABELS.manageBenchmarksDialog.dialogName);
7425
7569
  }
7426
- /** Open the Manage benchmarks dialog from the toolbar and wait for its table UI. Returns the scoped dialog. */
7570
+ /** Open the Manage Benchmarks dialog from the toolbar and wait for its toolbar. Returns the scoped dialog. */
7427
7571
  async function openManageBenchmarksDialog(scope) {
7428
7572
  await getManageBenchmarksButton(scope).click();
7429
7573
  const dialog = getManageBenchmarksDialog(scope);
7430
- await expect(dialog.getByRole('button', { name: EVALS_LABELS.manageBenchmarksDialog.newBenchmark })).toBeVisible({ timeout: 10000 });
7574
+ await expect(dialog.getByRole('button', {
7575
+ name: EVALS_LABELS.manageBenchmarksDialog.newBenchmark,
7576
+ exact: true,
7577
+ })).toBeVisible({ timeout: 10000 });
7431
7578
  return dialog;
7432
7579
  }
7433
7580
  /** Type into the benchmarks search input (aria-label "Search benchmarks") inside the manage dialog. */
@@ -7440,65 +7587,187 @@ async function searchBenchmarks(scope, query) {
7440
7587
  async function expectBenchmarkListed(scope, benchmarkName) {
7441
7588
  await expect(getManageBenchmarksDialog(scope).getByText(benchmarkName, { exact: true }).first()).toBeVisible({ timeout: 15000 });
7442
7589
  }
7590
+ /**
7591
+ * Show the Manual or CSV Upload tab of the Q&A form inside `dialog` and
7592
+ * return its panel. The panel is looked up by its tab's name, because the New
7593
+ * Benchmark dialog's Q&A step is a tabpanel around this one.
7594
+ */
7595
+ async function showQaFormTab(dialog, subTab) {
7596
+ const tabName = subTab === 'manual'
7597
+ ? EVALS_LABELS.addItemsDialog.manualTab
7598
+ : EVALS_LABELS.addItemsDialog.csvTab;
7599
+ const tab = dialog.getByRole('tab', { name: tabName, exact: true });
7600
+ await tab.click();
7601
+ await expect(tab).toHaveAttribute('data-state', 'active', { timeout: 5000 });
7602
+ const panel = dialog.getByRole('tabpanel', { name: tabName, exact: true });
7603
+ await expect(panel).toBeVisible({ timeout: 5000 });
7604
+ return panel;
7605
+ }
7606
+ /** Type one Q&A pair per row on the Manual tab of the Q&A form inside `dialog`, adding rows as needed. */
7607
+ async function fillQaRows(dialog, pairs) {
7608
+ const panel = await showQaFormTab(dialog, 'manual');
7609
+ for (let i = 0; i < pairs.length; i++) {
7610
+ if (i > 0) {
7611
+ await panel
7612
+ .getByRole('button', { name: EVALS_LABELS.addItemsDialog.addRow, exact: true })
7613
+ .click();
7614
+ }
7615
+ const question = panel
7616
+ .getByLabel(EVALS_LABELS.addItemsDialog.questionLabel, { exact: true })
7617
+ .nth(i);
7618
+ await expect(question).toBeVisible({ timeout: 5000 });
7619
+ await question.fill(pairs[i].question);
7620
+ const answer = pairs[i].answer;
7621
+ if (answer !== undefined) {
7622
+ await panel
7623
+ .getByLabel(EVALS_LABELS.addItemsDialog.answerLabel, { exact: true })
7624
+ .nth(i)
7625
+ .fill(answer);
7626
+ }
7627
+ }
7628
+ }
7629
+ /** Choose a CSV file on the CSV Upload tab of the Q&A form inside `dialog`. */
7630
+ async function chooseQaCsv(dialog, csvPath) {
7631
+ const panel = await showQaFormTab(dialog, 'csv');
7632
+ await panel
7633
+ .getByLabel(EVALS_LABELS.addItemsDialog.csvFileLabel, { exact: true })
7634
+ .setInputFiles(csvPath);
7635
+ }
7636
+ // ── New Benchmark dialog ───────────────────────────────────────────────
7443
7637
  /** Locator for the "New Benchmark" dialog. Scoped by aria-label — its title collides with the button that opens it. */
7444
7638
  function getCreateBenchmarkDialog(scope) {
7445
7639
  return dialogByName(scope, EVALS_LABELS.createBenchmarkDialog.dialogName);
7446
7640
  }
7447
7641
  /**
7448
- * Create a benchmark from inside the manage dialog: open "New Benchmark",
7449
- * fill the form, submit, and wait for both the dialog to close and the new
7450
- * row to appear.
7642
+ * Click "New Benchmark" in the open Manage Benchmarks dialog and wait for the
7643
+ * New Benchmark dialog on its Details step. Returns the dialog.
7451
7644
  */
7452
- async function createBenchmark(scope, opts) {
7453
- const manageDialog = getManageBenchmarksDialog(scope);
7454
- await manageDialog
7455
- .getByRole('button', { name: EVALS_LABELS.manageBenchmarksDialog.newBenchmark })
7645
+ async function openCreateBenchmarkDialog(scope) {
7646
+ await getManageBenchmarksDialog(scope)
7647
+ .getByRole('button', { name: EVALS_LABELS.manageBenchmarksDialog.newBenchmark, exact: true })
7456
7648
  .click();
7457
7649
  const dialog = getCreateBenchmarkDialog(scope);
7458
- await dialog.getByLabel(EVALS_LABELS.createBenchmarkDialog.nameLabel).fill(opts.name);
7650
+ await expect(dialog.getByLabel(EVALS_LABELS.createBenchmarkDialog.nameLabel, { exact: true })).toBeVisible({ timeout: 10000 });
7651
+ return dialog;
7652
+ }
7653
+ /**
7654
+ * Click "Next" on the Details step of the open New Benchmark dialog and wait
7655
+ * for the Q&A step: its tab active and its note showing. "Next" is disabled
7656
+ * until Name has text. Returns the dialog.
7657
+ */
7658
+ async function goToBenchmarkQaStep(scope) {
7659
+ const labels = EVALS_LABELS.createBenchmarkDialog;
7660
+ const dialog = getCreateBenchmarkDialog(scope);
7661
+ await dialog.getByRole('button', { name: labels.next, exact: true }).click();
7662
+ await expect(dialog.getByRole('tab', { name: labels.qaTab, exact: true })).toHaveAttribute('data-state', 'active', { timeout: 5000 });
7663
+ // The Q&A step stays mounted while Details shows, so the note is checked for being on screen.
7664
+ const hint = dialog.getByTestId(labels.qaHintTestId);
7665
+ await expect(hint).toBeVisible({ timeout: 5000 });
7666
+ await expect(hint).toHaveText(labels.qaHint);
7667
+ return dialog;
7668
+ }
7669
+ // A CSV's count is not known here, and a large one is written with separators ("1,000").
7670
+ const CREATED_FROM_CSV_TOAST = /^Benchmark created with \S+ Q&A pairs?$/;
7671
+ /**
7672
+ * Fill both steps of the open New Benchmark dialog, submit, and wait for it to
7673
+ * close and for the toast that fits what was created.
7674
+ */
7675
+ async function submitCreateBenchmark(scope, opts) {
7676
+ var _a;
7677
+ const labels = EVALS_LABELS.createBenchmarkDialog;
7678
+ const dialog = getCreateBenchmarkDialog(scope);
7679
+ await dialog.getByLabel(labels.nameLabel, { exact: true }).fill(opts.name);
7459
7680
  if (opts.description !== undefined) {
7460
- await dialog
7461
- .getByLabel(EVALS_LABELS.createBenchmarkDialog.descriptionLabel)
7462
- .fill(opts.description);
7681
+ await dialog.getByLabel(labels.descriptionLabel, { exact: true }).fill(opts.description);
7682
+ }
7683
+ await goToBenchmarkQaStep(scope);
7684
+ const qaPairs = (_a = opts.qaPairs) !== null && _a !== void 0 ? _a : [];
7685
+ let createdToast = labels.successToastNoQa;
7686
+ if (opts.csvPath !== undefined) {
7687
+ await chooseQaCsv(dialog, opts.csvPath);
7688
+ createdToast = CREATED_FROM_CSV_TOAST;
7689
+ }
7690
+ else if (qaPairs.length > 0) {
7691
+ await fillQaRows(dialog, qaPairs);
7692
+ // The form leaves out a row whose question is blank.
7693
+ const saved = qaPairs.filter((pair) => pair.question.trim()).length;
7694
+ if (saved > 0) {
7695
+ createdToast = `Benchmark created with ${saved} Q&A pair${saved === 1 ? '' : 's'}`;
7696
+ }
7463
7697
  }
7464
- const submit = dialog.getByRole('button', {
7465
- name: EVALS_LABELS.createBenchmarkDialog.create,
7466
- exact: true,
7467
- });
7698
+ const submit = dialog.getByRole('button', { name: labels.create, exact: true });
7468
7699
  await expect(submit).toBeEnabled({ timeout: 5000 });
7469
7700
  await submit.click();
7470
- await expect(dialog).toBeHidden({ timeout: 15000 });
7701
+ // One click creates the benchmark and then saves its Q&A; the dialog closes after both.
7702
+ await expect(dialog).toBeHidden({ timeout: 30000 });
7703
+ // Checked here, right after the dialog closes: the toast is gone within seconds.
7704
+ await expect(getToast(scope, createdToast)).toBeVisible({ timeout: 10000 });
7705
+ }
7706
+ /**
7707
+ * Create a benchmark from inside the manage dialog: open "New Benchmark", fill
7708
+ * Name and Description, click "Next", fill the Q&A step when `qaPairs` or
7709
+ * `csvPath` is given, click "Create", and wait for the dialog to close, the
7710
+ * toast, and the new row. The Evals tab selects the new benchmark.
7711
+ *
7712
+ * Q&A is optional: with neither `qaPairs` nor `csvPath` the benchmark is
7713
+ * created without any, and "New Evaluation" stays disabled for it until Q&A
7714
+ * are added (see `expectBenchmarkNeedsQa`).
7715
+ */
7716
+ async function createBenchmark(scope, opts) {
7717
+ await openCreateBenchmarkDialog(scope);
7718
+ await submitCreateBenchmark(scope, opts);
7471
7719
  await expectBenchmarkListed(scope, opts.name);
7472
7720
  logger.info(`Created benchmark: ${opts.name}`);
7473
7721
  }
7474
- /** Locator for a benchmark's items dialog (Q&A list). Its title is the benchmark name (dynamic), hence the aria-label hook. */
7722
+ /**
7723
+ * Create an organization's first benchmark from the Evals tab's no-benchmarks
7724
+ * empty state: click its "New Benchmark" button, fill and submit the dialog as
7725
+ * `createBenchmark` does (Q&A optional), and wait for the tab to select the
7726
+ * new benchmark.
7727
+ */
7728
+ async function createBenchmarkFromEmptyState(scope, opts) {
7729
+ await getNoBenchmarksEmptyState(scope)
7730
+ .getByRole('button', { name: EVALS_LABELS.actions.newBenchmark, exact: true })
7731
+ .click();
7732
+ await submitCreateBenchmark(scope, opts);
7733
+ await expectSelectedBenchmark(scope, opts.name);
7734
+ logger.info(`Created benchmark: ${opts.name}`);
7735
+ }
7736
+ // ── Benchmark Q&A dialogs ──────────────────────────────────────────────
7737
+ /** Locator for a benchmark's Q&A dialog. Its title is the benchmark name (dynamic), hence the aria-label hook. */
7475
7738
  function getBenchmarkItemsDialog(scope) {
7476
7739
  return dialogByName(scope, EVALS_LABELS.benchmarkItemsDialog.dialogName);
7477
7740
  }
7478
7741
  /**
7479
- * Open a benchmark's Q&A items dialog from its row's eye button
7480
- * (`aria-label="View items in <name>"`). Returns the scoped items dialog.
7742
+ * Open a benchmark's Q&A dialog from its row's eye button
7743
+ * (`aria-label="View Q&A in <name>"`). Returns the scoped Q&A dialog.
7481
7744
  */
7482
7745
  async function openBenchmarkItems(scope, benchmarkName) {
7483
7746
  await getManageBenchmarksDialog(scope)
7484
- .getByRole('button', { name: `View items in ${benchmarkName}`, exact: true })
7747
+ .getByRole('button', { name: `View Q&A in ${benchmarkName}`, exact: true })
7485
7748
  .click();
7486
7749
  const dialog = getBenchmarkItemsDialog(scope);
7487
- await expect(dialog.getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.addItems })).toBeVisible({ timeout: 10000 });
7750
+ await expect(dialog.getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.addItems, exact: true })).toBeVisible({ timeout: 10000 });
7488
7751
  return dialog;
7489
7752
  }
7490
- /** Assert a Q&A item (matched by its question text) is listed in the items dialog. */
7753
+ /** Assert a Q&A pair (matched by its question text) is listed in the Q&A dialog. */
7491
7754
  async function expectQaItemListed(scope, questionText) {
7492
7755
  await expect(getBenchmarkItemsDialog(scope).getByText(questionText, { exact: false }).first()).toBeVisible({ timeout: 15000 });
7493
7756
  }
7757
+ /** Assert the Q&A dialog says the benchmark has no Q&A and offers "Add Q&A" as the next step. */
7758
+ async function expectBenchmarkItemsEmpty(scope) {
7759
+ const dialog = getBenchmarkItemsDialog(scope);
7760
+ await expect(dialog.getByText(EVALS_LABELS.benchmarkItemsDialog.empty, { exact: true })).toBeVisible({ timeout: 15000 });
7761
+ await expect(dialog.getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.addItems, exact: true })).toBeVisible();
7762
+ }
7494
7763
  /** Locator for the "Add Q&A" dialog (Manual / CSV Upload sub-tabs). */
7495
7764
  function getAddItemsDialog(scope) {
7496
7765
  return dialogByName(scope, EVALS_LABELS.addItemsDialog.dialogName);
7497
7766
  }
7498
- /** Open the "Add Q&A" dialog from the items dialog. Returns the scoped add dialog. */
7767
+ /** Open the "Add Q&A" dialog from a benchmark's Q&A dialog. Returns the scoped add dialog. */
7499
7768
  async function openAddItemsDialog(scope) {
7500
7769
  await getBenchmarkItemsDialog(scope)
7501
- .getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.addItems })
7770
+ .getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.addItems, exact: true })
7502
7771
  .click();
7503
7772
  const dialog = getAddItemsDialog(scope);
7504
7773
  // Both sub-tab triggers are unique to this dialog — a reliable ready signal.
@@ -7508,61 +7777,35 @@ async function openAddItemsDialog(scope) {
7508
7777
  /**
7509
7778
  * Switch to a sub-tab INSIDE the Add Q&A dialog. The tab query must stay
7510
7779
  * dialog-scoped: the edit-mentor modal behind it has its own `role="tab"`
7511
- * strip, and an unscoped `getByRole('tab')` can cross-match. Returns the
7512
- * active tabpanel so callers interact only with that sub-tab's controls
7513
- * (each panel has its own footer with its own Cancel/submit buttons).
7780
+ * strip, and an unscoped `getByRole('tab')` can cross-match. Returns that
7781
+ * sub-tab's tabpanel, which holds its fields; Cancel and the submit
7782
+ * button ("Add" or "Upload") are in the dialog's one footer, outside it.
7514
7783
  */
7515
7784
  async function switchToAddItemsSubTab(scope, subTab) {
7516
- const dialog = getAddItemsDialog(scope);
7517
- const tabName = subTab === 'manual'
7518
- ? EVALS_LABELS.addItemsDialog.manualTab
7519
- : EVALS_LABELS.addItemsDialog.csvTab;
7520
- const tab = dialog.getByRole('tab', { name: tabName, exact: true });
7521
- await tab.click();
7522
- await expect(tab).toHaveAttribute('data-state', 'active', { timeout: 5000 });
7523
- const panel = dialog.getByRole('tabpanel');
7524
- await expect(panel).toBeVisible({ timeout: 5000 });
7525
- return panel;
7785
+ return showQaFormTab(getAddItemsDialog(scope), subTab);
7526
7786
  }
7527
7787
  /**
7528
- * Add Q&A pairs via the Manual sub-tab. Fills one row per pair (clicking
7529
- * "Add row" as needed), submits, and waits for the dialog to close. Row
7530
- * inputs are matched by label WITHIN each row container so the Question /
7531
- * Answer labels of sibling rows can't collide.
7788
+ * Add Q&A pairs via the Manual sub-tab of the open Add Q&A dialog. Fills one
7789
+ * row per pair (clicking "Add Row" as needed), submits, and waits for the
7790
+ * dialog to close.
7532
7791
  */
7533
7792
  async function addQaPairsManually(scope, pairs) {
7534
7793
  const dialog = getAddItemsDialog(scope);
7535
- const panel = await switchToAddItemsSubTab(scope, 'manual');
7536
- for (let i = 0; i < pairs.length; i++) {
7537
- if (i > 0) {
7538
- await panel
7539
- .getByRole('button', { name: EVALS_LABELS.addItemsDialog.addRow, exact: true })
7540
- .click();
7541
- }
7542
- const question = panel.getByLabel(EVALS_LABELS.addItemsDialog.questionLabel).nth(i);
7543
- await expect(question).toBeVisible({ timeout: 5000 });
7544
- await question.fill(pairs[i].question);
7545
- if (pairs[i].answer !== undefined) {
7546
- await panel
7547
- .getByLabel(EVALS_LABELS.addItemsDialog.answerLabel)
7548
- .nth(i)
7549
- .fill(pairs[i].answer);
7550
- }
7551
- }
7552
- await panel.getByRole('button', { name: EVALS_LABELS.addItemsDialog.add, exact: true }).click();
7794
+ await fillQaRows(dialog, pairs);
7795
+ await dialog.getByRole('button', { name: EVALS_LABELS.addItemsDialog.add, exact: true }).click();
7553
7796
  await expect(dialog).toBeHidden({ timeout: 15000 });
7554
7797
  logger.info(`Added ${pairs.length} Q&A pair(s) manually`);
7555
7798
  }
7556
7799
  /**
7557
- * Upload Q&A pairs via the CSV Upload sub-tab. `csvPath` is a path on the
7558
- * test machine; the file needs `input` (required) and `expected_output`
7559
- * (optional) columns. Waits for the dialog to close on success.
7800
+ * Upload Q&A pairs via the CSV Upload sub-tab of the open Add Q&A dialog.
7801
+ * `csvPath` is a path on the test machine; the file needs `input` (required)
7802
+ * and `expected_output` (optional) columns. Waits for the dialog to close on
7803
+ * success.
7560
7804
  */
7561
7805
  async function uploadQaCsv(scope, csvPath) {
7562
7806
  const dialog = getAddItemsDialog(scope);
7563
- const panel = await switchToAddItemsSubTab(scope, 'csv');
7564
- await panel.getByLabel(EVALS_LABELS.addItemsDialog.csvFileLabel).setInputFiles(csvPath);
7565
- await panel
7807
+ await chooseQaCsv(dialog, csvPath);
7808
+ await dialog
7566
7809
  .getByRole('button', { name: EVALS_LABELS.addItemsDialog.upload, exact: true })
7567
7810
  .click();
7568
7811
  await expect(dialog).toBeHidden({ timeout: 30000 });
@@ -7573,16 +7816,19 @@ function getDeleteQaItemDialog(scope) {
7573
7816
  return dialogByName(scope, EVALS_LABELS.deleteItemDialog.dialogName);
7574
7817
  }
7575
7818
  /**
7576
- * Delete a Q&A item by its question text: click the row's trash button
7819
+ * Delete a Q&A pair by its question text: click the row's bin button
7577
7820
  * (`aria-label="Delete Q&A"`) and confirm. The confirm "Delete" button must
7578
- * stay scoped to the confirm dialog — the items dialog's rows keep their
7579
- * own delete buttons mounted behind it.
7821
+ * stay scoped to the confirm dialog — the Q&A dialog's rows keep their own
7822
+ * delete buttons mounted behind it.
7580
7823
  */
7581
7824
  async function deleteQaItem(scope, questionText) {
7582
7825
  const itemsDialog = getBenchmarkItemsDialog(scope);
7583
7826
  const row = itemsDialog.getByRole('row').filter({ hasText: questionText }).first();
7584
7827
  await row
7585
- .getByRole('button', { name: EVALS_LABELS.benchmarkItemsDialog.deleteItemAriaLabel })
7828
+ .getByRole('button', {
7829
+ name: EVALS_LABELS.benchmarkItemsDialog.deleteItemAriaLabel,
7830
+ exact: true,
7831
+ })
7586
7832
  .click();
7587
7833
  const confirmDialog = getDeleteQaItemDialog(scope);
7588
7834
  const confirm = confirmDialog.getByRole('button', {
@@ -7594,17 +7840,13 @@ async function deleteQaItem(scope, questionText) {
7594
7840
  await expect(confirmDialog).toBeHidden({ timeout: 15000 });
7595
7841
  logger.info(`Deleted Q&A item: ${questionText}`);
7596
7842
  }
7597
- /** Close the benchmark items dialog via Escape and wait for it to be gone. */
7843
+ /** Close a benchmark's Q&A dialog with its Close button and wait for it to be gone. */
7598
7844
  async function closeBenchmarkItemsDialog(scope) {
7599
- const dialog = getBenchmarkItemsDialog(scope);
7600
- await asPage$3(scope).keyboard.press('Escape');
7601
- await expect(dialog).toBeHidden({ timeout: 10000 });
7845
+ await closeDialogByButton(getBenchmarkItemsDialog(scope));
7602
7846
  }
7603
- /** Close the Manage benchmarks dialog via Escape and wait for it to be gone. */
7847
+ /** Close the Manage Benchmarks dialog with its Close button and wait for it to be gone. */
7604
7848
  async function closeManageBenchmarksDialog(scope) {
7605
- const dialog = getManageBenchmarksDialog(scope);
7606
- await asPage$3(scope).keyboard.press('Escape');
7607
- await expect(dialog).toBeHidden({ timeout: 10000 });
7849
+ await closeDialogByButton(getManageBenchmarksDialog(scope));
7608
7850
  }
7609
7851
 
7610
7852
  /**
@@ -9136,5 +9378,5 @@ function createPlaywrightConfig(options) {
9136
9378
  });
9137
9379
  }
9138
9380
 
9139
- export { AuthFlowBuilder, CHAT_PRIVACY_LABELS, CustomReporter, EVALS_LABELS, GRADER_LABELS, HISTORY_TAB_LABELS, LTI_LABELS, LTI_TEST_IDS, MEMORY_ADMIN_LABELS, MailsacClient, PRIVACY_LABELS, SCREENSHARE_LABELS, SPEND_LIMITS_LABELS, SUPPORT_LABELS, TASKS_LABELS, VOICE_LABELS, addAgentMemoryFromPopup, addCreditsDialog, addGraderCriterion, addManualScore, addMemory, addQaPairsManually, addSkillToAgent, addTenantUserSpendLimit, addTextResource, addUserGlobalMemory, addUserSpendLimit, addUserSpendLimitFromTenantBilling, agentLimitsRow, agentLimitsSection, applyVoiceInstructionsPreset, archiveFirstMemory, archiveMemoryByContent, billingAutoRechargeSection, billingCreditsSection, billingPlanSection, buildReportUrl, canChatWithEmbedMentor, cancelDeleteEvaluation, cancelDeleteInstance, cancelDeleteSkill, cancelDisconnectInstance, cancelEnableChatPrivacyMidSession, cancelKeyDelete, cancelLlmJudge, cancelNewInstanceDialog, cancelStartEvaluation, checkAdminStatus, checkRunStatus, choosePlan, clearAgentLimitsFilter, clearAgentMemoriesFilter, clearDateRangeFilter, clearGradeResultOverride, clearHistoryAgentFilter, clearInstanceSearch, clearUserFilter, clearUserLimitsFilter, clickBackHome, clickBillingAddCredits, clickBillingManageBilling, clickBillingManageUsage, clickBillingUpgrade, clickBillingUpgradeToStandard, clickChatPrivacyToggle, clickDownloadAgain, clickManualDownloadLink, clickSetSpendLimitForFilteredAgent, clickSetSpendLimitForFilteredUser, closeAgentLimitsPopup, closeAgentMemoriesPopup, closeBenchmarkItemsDialog, closeCreditBalanceDropdown, closeEvaluationDetailDialog, closeKeyDetail, closeManageBenchmarksDialog, closePlanPicker, closeUserMemoriesPopup, closeWithEsc, confirmEnableChatPrivacyMidSession, confirmKeyDelete, connectToInstance, copyEndpoint, createAgentSpendLimitViaFilter, createAuthSetup, createBenchmark, createEnvConfig, createInstance, createKey, createLink, createPlaywrightConfig, createSkill, createSupportTicketViaChatAndVerify, createTool, creditBalancePanel, creditBalancePlanBadge, creditBalanceTrigger, deleteAgentSpendLimit, deleteAgentSpendLimitFromTenantBilling, deleteEvaluation, deleteFirstMemory, deleteGraderCriterion, deleteInstance, deleteKey, deleteMemoryByContent, deleteQaItem, deleteResource, deleteSkill, deleteTask, deleteTenantUserSpendLimit, deleteUserGlobalMemory, deleteUserSpendLimit, deleteWorkspaceSpendLimit, disableSkill, disableSupport, disconnectInstance, downloadConversationCsv, downloadExportedReport, editAgentPrompt, editGraderCriterion, editInstance, editLink, editSkill, editTenantUserSpendLimit, editTextResource, editTool, editUserGlobalMemory, editUserSpendLimit, enableSkill, enableSupport, expandReview, expandTrace, expectAgentLimitsRowContent, expectAllEndpointsVisible, expectBenchmarkListed, expectBillingAutoRechargeSection, expectBillingCreditsSection, expectBillingPlanSection, expectBillingTabForCurrentPlan, expectBillingTabForFreePlan, expectBillingTabForPremiumPlan, expectBillingTabForStandardPlan, expectBillingTabForTrialPlan, expectCallConfigVisible, expectCallConfigVoiceTriggerShows, expectChatPrivacyConfirmDialogOpen, expectChatPrivacyLocked, expectChatPrivacySource, expectChatPrivacyState, expectChatPrivacyToggleVisible, expectCompletedTasks, expectCreditBalanceForCurrentPlan, expectCreditBalancePanelForFreePlan, expectCreditBalancePanelForPremiumPlan, expectCreditBalancePanelForStandardPlan, expectCreditBalancePanelForTrialPlan, expectCreditBalanceVisibilityForTenant, expectDetailWaitingForTraces, expectEndpointUrl, expectEntitySelected, expectFailedTasks, expectGradeResultRow, expectGraderMisconfiguredWarning, expectGraderTotalPoints, expectKeyDeleteError, expectKeyInList, expectKeyNotInList, expectKeysEmpty, expectLastCriterionDeleteDisabled, expectLinkInList, expectLinkNotInList, expectLinkStatus, expectLinkTargetUri, expectLinksEmpty, expectLogDetailsStatus, expectLogsForTask, expectLtiHeader, expectMentorVoiceTriggerShows, expectMessageInConversation, expectNewUserSpendLimitDefault, expectNoAccessibilityViolations, expectNoAccessibilityViolationsOnDialogs, expectNoBenchmarksNotice, expectNoLogsForSelectedTask, expectNoRepliesYet, expectNoTickets, expectOutputFilterEnabled, expectPlanPicker, expectPrivacyFieldsHidden, expectPrivacyFieldsVisible, expectPrivateModeSelected, expectPrivateModeTabReady, expectQaItemListed, expectReviewInList, expectReviewStatus, expectRunInTable, expectRunNotInTable, expectRunStatus, expectRunsTableEmpty, expectScheduleStartTimeInPastError, expectScreenShareDisabledHint, expectSelectedBenchmark, expectSttSelectDisabled, expectTaskInList, expectTaskNotInList, expectTaskStatus, expectTasksEmpty, expectTenantChatPrivacyEnabled, expectTenantChatPrivacyVisible, expectTicketClosedNotice, expectTicketDescriptionContains, expectTicketInList, expectTicketStatusInList, expectToolInList, expectToolNotInList, expectToolsEmpty, expectTotalTasks, expectTraceScore, expectTtsSelectDisabled, expectUserLimitsRowContent, expectVoiceInstructionsValue, expectVoiceProviderSelected, expectVoiceVisible, expectWorkspaceActualSpendStats, expectWorkspaceSpendStats, exportHistoryAndWaitForDownload, exportRunCsv, fillLinkName, fillToolForm, filterAgentLimits, filterAgentMemories, filterByAction, filterByActionAndVerify, filterByActor, filterByActorAndVerify, filterByDateRange, filterGradeResultsByEmail, filterHistoryByAgent, filterHistoryBySentiment, filterHistoryByTopic, filterTicketsByStatus, filterTicketsByUser, filterUserLimits, generateBrowserSetupProjects, generateProjectConfig, getAddItemsDialog, getAuditLogRowCount, getAvailableActors, getBenchmarkCombobox, getBenchmarkComboboxDropdown, getBenchmarkItemsDialog, getBillingAutoRechargeStatus, getBillingPlanLabel, getBrowserKey, getCallConfigForm, getChatInput, getChatPrivacyConfirmDialog, getChatPrivacyToggle, getConversationList, getConversationPreview, getConversationRows, getCreateBenchmarkDialog, getCreateKeyButton, getCreateLinkButton, getCreateToolButton, getCreditBalancePlanLabel, getCreditBalanceRemaining, getCurrentModel, getCurrentTenantShowPaywall, getDeleteEvaluationDialog, getDeleteQaItemDialog, getEditAgentDialog, getEndpoint, getEndpointCopyButton, getEndpointUrl, getEntityChip, getEvaluationDetailDialog, getExportRowsByState, getExportsTable, getInstanceHealthLabel, getInstanceRowCount, getInstanceStatusLabel, getKeyActionsTrigger, getKeyCreateModal, getKeyCreateNameInput, getKeyDeleteModal, getKeyDetailModal, getKeyDetailNameInput, getKeyPublicJwkField, getKeyPublicKeyField, getKeyRow, getKeysEmptyState, getKeysSection, getLinkCopyTargetUriButton, getLinkEditButton, getLinkModal, getLinkNameInput, getLinkRow, getLinksEmptyState, getLinksRefreshButton, getLinksSection, getLlmJudgeDialog, getLlmPickerDialog, getLtiSubTab, getManageBenchmarksButton, getManageBenchmarksDialog, getMemoryCount, getMentorIdFromUrl, getNewEvaluationButton, getOutputFilterSwitch, getPaginationInfo, getPlanPickerKeys, getPrivateModeCard, getProfileDialog, getReplyComposer, getReviewRow, getRunRow, getScheduleTaskButton, getSearchInput, getSkillRowCount, getStartEvaluationDialog, getStatusFilter, getSupportInfoBox, getSupportTabTrigger, getSupportToggle, getTaskRow, getTenantChatPrivacyRow, getTenantChatPrivacySwitch, getTicketDescription, getTicketDetail, getTicketList, getTicketRow, getTicketRowByIndex, getToolEditButton, getToolKeySelect, getToolModal, getToolRow, getToolsEmptyState, getToolsSection, getTraceRow, getUserFilter, getVoiceProviderCard, getVoiceRow, goToFirstPage, goToLastPage, goToNextPage, goToPage, goToPreviousPage, goToSkillsListPage, graderTabBody, inviteUserTest, isEvaluationTabVisible, isFirefox, isGraderTabVisible, isGradingEnabled, isJSON, isLtiTabVisible, isMemoryTabVisible, isOnFirstPage, isOnLastPage, isPrivacyTabVisible, isPrivateModeTabVisible, isSandboxKindEnabled, isSandboxTabVisible, isScreenShareTabVisible, isSkillEnabled, isSpendLimitsTabVisible, isSupportEnabled, isSupportTabVisible, isTasksTabVisible, isTenantMemoryTabVisible, isVoiceTabVisible, logger, loginWithEmailAndPassword, loginWithMicrosoftIdp, memoryAdminAgentRow, memoryAdminAgentSection, memoryAdminGlobalSection, memoryAdminUserRow, memoryRowByContent, navigateToAccountComponent, navigateToAuditLog, navigateToAuditLogAndWaitForData, navigateToDataReports, navigateToReportDownload, newUserDefaultSection, openAddItemsDialog, openAddMemoryDialog, openAgentLimitsManage, openAgentMemoriesPopup, openAgentPromptEditModal, openBenchmarkItems, openCallConfigVoicePicker, openCreateKeyModal, openCreateLinkModal, openCreateToolModal, openCreditBalanceDropdown, openEditInstanceDialog, openEditLinkModal, openEditSkillDialog, openEditToolModal, openFirstLogDetails, openHistoryTab, openInstanceActionsMenu, openKeyActionsMenu, openKeyDelete, openKeyDetail, openLLMProviderPicker, openManageBenchmarksDialog, openMentorVoicePicker, openNewInstanceDialog, openNewReviewForRun, openNewReviewFromDetail, openNewSkillDialog, openPlanPickerFromBilling, openPlanPickerFromCreditBalance, openRunActionsMenu, openRunResults, openScheduleTaskDialog, openScreenSharePromptEditor, openSkillActionsMenu, openSlashSkillPicker, openStartEvaluationDialog, openTicket, openUserMemoriesPopup, openVoiceInstructionsEditor, overrideGradeResult, parseReportUrlParams, planCard, planPickerDialog, previewCallConfigVoiceInline, previewMentorVoiceInline, previewVoice, pushConfiguration, readEndpointUrl, readKeyPublicJwk, readKeyPublicKey, refreshEvaluationDetail, refreshLinks, refreshTickets, reliableClick, reliableFill, removeSkillFromAgent, removeTraceScore, renameKey, replyToTicket, requestSupportTicketViaChat, resetCallConfig, retry, retryFailedLink, runConnectedInstanceChecks, runInstanceChecks, safeWaitForURL, saveCallConfig, saveGraderConfig, saveScreenSharePrompts, saveVoiceSettings, scheduleTask, searchBenchmarks, searchInstances, searchMemoryAdminUsers, searchTasks, searchVoices, selectBenchmark, selectCallMode, selectConversation, selectDateFromCalendar, selectJudgeLlm, selectLLMModel, selectLlmProvider, selectPrivacyAction, selectPrivateMode, selectSlashSkill, selectSttProvider, selectTaskInList, selectToolKeySetMode, selectToolSigningKey, selectTtsProvider, selectVoice, selectVoiceProvider, sendAgentChatMessage, setAgentLimitsRowEnabled, setAgentSpendLimit, setAgentSpendLimitFromTenantBilling, setBlockMessage, setCallLanguage, setCatalogSkillEnabled, setEnableVideo, setEntitySelected, setGradingEnabled, setNewUserSpendLimitDefault, setOutputFilterEnabled, setScreenSharePrompt, setSupportEnabled, setTenantChatPrivacyEnabled, setTicketStatus, setUseFunctionCallingEnabled, setUserLimitsRowEnabled, setUserSpendLimitEnabled, setVoiceInstructions, setWorkspaceSpendLimit, setupSandboxInstance, shouldAddNewRowWhenClickingAddRowButton, shouldAllowEditingCellValuesInCSVEditor, shouldCancelCombiningReports, shouldCloseCSVEditorWhenClickingCloseButton, shouldCloseCSVEditorWithoutSavingWhenClickingCancel, shouldCombineRecommendationReports, shouldDirectlyDownloadChatHistoryReportWithoutCSVEditor, shouldDisableOtherDownloadButtonsWhileGeneratingReport, shouldDisplayCSVInEditableTableFormat, shouldDisplayReportCards, shouldHaveCombinedReportDataTestIds, shouldOpenCSVEditorDialog, shouldOpenCSVEditorForUserMetadataReport, shouldSaveEditedCSVAndTriggerDownload, shouldShowCombiningReportsDialog, shouldVerifyCSVEditorDialogAccessibility, signUpWithEmailAndPassword, spendLimitsTabBody, startEvaluation, startHistoryExport, submitAddCreditsAmount, submitLinkModal, submitLlmJudge, submitToolModal, switchHistorySubTab, switchToAddItemsSubTab, switchToAgentLimits, switchToAgentSkillsSubTab, switchToAvailableSkillsSubTab, switchToEvaluationTab, switchToGraderSubTab, switchToGraderTab, switchToLtiSubTab, switchToLtiTab, switchToMemoryAdminSubTab, switchToMemoryTab, switchToPlanAndCredits, switchToPrivacyTab, switchToPrivateModeTab, switchToSandboxTab, switchToScreenShareTab, switchToSkillResourcesSubTab, switchToSkillsTab, switchToSpendLimitsSubTab, switchToSpendLimitsTab, switchToSupportTab, switchToTasksTab, switchToTenantMemoryTab, switchToUserLimits, switchToVoiceSubTab, switchToVoiceTab, switchToWorkspaceSpendLimits, teardownSandboxInstance, test, toggleAutoPush, toggleMemorySwitch, toggleSandboxKind, toggleSkill, toggleUserMemoryAdminSetting, toolFields, turnOffChatPrivacyMidSession, uploadAssetResource, uploadQaCsv, userLimitsRow, userLimitsSection, userSpendLimitRow, verifyAgentConfigPromptsVisible, verifyAgentSkillsEmptyState, verifyAuditLogEmptyState, verifyAuditLogEntryStructure, verifyAuditLogGenericError, verifyAuditLogLoading, verifyAuditLogPermissionError, verifyAuditLogTableVisible, verifyClawSectionHidden, verifyClawSectionVisible, verifyConnectDisabledForUnhealthy, verifyConnectedInstanceCard, verifyCurrentPage, verifyDonePhase, verifyDownloadingPhase, verifyErrorPhase, verifyInstanceTableEmpty, verifyInstanceTableVisible, verifyMemoryExists, verifyMemoryNotExists, verifyMemoryTabMemoriesList, verifyMemoryTabSettings, verifyPreparingPhase, verifySandboxKindsVisible, verifySkillAdded, verifySkillVisible, verifySkillsEmptyState, verifySkillsTabVisible, waitForAgentChatResponse, waitForAuditLogDataLoaded, waitForBillingTabReady, waitForCompletedExportRow, waitForConversations, waitForCreditBalanceLoaded, waitForDialogReady, waitForElementStable, waitForLinkReady, waitForPageLoad, waitForPageReady, waitForPlanPicker, waitForReportDownload, waitForTicketInList, workspaceSpendLimitSection };
9381
+ export { AuthFlowBuilder, CHAT_PRIVACY_LABELS, CustomReporter, EVALS_LABELS, GRADER_LABELS, HISTORY_TAB_LABELS, LTI_LABELS, LTI_TEST_IDS, MEMORY_ADMIN_LABELS, MailsacClient, PRIVACY_LABELS, SCREENSHARE_LABELS, SPEND_LIMITS_LABELS, SUPPORT_LABELS, TASKS_LABELS, VOICE_LABELS, addAgentMemoryFromPopup, addCreditsDialog, addGraderCriterion, addManualScore, addMemory, addQaPairsManually, addSkillToAgent, addTenantUserSpendLimit, addTextResource, addUserGlobalMemory, addUserSpendLimit, addUserSpendLimitFromTenantBilling, agentLimitsRow, agentLimitsSection, applyVoiceInstructionsPreset, archiveFirstMemory, archiveMemoryByContent, billingAutoRechargeSection, billingCreditsSection, billingPlanSection, buildReportUrl, canChatWithEmbedMentor, cancelDeleteEvaluation, cancelDeleteInstance, cancelDeleteSkill, cancelDisconnectInstance, cancelEnableChatPrivacyMidSession, cancelKeyDelete, cancelLlmJudge, cancelNewInstanceDialog, cancelStartEvaluation, checkAdminStatus, checkRunStatus, choosePlan, clearAgentLimitsFilter, clearAgentMemoriesFilter, clearDateRangeFilter, clearGradeResultOverride, clearHistoryAgentFilter, clearInstanceSearch, clearUserFilter, clearUserLimitsFilter, clickBackHome, clickBillingAddCredits, clickBillingManageBilling, clickBillingManageUsage, clickBillingUpgrade, clickBillingUpgradeToStandard, clickChatPrivacyToggle, clickDownloadAgain, clickManualDownloadLink, clickSetSpendLimitForFilteredAgent, clickSetSpendLimitForFilteredUser, closeAgentLimitsPopup, closeAgentMemoriesPopup, closeBenchmarkItemsDialog, closeCreditBalanceDropdown, closeEvaluationDetailDialog, closeKeyDetail, closeManageBenchmarksDialog, closePlanPicker, closeUserMemoriesPopup, closeWithEsc, confirmEnableChatPrivacyMidSession, confirmKeyDelete, connectToInstance, copyEndpoint, createAgentSpendLimitViaFilter, createAuthSetup, createBenchmark, createBenchmarkFromEmptyState, createEnvConfig, createInstance, createKey, createLink, createPlaywrightConfig, createSkill, createSupportTicketViaChatAndVerify, createTool, creditBalancePanel, creditBalancePlanBadge, creditBalanceTrigger, deleteAgentSpendLimit, deleteAgentSpendLimitFromTenantBilling, deleteEvaluation, deleteFirstMemory, deleteGraderCriterion, deleteInstance, deleteKey, deleteMemoryByContent, deleteQaItem, deleteResource, deleteSkill, deleteTask, deleteTenantUserSpendLimit, deleteUserGlobalMemory, deleteUserSpendLimit, deleteWorkspaceSpendLimit, disableSkill, disableSupport, disconnectInstance, downloadConversationCsv, downloadExportedReport, editAgentPrompt, editGraderCriterion, editInstance, editLink, editSkill, editTenantUserSpendLimit, editTextResource, editTool, editUserGlobalMemory, editUserSpendLimit, enableSkill, enableSupport, expandReview, expandTrace, expectAgentLimitsRowContent, expectAllEndpointsVisible, expectBenchmarkItemsEmpty, expectBenchmarkListed, expectBenchmarkNeedsQa, expectBillingAutoRechargeSection, expectBillingCreditsSection, expectBillingPlanSection, expectBillingTabForCurrentPlan, expectBillingTabForFreePlan, expectBillingTabForPremiumPlan, expectBillingTabForStandardPlan, expectBillingTabForTrialPlan, expectCallConfigVisible, expectCallConfigVoiceTriggerShows, expectChatPrivacyConfirmDialogOpen, expectChatPrivacyLocked, expectChatPrivacySource, expectChatPrivacyState, expectChatPrivacyToggleVisible, expectCompletedTasks, expectCreditBalanceForCurrentPlan, expectCreditBalancePanelForFreePlan, expectCreditBalancePanelForPremiumPlan, expectCreditBalancePanelForStandardPlan, expectCreditBalancePanelForTrialPlan, expectCreditBalanceVisibilityForTenant, expectDetailWaitingForTraces, expectEndpointUrl, expectEntitySelected, expectFailedTasks, expectGradeResultRow, expectGraderMisconfiguredWarning, expectGraderTotalPoints, expectKeyDeleteError, expectKeyInList, expectKeyNotInList, expectKeysEmpty, expectLastCriterionDeleteDisabled, expectLinkInList, expectLinkNotInList, expectLinkStatus, expectLinkTargetUri, expectLinksEmpty, expectLogDetailsStatus, expectLogsForTask, expectLtiHeader, expectMentorVoiceTriggerShows, expectMessageInConversation, expectNewUserSpendLimitDefault, expectNoAccessibilityViolations, expectNoAccessibilityViolationsOnDialogs, expectNoBenchmarksNotice, expectNoLogsForSelectedTask, expectNoRepliesYet, expectNoTickets, expectOutputFilterEnabled, expectPlanPicker, expectPrivacyFieldsHidden, expectPrivacyFieldsVisible, expectPrivateModeSelected, expectPrivateModeTabReady, expectQaItemListed, expectReviewInList, expectReviewStatus, expectRunInTable, expectRunNotInTable, expectRunStatus, expectRunsTableEmpty, expectScheduleStartTimeInPastError, expectScreenShareDisabledHint, expectSelectedBenchmark, expectSttSelectDisabled, expectTaskInList, expectTaskNotInList, expectTaskStatus, expectTasksEmpty, expectTenantChatPrivacyEnabled, expectTenantChatPrivacyVisible, expectTicketClosedNotice, expectTicketDescriptionContains, expectTicketInList, expectTicketStatusInList, expectToolInList, expectToolNotInList, expectToolsEmpty, expectTotalTasks, expectTraceScore, expectTtsSelectDisabled, expectUserLimitsRowContent, expectVoiceInstructionsValue, expectVoiceProviderSelected, expectVoiceVisible, expectWorkspaceActualSpendStats, expectWorkspaceSpendStats, exportHistoryAndWaitForDownload, exportRunCsv, fillLinkName, fillToolForm, filterAgentLimits, filterAgentMemories, filterByAction, filterByActionAndVerify, filterByActor, filterByActorAndVerify, filterByDateRange, filterGradeResultsByEmail, filterHistoryByAgent, filterHistoryBySentiment, filterHistoryByTopic, filterTicketsByStatus, filterTicketsByUser, filterUserLimits, generateBrowserSetupProjects, generateProjectConfig, getAddItemsDialog, getAuditLogRowCount, getAvailableActors, getBenchmarkCombobox, getBenchmarkComboboxDropdown, getBenchmarkItemsDialog, getBillingAutoRechargeStatus, getBillingPlanLabel, getBrowserKey, getCallConfigForm, getChatInput, getChatPrivacyConfirmDialog, getChatPrivacyToggle, getConversationList, getConversationPreview, getConversationRows, getCreateBenchmarkDialog, getCreateKeyButton, getCreateLinkButton, getCreateToolButton, getCreditBalancePlanLabel, getCreditBalanceRemaining, getCurrentModel, getCurrentTenantShowPaywall, getDeleteEvaluationDialog, getDeleteQaItemDialog, getEditAgentDialog, getEndpoint, getEndpointCopyButton, getEndpointUrl, getEntityChip, getEvaluationDetailDialog, getExportRowsByState, getExportsTable, getInstanceHealthLabel, getInstanceRowCount, getInstanceStatusLabel, getKeyActionsTrigger, getKeyCreateModal, getKeyCreateNameInput, getKeyDeleteModal, getKeyDetailModal, getKeyDetailNameInput, getKeyPublicJwkField, getKeyPublicKeyField, getKeyRow, getKeysEmptyState, getKeysSection, getLinkCopyTargetUriButton, getLinkEditButton, getLinkModal, getLinkNameInput, getLinkRow, getLinksEmptyState, getLinksRefreshButton, getLinksSection, getLlmJudgeDialog, getLlmPickerDialog, getLtiSubTab, getManageBenchmarksButton, getManageBenchmarksDialog, getMemoryCount, getMentorIdFromUrl, getNewEvaluationButton, getNoBenchmarksEmptyState, getNoQaPrompt, getOutputFilterSwitch, getPaginationInfo, getPlanPickerKeys, getPrivateModeCard, getProfileDialog, getReplyComposer, getReviewRow, getRunRow, getScheduleTaskButton, getSearchInput, getSkillRowCount, getStartEvaluationDialog, getStatusFilter, getSupportInfoBox, getSupportTabTrigger, getSupportToggle, getTaskRow, getTenantChatPrivacyRow, getTenantChatPrivacySwitch, getTicketDescription, getTicketDetail, getTicketList, getTicketRow, getTicketRowByIndex, getToolEditButton, getToolKeySelect, getToolModal, getToolRow, getToolsEmptyState, getToolsSection, getTraceRow, getUserFilter, getVoiceProviderCard, getVoiceRow, goToBenchmarkQaStep, goToFirstPage, goToLastPage, goToNextPage, goToPage, goToPreviousPage, goToSkillsListPage, graderTabBody, inviteUserTest, isEvaluationTabVisible, isFirefox, isGraderTabVisible, isGradingEnabled, isJSON, isLtiTabVisible, isMemoryTabVisible, isOnFirstPage, isOnLastPage, isPrivacyTabVisible, isPrivateModeTabVisible, isSandboxKindEnabled, isSandboxTabVisible, isScreenShareTabVisible, isSkillEnabled, isSpendLimitsTabVisible, isSupportEnabled, isSupportTabVisible, isTasksTabVisible, isTenantMemoryTabVisible, isVoiceTabVisible, logger, loginWithEmailAndPassword, loginWithMicrosoftIdp, memoryAdminAgentRow, memoryAdminAgentSection, memoryAdminGlobalSection, memoryAdminUserRow, memoryRowByContent, navigateToAccountComponent, navigateToAuditLog, navigateToAuditLogAndWaitForData, navigateToDataReports, navigateToReportDownload, newUserDefaultSection, openAddItemsDialog, openAddMemoryDialog, openAddQaFromEvalsTab, openAgentLimitsManage, openAgentMemoriesPopup, openAgentPromptEditModal, openBenchmarkItems, openCallConfigVoicePicker, openCreateBenchmarkDialog, openCreateKeyModal, openCreateLinkModal, openCreateToolModal, openCreditBalanceDropdown, openEditInstanceDialog, openEditLinkModal, openEditSkillDialog, openEditToolModal, openFirstLogDetails, openHistoryTab, openInstanceActionsMenu, openKeyActionsMenu, openKeyDelete, openKeyDetail, openLLMProviderPicker, openManageBenchmarksDialog, openMentorVoicePicker, openNewInstanceDialog, openNewReviewForRun, openNewReviewFromDetail, openNewSkillDialog, openPlanPickerFromBilling, openPlanPickerFromCreditBalance, openRunActionsMenu, openRunResults, openScheduleTaskDialog, openScreenSharePromptEditor, openSkillActionsMenu, openSlashSkillPicker, openStartEvaluationDialog, openTicket, openUserMemoriesPopup, openVoiceInstructionsEditor, overrideGradeResult, parseReportUrlParams, planCard, planPickerDialog, previewCallConfigVoiceInline, previewMentorVoiceInline, previewVoice, pushConfiguration, readEndpointUrl, readKeyPublicJwk, readKeyPublicKey, refreshEvaluationDetail, refreshLinks, refreshTickets, reliableClick, reliableFill, removeSkillFromAgent, removeTraceScore, renameKey, replyToTicket, requestSupportTicketViaChat, resetCallConfig, retry, retryFailedLink, runConnectedInstanceChecks, runInstanceChecks, safeWaitForURL, saveCallConfig, saveGraderConfig, saveScreenSharePrompts, saveVoiceSettings, scheduleTask, searchBenchmarks, searchInstances, searchMemoryAdminUsers, searchTasks, searchVoices, selectBenchmark, selectCallMode, selectConversation, selectDateFromCalendar, selectJudgeLlm, selectLLMModel, selectLlmProvider, selectPrivacyAction, selectPrivateMode, selectSlashSkill, selectSttProvider, selectTaskInList, selectToolKeySetMode, selectToolSigningKey, selectTtsProvider, selectVoice, selectVoiceProvider, sendAgentChatMessage, setAgentLimitsRowEnabled, setAgentSpendLimit, setAgentSpendLimitFromTenantBilling, setBlockMessage, setCallLanguage, setCatalogSkillEnabled, setEnableVideo, setEntitySelected, setGradingEnabled, setNewUserSpendLimitDefault, setOutputFilterEnabled, setScreenSharePrompt, setSupportEnabled, setTenantChatPrivacyEnabled, setTicketStatus, setUseFunctionCallingEnabled, setUserLimitsRowEnabled, setUserSpendLimitEnabled, setVoiceInstructions, setWorkspaceSpendLimit, setupSandboxInstance, shouldAddNewRowWhenClickingAddRowButton, shouldAllowEditingCellValuesInCSVEditor, shouldCancelCombiningReports, shouldCloseCSVEditorWhenClickingCloseButton, shouldCloseCSVEditorWithoutSavingWhenClickingCancel, shouldCombineRecommendationReports, shouldDirectlyDownloadChatHistoryReportWithoutCSVEditor, shouldDisableOtherDownloadButtonsWhileGeneratingReport, shouldDisplayCSVInEditableTableFormat, shouldDisplayReportCards, shouldHaveCombinedReportDataTestIds, shouldOpenCSVEditorDialog, shouldOpenCSVEditorForUserMetadataReport, shouldSaveEditedCSVAndTriggerDownload, shouldShowCombiningReportsDialog, shouldVerifyCSVEditorDialogAccessibility, signUpWithEmailAndPassword, spendLimitsTabBody, startEvaluation, startHistoryExport, submitAddCreditsAmount, submitLinkModal, submitLlmJudge, submitToolModal, switchHistorySubTab, switchToAddItemsSubTab, switchToAgentLimits, switchToAgentSkillsSubTab, switchToAvailableSkillsSubTab, switchToEvaluationTab, switchToGraderSubTab, switchToGraderTab, switchToLtiSubTab, switchToLtiTab, switchToMemoryAdminSubTab, switchToMemoryTab, switchToPlanAndCredits, switchToPrivacyTab, switchToPrivateModeTab, switchToSandboxTab, switchToScreenShareTab, switchToSkillResourcesSubTab, switchToSkillsTab, switchToSpendLimitsSubTab, switchToSpendLimitsTab, switchToSupportTab, switchToTasksTab, switchToTenantMemoryTab, switchToUserLimits, switchToVoiceSubTab, switchToVoiceTab, switchToWorkspaceSpendLimits, teardownSandboxInstance, test, toggleAutoPush, toggleMemorySwitch, toggleSandboxKind, toggleSkill, toggleUserMemoryAdminSetting, toolFields, turnOffChatPrivacyMidSession, uploadAssetResource, uploadQaCsv, userLimitsRow, userLimitsSection, userSpendLimitRow, verifyAgentConfigPromptsVisible, verifyAgentSkillsEmptyState, verifyAuditLogEmptyState, verifyAuditLogEntryStructure, verifyAuditLogGenericError, verifyAuditLogLoading, verifyAuditLogPermissionError, verifyAuditLogTableVisible, verifyClawSectionHidden, verifyClawSectionVisible, verifyConnectDisabledForUnhealthy, verifyConnectedInstanceCard, verifyCurrentPage, verifyDonePhase, verifyDownloadingPhase, verifyErrorPhase, verifyInstanceTableEmpty, verifyInstanceTableVisible, verifyMemoryExists, verifyMemoryNotExists, verifyMemoryTabMemoriesList, verifyMemoryTabSettings, verifyPreparingPhase, verifySandboxKindsVisible, verifySkillAdded, verifySkillVisible, verifySkillsEmptyState, verifySkillsTabVisible, waitForAgentChatResponse, waitForAuditLogDataLoaded, waitForBillingTabReady, waitForCompletedExportRow, waitForConversations, waitForCreditBalanceLoaded, waitForDialogReady, waitForElementStable, waitForLinkReady, waitForPageLoad, waitForPageReady, waitForPlanPicker, waitForReportDownload, waitForTicketInList, workspaceSpendLimitSection };
9140
9382
  //# sourceMappingURL=index.esm.js.map