minitest-cli 0.47.0__tar.gz → 0.48.0__tar.gz

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 (310) hide show
  1. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/PKG-INFO +2 -2
  2. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/README.md +1 -1
  3. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/pyproject.toml +1 -1
  4. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/SKILL.md +10 -2
  5. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/reference/suite-schemas.md +3 -2
  6. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/reference/test-suite-design.md +67 -66
  7. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/batch.py +4 -2
  8. minitest_cli-0.48.0/src/minitest_cli/commands/batch_delete.py +77 -0
  9. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_batch_commands.py +79 -1
  10. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/uv.lock +1 -1
  11. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.env.example +0 -0
  12. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.github/workflows/ci.yml +0 -0
  13. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.github/workflows/install-scripts.yml +0 -0
  14. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.github/workflows/release.yml +0 -0
  15. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.gitignore +0 -0
  16. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/.opencode/skill/release/SKILL.md +0 -0
  17. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/AGENTS.md +0 -0
  18. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/RELEASE.md +0 -0
  19. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/install.ps1 +0 -0
  20. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/install.sh +0 -0
  21. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/pyrightconfig.json +0 -0
  22. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/renovate.json +0 -0
  23. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/scripts/check-dist-skill.py +0 -0
  24. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/scripts/sync-skill-references.py +0 -0
  25. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/README.md +0 -0
  26. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/commands/app-skills.md +0 -0
  27. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/commands/draft-features.md +0 -0
  28. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/commands/drafts.md +0 -0
  29. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/commands/knowledge.md +0 -0
  30. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/commands/schedules.md +0 -0
  31. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/skills/minitest-cli/reference/index.json +0 -0
  32. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/__init__.py +0 -0
  33. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/api/__init__.py +0 -0
  34. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/api/apps_manager_client.py +0 -0
  35. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/api/client.py +0 -0
  36. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/api/errors.py +0 -0
  37. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/assets/__init__.py +0 -0
  38. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/assets/callback.html +0 -0
  39. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/__init__.py +0 -0
  40. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/_abi_mismatch.py +0 -0
  41. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/_response_errors.py +0 -0
  42. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_knowledge.py +0 -0
  43. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_knowledge_helpers.py +0 -0
  44. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_network_helpers.py +0 -0
  45. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_skill.py +0 -0
  46. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_skill_helpers.py +0 -0
  47. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_skill_secret.py +0 -0
  48. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/app_skill_write.py +0 -0
  49. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/apps.py +0 -0
  50. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/apps_dependencies.py +0 -0
  51. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/apps_get.py +0 -0
  52. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/apps_helpers.py +0 -0
  53. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/apps_table.py +0 -0
  54. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/auth.py +0 -0
  55. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/auth_api_key.py +0 -0
  56. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/backend_api.py +0 -0
  57. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/backend_api_helpers.py +0 -0
  58. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/batch_helpers.py +0 -0
  59. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/batch_warnings.py +0 -0
  60. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build.py +0 -0
  61. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_bundle.py +0 -0
  62. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_commit.py +0 -0
  63. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_conversion.py +0 -0
  64. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_fix_prompts.py +0 -0
  65. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_get.py +0 -0
  66. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_helpers.py +0 -0
  67. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/build_signing.py +0 -0
  68. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/capabilities.py +0 -0
  69. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/commit_helpers.py +0 -0
  70. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/deprecated_alias.py +0 -0
  71. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft.py +0 -0
  72. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_asks.py +0 -0
  73. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_conversation.py +0 -0
  74. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_create.py +0 -0
  75. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_feature.py +0 -0
  76. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_feature_apply.py +0 -0
  77. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_feature_conflicts.py +0 -0
  78. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_feature_helpers.py +0 -0
  79. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_feature_show.py +0 -0
  80. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_helpers.py +0 -0
  81. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_legacy.py +0 -0
  82. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_lifecycle.py +0 -0
  83. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_pr.py +0 -0
  84. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_render.py +0 -0
  85. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_update.py +0 -0
  86. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/draft_watch.py +0 -0
  87. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/env.py +0 -0
  88. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/env_helpers.py +0 -0
  89. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/feedback_scope.py +0 -0
  90. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/flow_types.py +0 -0
  91. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/flow_types_helpers.py +0 -0
  92. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/flow_types_write.py +0 -0
  93. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/glossary.py +0 -0
  94. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/init.py +0 -0
  95. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/init_helpers.py +0 -0
  96. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/init_playbook.py +0 -0
  97. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues.py +0 -0
  98. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_fix.py +0 -0
  99. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_get.py +0 -0
  100. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_helpers.py +0 -0
  101. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_not_a_bug.py +0 -0
  102. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_projection.py +0 -0
  103. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_service.py +0 -0
  104. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_severity.py +0 -0
  105. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/issues_ticket.py +0 -0
  106. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/knowledge.py +0 -0
  107. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/knowledge_documents.py +0 -0
  108. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/knowledge_helpers.py +0 -0
  109. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/private_network.py +0 -0
  110. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/private_network_helpers.py +0 -0
  111. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/recording_timeline.py +0 -0
  112. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/reference.py +0 -0
  113. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/residential_proxy.py +0 -0
  114. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run.py +0 -0
  115. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_asks.py +0 -0
  116. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_asks_helpers.py +0 -0
  117. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_backend_override.py +0 -0
  118. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_cancel.py +0 -0
  119. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_commit.py +0 -0
  120. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_commit_helpers.py +0 -0
  121. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_display.py +0 -0
  122. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_feedback.py +0 -0
  123. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_helpers.py +0 -0
  124. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_list.py +0 -0
  125. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_logs.py +0 -0
  126. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_recording.py +0 -0
  127. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_recording_helpers.py +0 -0
  128. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_tagged.py +0 -0
  129. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/run_targets.py +0 -0
  130. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules.py +0 -0
  131. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_body.py +0 -0
  132. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_create.py +0 -0
  133. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_display.py +0 -0
  134. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_helpers.py +0 -0
  135. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_history.py +0 -0
  136. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_options.py +0 -0
  137. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_scope.py +0 -0
  138. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_spec.py +0 -0
  139. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_update.py +0 -0
  140. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/schedules_where.py +0 -0
  141. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/screens.py +0 -0
  142. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/screens_detail.py +0 -0
  143. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/screens_format.py +0 -0
  144. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/screens_helpers.py +0 -0
  145. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/screens_tree.py +0 -0
  146. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret.py +0 -0
  147. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_create.py +0 -0
  148. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_extract.py +0 -0
  149. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_helpers.py +0 -0
  150. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_lint.py +0 -0
  151. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_manage.py +0 -0
  152. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_state.py +0 -0
  153. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/secret_value.py +0 -0
  154. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/skill.py +0 -0
  155. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/tags.py +0 -0
  156. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/tags_helpers.py +0 -0
  157. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_file.py +0 -0
  158. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_file_helpers.py +0 -0
  159. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_file_list.py +0 -0
  160. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile.py +0 -0
  161. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_card.py +0 -0
  162. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_default.py +0 -0
  163. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_helpers.py +0 -0
  164. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_list.py +0 -0
  165. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_secret_refs.py +0 -0
  166. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/test_profile_update.py +0 -0
  167. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/upgrade.py +0 -0
  168. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/upgrade_auto.py +0 -0
  169. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story.py +0 -0
  170. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_app_skills.py +0 -0
  171. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_bindings.py +0 -0
  172. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_camera.py +0 -0
  173. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_create.py +0 -0
  174. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_criteria.py +0 -0
  175. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_criterion_edits.py +0 -0
  176. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_delete.py +0 -0
  177. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_device_count.py +0 -0
  178. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_helpers.py +0 -0
  179. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_iap.py +0 -0
  180. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_modify.py +0 -0
  181. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_overrides.py +0 -0
  182. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_profiles.py +0 -0
  183. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_setup_commit.py +0 -0
  184. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/user_story_update.py +0 -0
  185. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/verdicts.py +0 -0
  186. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/verdicts_helpers.py +0 -0
  187. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/commands/verdicts_projection.py +0 -0
  188. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/__init__.py +0 -0
  189. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/app_context.py +0 -0
  190. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/auth.py +0 -0
  191. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/config.py +0 -0
  192. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/credentials.py +0 -0
  193. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/oauth.py +0 -0
  194. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/tenants.py +0 -0
  195. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/core/token_exchange.py +0 -0
  196. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/main.py +0 -0
  197. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/__init__.py +0 -0
  198. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/app.py +0 -0
  199. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/app_env_vars.py +0 -0
  200. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/app_failure.py +0 -0
  201. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/app_screen.py +0 -0
  202. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/base.py +0 -0
  203. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/batch.py +0 -0
  204. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/build.py +0 -0
  205. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/commit_build.py +0 -0
  206. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/draft_feature.py +0 -0
  207. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/flow_type.py +0 -0
  208. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/issues.py +0 -0
  209. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/lane_logs.py +0 -0
  210. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/recording.py +0 -0
  211. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/story_run.py +0 -0
  212. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/tag.py +0 -0
  213. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/targets.py +0 -0
  214. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/models/user_story.py +0 -0
  215. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/__init__.py +0 -0
  216. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/auto_update.py +0 -0
  217. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/auto_update_notices.py +0 -0
  218. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/auto_update_state.py +0 -0
  219. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/auto_update_worker.py +0 -0
  220. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/confirm.py +0 -0
  221. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/dependency_sim.py +0 -0
  222. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/mermaid.py +0 -0
  223. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/output.py +0 -0
  224. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/reference_docs.py +0 -0
  225. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/skill_bundle.py +0 -0
  226. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/skill_refresh.py +0 -0
  227. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/update_check.py +0 -0
  228. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/versions.py +0 -0
  229. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/src/minitest_cli/utils/video_frames.py +0 -0
  230. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/__init__.py +0 -0
  231. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/_backend_api_server.py +0 -0
  232. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/_commit_transport.py +0 -0
  233. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/_network_settings_server.py +0 -0
  234. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/_secret_router.py +0 -0
  235. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/conftest.py +0 -0
  236. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/draft_backend.py +0 -0
  237. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_api_client_auth.py +0 -0
  238. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_app_context.py +0 -0
  239. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_app_knowledge_commands.py +0 -0
  240. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_app_skill_commands.py +0 -0
  241. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_apps_commands.py +0 -0
  242. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_apps_dependencies.py +0 -0
  243. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_auth.py +0 -0
  244. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_auth_api_key.py +0 -0
  245. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_auth_commands.py +0 -0
  246. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_auto_update.py +0 -0
  247. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_auto_update_worker.py +0 -0
  248. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_backend_api_commands.py +0 -0
  249. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_backend_api_openapi_auth.py +0 -0
  250. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_build_bundle.py +0 -0
  251. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_build_commands.py +0 -0
  252. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_build_fix_prompts.py +0 -0
  253. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_build_get_command.py +0 -0
  254. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_capabilities_command.py +0 -0
  255. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_channel_header.py +0 -0
  256. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_code_quality.py +0 -0
  257. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_commit_commands.py +0 -0
  258. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_config_defaults.py +0 -0
  259. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_deprecated_commands.py +0 -0
  260. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_deprecated_flags.py +0 -0
  261. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_draft_ask_commands.py +0 -0
  262. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_draft_commands.py +0 -0
  263. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_draft_conversation_commands.py +0 -0
  264. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_draft_feature_commands.py +0 -0
  265. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_draft_pr_commands.py +0 -0
  266. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_env.py +0 -0
  267. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_flow_types_commands.py +0 -0
  268. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_init_command.py +0 -0
  269. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_issues.py +0 -0
  270. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_issues_commands.py +0 -0
  271. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_issues_get_command.py +0 -0
  272. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_issues_triage_commands.py +0 -0
  273. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_knowledge_v2_commands.py +0 -0
  274. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_mermaid.py +0 -0
  275. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_native_device_types.py +0 -0
  276. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_network_errors.py +0 -0
  277. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_physical_devices.py +0 -0
  278. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_private_network_commands.py +0 -0
  279. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_recording_timeline.py +0 -0
  280. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_reference_command.py +0 -0
  281. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_residential_proxy_commands.py +0 -0
  282. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_response_errors.py +0 -0
  283. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_ask_commands.py +0 -0
  284. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_backend_override.py +0 -0
  285. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_cancel.py +0 -0
  286. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_commands.py +0 -0
  287. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_frames_command.py +0 -0
  288. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_run_logs_command.py +0 -0
  289. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_schedules_commands.py +0 -0
  290. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_screens_commands.py +0 -0
  291. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_secret_commands.py +0 -0
  292. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_secret_create.py +0 -0
  293. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_secret_lint_extract.py +0 -0
  294. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_skill_bundle.py +0 -0
  295. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_skill_command.py +0 -0
  296. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_suite_change_provenance.py +0 -0
  297. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_tags_commands.py +0 -0
  298. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_test_file_commands.py +0 -0
  299. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_test_profile_commands.py +0 -0
  300. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_test_profile_secret_refs.py +0 -0
  301. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_upgrade_auto_command.py +0 -0
  302. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_upgrade_command.py +0 -0
  303. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_user_story_bindings_commands.py +0 -0
  304. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_user_story_commands.py +0 -0
  305. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_user_story_in_app_purchases.py +0 -0
  306. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_user_story_setup_commit.py +0 -0
  307. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_verdicts.py +0 -0
  308. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_verdicts_actionable.py +0 -0
  309. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/test_version.py +0 -0
  310. {minitest_cli-0.47.0 → minitest_cli-0.48.0}/tests/triage_backend.py +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.5
2
2
  Name: minitest-cli
3
- Version: 0.47.0
3
+ Version: 0.48.0
4
4
  Summary: Minitest CLI – command-line interface for the Minitest testing platform
5
5
  Project-URL: Homepage, https://minitap.ai/
6
6
  Project-URL: Source, https://github.com/minitap-ai/minitest-cli
@@ -247,7 +247,7 @@ app defaults.
247
247
  | `minitest screens` | Inspect the screens exploration mapped for an app |
248
248
  | `minitest build` | Native iOS/Android build management |
249
249
  | `minitest run` | Start runs and inspect their scenario runs |
250
- | `minitest batch` | List, inspect and cancel runs (a batch is a Run in the webapp) |
250
+ | `minitest batch` | List, inspect, cancel and delete runs (a batch is a Run in the webapp) |
251
251
  | `minitest schedules` | Scheduled runs: run scenarios at a set time and post a Slack summary |
252
252
  | `minitest issues` | Read app issues and their fix prompts, mark them fixed |
253
253
  | `minitest skill` | Print the CLI skill for your AI coding agent (not app skills) |
@@ -213,7 +213,7 @@ app defaults.
213
213
  | `minitest screens` | Inspect the screens exploration mapped for an app |
214
214
  | `minitest build` | Native iOS/Android build management |
215
215
  | `minitest run` | Start runs and inspect their scenario runs |
216
- | `minitest batch` | List, inspect and cancel runs (a batch is a Run in the webapp) |
216
+ | `minitest batch` | List, inspect, cancel and delete runs (a batch is a Run in the webapp) |
217
217
  | `minitest schedules` | Scheduled runs: run scenarios at a set time and post a Slack summary |
218
218
  | `minitest issues` | Read app issues and their fix prompts, mark them fixed |
219
219
  | `minitest skill` | Print the CLI skill for your AI coding agent (not app skills) |
@@ -1,6 +1,6 @@
1
1
  [project]
2
2
  name = "minitest-cli"
3
- version = "0.47.0"
3
+ version = "0.48.0"
4
4
  description = "Minitest CLI – command-line interface for the Minitest testing platform"
5
5
  readme = "README.md"
6
6
  license = "MIT"
@@ -8,7 +8,7 @@ description: >
8
8
  affects UI, navigation, or user journeys to check if existing tests need
9
9
  to be updated.
10
10
  metadata:
11
- cli-version: "0.46.0"
11
+ cli-version: "0.47.0"
12
12
  ---
13
13
 
14
14
  # Minitest CLI
@@ -1136,7 +1136,7 @@ get`.
1136
1136
  ### 7. Work with batches
1137
1137
 
1138
1138
  A batch groups runs triggered together (by `run all`, CI, or a single
1139
- `run start`). Use the `batch` group to inspect or cancel them.
1139
+ `run start`). Use the `batch` group to inspect, cancel or delete them.
1140
1140
 
1141
1141
  ```bash
1142
1142
  minitest --json --app <app_id> batch list # recent batches
@@ -1145,8 +1145,15 @@ minitest --json --app <app_id> batch list --commit-sha abc1234
1145
1145
  minitest --json --app <app_id> batch list --user-story <id>
1146
1146
  minitest --json --app <app_id> batch get <batch_id> # batch + its runs
1147
1147
  minitest --json --app <app_id> batch cancel <batch_id> # cancels all pending/running runs
1148
+ minitest --json --app <app_id> batch delete <batch_id>... --yes # deletes finished runs
1148
1149
  ```
1149
1150
 
1151
+ `batch delete` takes several ids and answers `{"results": [{"id", "deleted", "status", "error"}]}`,
1152
+ one entry per id; it exits non-zero if any id failed. A deleted run leaves the runs list, scenario
1153
+ history and health for everyone in the workspace, and its share links stop working. There is no
1154
+ undo from the CLI, so only delete a run the user asked you to delete. A run that has not finished
1155
+ is refused (409): cancel it or wait for it first. It needs an owner, admin or API key.
1156
+
1150
1157
  **Batch statuses:** `pending` | `awaiting_build` | `running` | `completed` | `failed` | `cancelled`
1151
1158
 
1152
1159
  ### 8. Read a whole batch’s verdicts in one call
@@ -1601,6 +1608,7 @@ the runs. Use `run verdicts <batch_id>` when you actually want the outcomes.
1601
1608
  | Frame strips from a video | `minitest run frames <video> [--start S] [--end E] [--every SEC] [--at T ...] [--columns N] [--width PX]` |
1602
1609
  | Target logs (device log, crash report, console, HAR) | `minitest --json --app ID run logs <run_id> [--platform P] [--target-id ID] [--out DIR]` |
1603
1610
  | Cancel batch | `minitest --json --app ID batch cancel <batch_id>` |
1611
+ | Delete finished runs | `minitest --json --app ID batch delete <batch_id>... --yes` |
1604
1612
  | Create a schedule | `minitest --json --app ID schedules create --name N --at 03:00 [--days weekdays] [--timezone Europe/Oslo] [--tag T \| --scenario S] (--url U --browser chrome \| --android [--android-build ID]) --slack-channel C…` — see [`commands/schedules.md`](commands/schedules.md) |
1605
1613
  | Preview a schedule | `minitest --json --app ID schedules preview <schedule>` (or the `create` flags): duration, builds tonight would pick |
1606
1614
  | List / show schedules | `minitest --json --app ID schedules list` / `schedules get <schedule>` |
@@ -154,5 +154,6 @@ Steps must be executable as-is by someone who has never seen the app:
154
154
  - Every capability-matrix row is covered by ≥ 1 scenario.
155
155
  - Every step is atomic and human-reproducible per the rules above.
156
156
  - No acceptance criterion mentions backend, database, network, or any non-visible
157
- effect, except a backend state the app knowledge documents a read call for,
158
- asserted next to a criterion for what the user sees.
157
+ effect, except a backend state the app knowledge, an app skill or the app's
158
+ Backend API documents a read call for, asserted next to a criterion for what the
159
+ user sees.
@@ -152,19 +152,19 @@ with evidence. **Do not begin Wave 3 until `02-capability-matrix.md` exists.**
152
152
  Then **you** derive `03-personas.yaml`, applying the account-gating
153
153
  and persona doctrine below — the minimal basis set, never the cross-product.
154
154
 
155
- **What separates one account from another is the highest-value signal in the app — and the easiest to miss.** Two users on the same screen can see different things: a Pro badge, an extra tab, a paywall, a region-only feature, an experiment variant. Hunt these differentiators on purpose while you explore; don't treat them as incidental. Each one you find is a persona worth testing and a flow worth covering; each one you miss is a whole surface that never gets tested.
155
+ **What separates one account from another is the highest-value signal in the app — and the easiest to miss.** Two users on the same screen can see different things: a Pro badge, an extra tab, a paywall, a region-only feature, an experiment variant. Hunt these differentiators on purpose while you explore; don't treat them as incidental. Each one you find is a test account worth creating and a flow worth covering; each one you miss is a whole surface that never gets tested.
156
156
 
157
157
  Look for every axis that gates what a user can reach or do:
158
158
 
159
159
  - **Tiers and entitlements** — free vs paid plans, subscription levels, `capabilities`/`permissions` on the user, paywalls (hard = blocks the feature, soft = dismissible upsell).
160
160
  - **Roles and permission guards** — route or screen guards keyed on a role (Driver vs Diner, Admin vs Member), features hidden behind a permission.
161
- - **Feature flags and remote config** — static flags in the build and server-driven ones (LaunchDarkly, Firebase Remote Config, a homegrown `/config` call). A remote flag's real value is unknowable from the code alone: you can see the flag exists, not what it's set to. Model the variant most users see, record the assumption in app knowledge, and never write a criterion that silently assumes one side of a flag you can't read.
161
+ - **Feature flags and remote config** — static flags in the build and server-driven ones (LaunchDarkly, Firebase Remote Config, a homegrown `/config` call). A remote flag's real value is unknowable from the code alone: you can see the flag exists, not what it's set to. Model the variant most users see, record the assumption in the App guide, and never write a criterion that silently assumes one side of a flag you can't read.
162
162
  - **Locale, region, and verification state** — country/language gates, KYC/age/verification walls, features that only appear once an account is verified.
163
163
  - **Data state** — empty vs populated (a fresh account with no orders vs a returning one with history) often renders entirely different screens.
164
164
 
165
- **Cover each differentiated capability once, with the cheapest persona that unlocks it — not the cross-product of every feature by every persona.** If a feature behaves the same for Free and Pro, test it once as the cheaper persona; spend a Pro-only scenario on what Pro actually changes. The goal is every capability covered, not every persona re-walking every flow.
165
+ **Cover each differentiated capability once, with the cheapest test account that unlocks it — not every feature crossed with every account.** If a feature behaves the same for Free and Pro, test it once on the cheaper account; spend a Pro-only scenario on what Pro actually changes. The goal is every capability covered, not every account re-walking every flow.
166
166
 
167
- **"Cheapest" means cheapest to run, not cheapest to set up.** The two point in opposite directions and the wrong reading is the tempting one: the system "New user" persona looks cheapest because it needs no credentials and no provisioning, while at run time it is the most expensive entry the app has — a full registration funnel before a single assertion, every run. A pre-existing account costs one sign-in. Rank personas by what each run pays, and see ""New user" is for stories whose SUBJECT is being new" in the personas guidance for why that same choice also decides whether the story can have a parent at all.
167
+ **"Cheapest" means cheapest to run, not cheapest to set up.** The two point in opposite directions and the wrong reading is the tempting one: the system "New user" account looks cheapest because it needs no credentials and no provisioning, while at run time it is the most expensive entry the app has — a full registration funnel before a single assertion, every run. A pre-existing account costs one sign-in. Rank test accounts by what each run pays, and see ""New user" is for scenarios whose SUBJECT is being new" in the test accounts guidance for why that same choice also decides whether the scenario can have a parent at all.
168
168
 
169
169
 
170
170
  A **test account** (a *test profile* in the CLI and API) is one specific user identity and state. Scenarios (*user stories* in the API) bind test accounts many-to-many: a scenario may bind several, none is primary, and the tester picks whichever fits each step. Dependency checks compare the alphabetically first one bound.
@@ -255,37 +255,37 @@ wiring are now **derived facts**, not guesses.
255
255
  Compose `suite.yaml` in `$WORK` (schema + step-writing rules in `minitest reference suite-schemas`),
256
256
  governed by the doctrine below.
257
257
 
258
- **A user story = a full session, and it is self-contained.** Each story runs as a fresh tester session and owns its own path to the exact state it tests — signing in, onboarding, navigating, and setting up whatever cheap in-app state it needs. It never assumes a previous story left the device in a particular state — in a chain the device and app are shared, so it may start wherever the previous story stopped, and it checks where it is before taking its path. Even a story with dependencies must be runnable on its own. Put that setup path in the **description** (as straightforwardly as possible, so the agent reaches the testable state fast); the acceptance criteria stay focused on WHAT to verify once there. Getting a device, loading the app, signing in, and reaching the feature is the bulk of the cost — once the agent is there, it should exercise the feature exhaustively before the session is torn down.
258
+ **A scenario = a full session, and it is self-contained.** Each scenario runs as a fresh tester session and owns its own path to the exact state it tests — signing in, onboarding, navigating, and setting up whatever cheap in-app state it needs. It never assumes a previous scenario left the device in a particular state — in a chain the device and app are shared, so it may start wherever the previous scenario stopped, and it checks where it is before taking its path. Even a scenario with dependencies must be runnable on its own. Put that setup path in the **description** (as straightforwardly as possible, so the agent reaches the testable state fast); the acceptance criteria stay focused on WHAT to verify once there. Getting a device, loading the app, signing in, and reaching the feature is the bulk of the cost — once the agent is there, it should exercise the feature exhaustively before the session is torn down.
259
259
 
260
- The only state a story does **not** recreate itself is state that is expensive or irreversible to produce (an upgraded account, a placed order, moderated content). That comes from a real dependency (see the dependencies guidance). Identity comes from the persona bound to the story, never from an earlier story. Everything cheap and reversible, the story does itself.
260
+ The only state a scenario does **not** recreate itself is state that is expensive or irreversible to produce (an upgraded account, a placed order, moderated content). That comes from a real dependency (see the dependencies guidance). Identity comes from the test account bound to the scenario, never from an earlier scenario. Everything cheap and reversible, the scenario does itself.
261
261
 
262
- **Default to ONE story per feature area.** Pack everything reachable in a single uninterrupted session into the same story's acceptance criteria. The criteria list IS the test script; long criteria lists are GOOD, not a smell.
262
+ **Default to ONE scenario per feature area.** Pack everything reachable in a single uninterrupted session into the same scenario's acceptance criteria. The criteria list IS the test script; long criteria lists are GOOD, not a smell.
263
263
 
264
- **Only create a separate story when starting from scratch is actually required.** Split a flow into multiple stories ONLY if one of these is true:
264
+ **Only create a separate scenario when starting from scratch is actually required.** Split a flow into multiple scenarios ONLY if one of these is true:
265
265
 
266
266
  1. **App state must change between scenarios** in a way the agent cannot undo in the same session (e.g. "first-launch onboarding" vs. "returning user home screen" — the first needs a clean environment).
267
- 2. **A different profile/persona is needed** (e.g. "Driver accepts delivery" vs. "Diner places order" — different accounts, different permissions). **Exception — collaborative journeys:** when a single flow genuinely spans two accounts (sign in as A and send a message, sign out, sign in as B and confirm B received it), keep it as ONE story and bind BOTH personas to it. The tester drives both accounts sequentially on one device, switching accounts as its own setup steps — set an explicit device count of 1, since binding two personas otherwise gets one device per persona. Split only when the two roles run genuinely separate journeys, not when one journey needs both.
267
+ 2. **A different test account is needed** (e.g. "Driver accepts delivery" vs. "Diner places order" — different accounts, different permissions). **Exception — collaborative journeys:** when a single flow genuinely spans two accounts (sign in as A and send a message, sign out, sign in as B and confirm B received it), keep it as ONE scenario and bind BOTH test accounts to it. The tester drives both accounts sequentially on one device, switching accounts as its own setup steps — set an explicit device count of 1, since binding two test accounts otherwise gets one device per test account. Split only when the two roles run genuinely separate journeys, not when one journey needs both.
268
268
  3. **A different entry path is required** that cannot be reached by in-app navigation (e.g. "deep link opens product page from external app").
269
- 4. **The feature is genuinely independent** and isolating it gives clearer failure signals (e.g. "Login" stays its own story even though every other story logs in — because login failure should not cascade into ten unrelated reds).
269
+ 4. **The feature is genuinely independent** and isolating it gives clearer failure signals (e.g. "Login" stays its own scenario even though every other scenario logs in — because login failure should not cascade into ten unrelated reds).
270
270
 
271
- **If none of those apply, fold the scenarios into ONE story.** Examples of what to collapse:
271
+ **If none of those apply, fold the scenarios into ONE scenario.** Examples of what to collapse:
272
272
 
273
- - Five separate stories about viewing, updating, and managing a single resource (e.g. orders, bookings, listings) → ONE story covering the full lifecycle in a single session.
274
- - "Add to Cart" + "View Cart" + "Update Quantity" + "Remove from Cart" → ONE story "Cart Management" with the four flows as sequential criteria.
275
- - "Open Settings" + "Toggle Notifications" + "Change Language" + "View About" → ONE story "Settings Screen" walking through each toggle and screen.
273
+ - Five separate scenarios about viewing, updating, and managing a single resource (e.g. orders, bookings, listings) → ONE scenario covering the full lifecycle in a single session.
274
+ - "Add to Cart" + "View Cart" + "Update Quantity" + "Remove from Cart" → ONE scenario "Cart Management" with the four flows as sequential criteria.
275
+ - "Open Settings" + "Toggle Notifications" + "Change Language" + "View About" → ONE scenario "Settings Screen" walking through each toggle and screen.
276
276
 
277
- **Test before splitting:** ask yourself "could a tester, already logged in and standing on the home screen, walk through both of these without resetting the environment, switching accounts, or restarting the app?" If yes — same story.
277
+ **Test before splitting:** ask yourself "could a tester, already logged in and standing on the home screen, walk through both of these without resetting the environment, switching accounts, or restarting the app?" If yes — same scenario.
278
278
 
279
- **Coverage:** cover every major user-facing flow. Login, registration, onboarding, the app's core value prop, search, profile, settings — if the app has it, there must be a story. Cover the happy path end-to-end AND the paths that can break: failure states, validation errors, permission/auth denials, empty states, and edge cases. Prefer fewer, well-defined stories over many vague ones.
279
+ **Coverage:** cover every major user-facing flow. Login, registration, onboarding, the app's core value prop, search, profile, settings — if the app has it, there must be a scenario. Cover the happy path end-to-end AND the paths that can break: failure states, validation errors, permission/auth denials, empty states, and edge cases. Prefer fewer, well-defined scenarios over many vague ones.
280
280
 
281
- **Feature flags:** when a flow's visible version depends on a rollout or feature flag, model the most common variant (the one most users see). If a major variant is gated to a small rollout, note it in the story description or app knowledge rather than authoring ambiguous criteria.
281
+ **Feature flags:** when a flow's visible version depends on a rollout or feature flag, model the most common variant (the one most users see). If a major variant is gated to a small rollout, note it in the scenario description or the App guide rather than authoring ambiguous criteria.
282
282
 
283
283
 
284
- A dependency is **not** a way to put the device into a prior state. Every story is **self-contained**: it reaches the exact state it tests with its own setup path, then verifies its criteria. Any story — even a leaf deep in a chain — can run on its own; ancestors are never replayed. A dependency does three things:
284
+ A dependency is **not** a way to put the device into a prior state. Every scenario is **self-contained**: it reaches the exact state it tests with its own setup path, then verifies its criteria. Any scenario — even a leaf deep in a chain — can run on its own; ancestors are never replayed. A dependency does three things:
285
285
 
286
286
  1. **Fail-fast.** When a parent's run fails (infra error, crash, cancellation), is escalated for review, or completes with its tester reporting that the dependents' starting state is now unreachable (judged by reachability, not severity), the batch engine **cascades a SKIPPED state** to every not-yet-started transitive dependent on the same platform. A parent that merely ends with a critical verdict does **not** skip its dependents on its own. The user sees one root-cause red instead of a wall of twenty.
287
- 2. **Device reuse.** The engine runs a chain on one device, in dependency order, launching the app once for the whole chain — so a dependent may start wherever its parent left the app, and must check and reach its own state. This is an optimization the engine applies — never a correctness crutch the story leans on.
288
- 3. **Ordering.** A dependent never starts until each of its parents in the batch has finished and been scored, so backend state a parent produced already exists when the dependent runs. A story whose parent is not in the batch runs as a root: batch creation never pulls ancestors in.
287
+ 2. **Device reuse.** The engine runs a chain on one device, in dependency order, launching the app once for the whole chain — so a dependent may start wherever its parent left the app, and must check and reach its own state. This is an optimization the engine applies — never a correctness crutch the scenario leans on.
288
+ 3. **Ordering.** A dependent never starts until each of its parents in the batch has finished and been scored, so backend state a parent produced already exists when the dependent runs. An ordinary run never pulls ancestors in: a scenario whose parent is not in the run runs as a root.
289
289
 
290
290
  Declare these relationships up front — the cascade engine cannot infer them.
291
291
 
@@ -301,24 +301,24 @@ Two questions, in order:
301
301
  > **2. "To run B, some state must already exist. Can B create that state itself — cheaply and safely — with its own setup (setup ≠ acceptance criteria)?"**
302
302
 
303
303
  - **Yes → reducible.** Inline the setup into B's own description; declare **no** dependency for it.
304
- - **No → irreducible.** The state is expensive or irreversible to produce (account upgraded, order placed, content moderated). Give it a dedicated story and declare a real dependency on that story.
304
+ - **No → irreducible.** The state is expensive or irreversible to produce (account upgraded, order placed, content moderated). Give it a dedicated scenario and declare a real dependency on that scenario.
305
305
 
306
- The bar is **deliberately high**. Cheap in-app state (an item in the cart, a filter applied, a screen navigated to) is always reducible — the story does it itself. The feature's value comes from the few foundational flows (login, onboarding, subscription upgrade) and from genuinely irreversible produced state, not from modeling the whole app graph.
306
+ The bar is **deliberately high**. Cheap in-app state (an item in the cart, a filter applied, a screen navigated to) is always reducible — the scenario does it itself. The feature's value comes from the few foundational flows (login, onboarding, subscription upgrade) and from genuinely irreversible produced state, not from modeling the whole app graph.
307
307
 
308
- **A record only a commit makes is irreducible, even when the commit is cheap.** A booking or an order is real state under the persona's address, and the agents that explore and check feasibility stop at a commit step unless the customer explicitly allowed that story's commit (`minitest scenario allow-setup-commit`). With that permission, the producer's feasibility check performs its commit, and a direct dependent's check runs the producer first on the same device. Without it, a story that makes the record inline as setup is only ever checked up to that setup step. Give it a producer story on the same persona, and depend on it. Where the customer does not allow that commit, no story can produce the record: scope the reading story to where it stops and say so, rather than inventing a producer.
308
+ **A record only a commit makes is irreducible, even when the commit is cheap.** A booking or an order is real state under the test account's address, and the agents that explore and check feasibility stop at a commit step unless the customer explicitly allowed that scenario's commit (`minitest scenario allow-setup-commit <id> --customer-quote "<their words>"`). With that permission, the producer's feasibility check performs its commit, and a direct dependent's check runs the producer first on the same device. Without it, a scenario that makes the record inline as setup is only ever checked up to that setup step. Give it a producer scenario on the same test account, and depend on it. Where the customer does not allow that commit, no scenario can produce the record: scope the reading scenario to where it stops and say so, rather than inventing a producer.
309
309
 
310
- **Every non-gate edge names its state.** A dependency that exists for device-reuse (not just fail-fast) has to justify itself: name the exact state the parent *produces* and the child *consumes*, and say why that state is expensive or irreversible to re-create — that's what keeps it above the reducible bar. ✅ "Checkout depends on Place Order: Place Order leaves the account with a placed, non-cancellable order that Checkout's refund flow reads." ✅ "Manage a booking depends on Make a booking: Make a booking leaves an active reservation under the persona's `@qa.minitap.ai` address, whose confirmation number Manage a booking reads from that inbox." ❌ "Checkout depends on Add to Cart" — a cart item is cheap in-app state the child sets up itself. **No such state, no edge:** if you can't name it, the dependency is reducible and shouldn't exist. Gate (sign-in/onboarding) edges are exempt — they exist purely for fail-fast, so "requires the persona's session" is reason enough.
310
+ **Every non-gate edge names its state.** A dependency that exists for device-reuse (not just fail-fast) has to justify itself: name the exact state the parent *produces* and the child *consumes*, and say why that state is expensive or irreversible to re-create — that's what keeps it above the reducible bar. ✅ "Checkout depends on Place Order: Place Order leaves the account with a placed, non-cancellable order that Checkout's refund flow reads." ✅ "Manage a booking depends on Make a booking: Make a booking leaves an active reservation under the test account's `@qa.minitap.ai` address, whose confirmation number Manage a booking reads from that inbox." ❌ "Checkout depends on Add to Cart" — a cart item is cheap in-app state the child sets up itself. **No such state, no edge:** if you can't name it, the dependency is reducible and shouldn't exist. Gate (sign-in/onboarding) edges are exempt — they exist purely for fail-fast, so "requires the test account's session" is reason enough.
311
311
 
312
- **Keep chains shallow** (advice — the server enforces no depth). The dependency graph is a shallow DAG. Counting only non-gate edges (gate roots don't add depth), no chain should run more than **three** levels deep. A deeper chain is a smell: a story buried under that many prerequisites is almost always one that should have inlined its own cheap state instead.
312
+ **Keep chains shallow** (advice — the server enforces no depth). The dependency graph is a shallow DAG. Counting only non-gate edges (gate roots don't add depth), no chain should run more than **three** levels deep. A deeper chain is a smell: a scenario buried under that many prerequisites is almost always one that should have inlined its own cheap state instead.
313
313
 
314
314
  ### Isolate write scenarios
315
315
 
316
316
  A scenario that **produces irreversible state** (places an order, upgrades an account, posts content, cancels a subscription) is a **write** scenario. Keep write scenarios apart:
317
317
 
318
- - **One writer per irreversible resource.** Never let two scenarios mutate the same irreversible state on the same account in ways that can run concurrently — the winner is nondeterministic and both readings become flaky. If two flows both need to place an order, give each its own persona (its own chain) so they can never collide.
318
+ - **One writer per irreversible resource.** Never let two scenarios mutate the same irreversible state on the same account in ways that can run concurrently — the winner is nondeterministic and both readings become flaky. If two flows both need to place an order, give each its own test account (its own chain) so they can never collide.
319
319
  - **A write scenario is a chain leaf for anything that reads what it wrote.** If B verifies the result of A's write, B depends on A (irreducible state) and A's write is not re-run underneath B. Do not fan out multiple concurrent writers under one shared setup root.
320
- - **Reads are free to parallelize; writes are not.** Read-only journeys on the same persona can run independently; write journeys on the same persona must be serialized by the dependency chain that owns that state.
321
- - **A persona marked `exclusive_use`** is held by at most one running test at a time: the dispatcher waits rather than run two tests on it concurrently. Use it for an account whose state any concurrent run would corrupt.
320
+ - **Reads are free to parallelize; writes are not.** Read-only journeys on the same test account can run independently; write journeys on the same test account must be serialized by the dependency chain that owns that state.
321
+ - **A test account marked `exclusive_use`** is held by at most one running test at a time: the dispatcher waits rather than run two tests on it concurrently. Use it for an account whose state any concurrent run would corrupt.
322
322
 
323
323
  ### Concrete examples
324
324
 
@@ -328,43 +328,43 @@ A scenario that **produces irreversible state** (places an order, upgrades an ac
328
328
  - **`Edit Profile` is *not* a dependency of `View Profile`** — viewing works whether or not editing did.
329
329
  - **`Search` is *not* a dependency of `Settings`** — independent journeys that share an app shell.
330
330
 
331
- ### One persona per chain (hard rule)
331
+ ### One test account per chain (hard rule)
332
332
 
333
- Every story in a chain shares the SAME persona. Identity is **not** carried along a chain: the tester receives the bound persona's credentials on every story, and no signed-in session travels to another device. The system "New user" persona signs up fresh on every story it runs (a new `<random>@qa.minitap.ai` address each time), so a "New user" dependent never sees an account its parent created — bind a standard persona to any chain whose dependents read what the parent made.
333
+ Every scenario in a chain shares the SAME test account. Identity is **not** carried along a chain: the tester receives the bound test account's credentials on every scenario, and no signed-in session travels to another device. The system "New user" test account signs up fresh on every scenario it runs (a new `<random>@qa.minitap.ai` address each time), so a "New user" dependent never sees an account its parent created — bind a standard test account to any chain whose dependents read what the parent made.
334
334
 
335
- Give each persona one **"Sign in as `<Persona>`" root story** bound to that persona's profile, then chain that persona's stories off it — every node, root included, carries that profile. Two personas must be two separate chains, each rooted in its own sign-in story. The backend rejects an edge whose parent carries a different persona than the child's primary persona with `422 persona_mismatch` (a parent with no persona is accepted), so a chain wired wrong will fail to save. **Never change a story's persona just to make an edge legal** — the fix is always to point the story at its own persona's sign-in root.
335
+ Give each test account one **"Sign in as `<Account>`" root scenario** bound to it, then chain that test account's scenarios off it — every node, root included, carries that test account. Two test accounts must be two separate chains, each rooted in its own sign-in scenario. The backend rejects an edge whose parent carries a different test account than the child's primary test account with `422 persona_mismatch` (a parent with no test account is accepted), so a chain wired wrong will fail to save. **Never change a scenario's test account just to make an edge legal** — the fix is always to point the scenario at its own test account's sign-in root.
336
336
 
337
- - ✅ `Sign in as Diner (Diner) → Browse Restaurants (Diner) → Checkout (Diner)` — one persona, one sign-in root, same profile on every node.
338
- - ✅ `Sign in as Driver (Driver) → Accept Delivery (Driver)` and a separate `Sign in as Diner (Diner) → Place Order (Diner)` — two personas, two independent chains.
337
+ - ✅ `Sign in as Diner (Diner) → Browse Restaurants (Diner) → Checkout (Diner)` — one test account, one sign-in root, same test account on every node.
338
+ - ✅ `Sign in as Driver (Driver) → Accept Delivery (Driver)` and a separate `Sign in as Diner (Diner) → Place Order (Diner)` — two test accounts, two independent chains.
339
339
  - ❌ `Sign in as Driver (Driver) → Place Order (Diner)` — two accounts in one chain; Place Order belongs under the Diner sign-in root.
340
- - ❌ Re-binding a child to a different persona purely to satisfy the edge — keep its identity, rewire the edge.
340
+ - ❌ Re-binding a child to a different test account purely to satisfy the edge — keep its identity, rewire the edge.
341
341
 
342
- Moving a story and everything that depends on it to another persona is the one legitimate reason to change a bound story's profile. Updating a story with `cascadePersonaToDependents: true` rebinds its entire dependent subtree to the new persona in one write instead of rejecting the change; without it, the update is refused whenever a dependent would be left on the old persona. It rewrites each descendant's bindings to exactly the new persona, dropping any others, and it is a REST `PATCH` field only — minitest-cli does not expose it.
342
+ Moving a scenario and everything that depends on it to another test account is the one legitimate reason to change a bound scenario's test account. Updating a scenario with `cascadePersonaToDependents: true` rebinds its entire dependent subtree to the new test account in one write instead of rejecting the change; without it, the update is refused whenever a dependent would be left on the old test account. A descendant whose test account changes is rebound to exactly the new account, dropping any others; one already on it keeps its bindings. It is a REST `PATCH` field only — minitest-cli does not expose it.
343
343
 
344
344
  ### Constraints (server-validated)
345
345
 
346
- - **Same-app only.** A story can't depend on a story from another app.
346
+ - **Same-app only.** A scenario can't depend on a scenario from another app.
347
347
  - **No cycles.** Direct (A→A), transitive (A→B→A), or longer chains are all rejected.
348
- - **References must exist.** Don't invent IDs; pull them from the story list. A deleted story counts as missing.
349
- - **Replace-all semantics.** `depends_on=[]` clears the set; omitting it leaves it untouched. In minitest-cli, `--depends-on` replaces the whole set, `--remove-dependency` drops individual parents (ignored when `--depends-on` is also given), and `create --depends-on` sets the edges in a second call after the story is created — a rejected edge leaves the story saved without it.
348
+ - **References must exist.** Don't invent IDs; pull them from the scenario list. A deleted scenario counts as missing.
349
+ - **Replace-all semantics.** `depends_on=[]` clears the set; omitting it leaves it untouched. In minitest-cli, `--depends-on` replaces the whole set, `--remove-dependency` drops individual parents (ignored when `--depends-on` is also given), and `create --depends-on` sets the edges in a second call after the scenario is created — a rejected edge leaves the scenario saved without it.
350
350
  - **Errors.** A violation returns 422 with `detail.kind` ∈ `self_dependency | references_missing | cross_app | persona_mismatch | cycle` and the offending `ids`, one kind at a time.
351
- - **Per-platform isolation.** Dependencies are resolved separately per platform — a failure on one target never skips another target's story. If a parent does not run on a platform (disabled or absent there), its dependents attach to the nearest ancestor that does and are not skipped. You don't declare per-platform; the engine handles it.
351
+ - **Per-platform isolation.** Dependencies are resolved separately per platform — a failure on one target never skips another target's scenario. If a parent does not run on a platform (disabled or absent there), its dependents attach to the nearest ancestor that does and are not skipped. You don't declare per-platform; the engine handles it.
352
352
 
353
353
 
354
- Every story has three parts: **title**, **description**, and **acceptance criteria**. Always author all three — the API accepts a story without a description or criteria, but the tester then only has the title as intent and nothing to verify. Limits: title ≤255 characters, description ≤5000, each criterion 1–2000 (no cap on how many).
354
+ Every scenario has three parts: **title**, **description**, and **acceptance criteria**. Always author all three — the API accepts a scenario without a description or criteria, but the tester then only has the title as intent and nothing to verify. Limits: title ≤255 characters, description ≤5000, each criterion 1–2000 (no cap on how many).
355
355
 
356
356
  **Title** — starts with a verb. Action-oriented and user-facing (e.g. "Add Item to Cart", "Comment on Own Game", "Discover Restaurants and Manage Cards"). Never noun phrases like "Restaurant Discovery and Card Management".
357
357
 
358
- **Description** — one or two sentences explaining what the user does and any important scope constraints. This tells the tester agent the intent of the story without reading all criteria. Example: "Logged-in user opens the comments drawer on one of their OWN published games and posts a comment. Do not comment on other users' games." When reaching the tested state is not obvious from the criteria alone, the description is also where you spell out that setup path (how to get there) — the criteria then stay focused on what to verify once there. Keep it a path, not choreography: name the state to reach, not every tap.
358
+ **Description** — one or two sentences explaining what the user does and any important scope constraints. This tells the tester agent the intent of the scenario without reading all criteria. Example: "Logged-in user opens the comments drawer on one of their OWN published games and posts a comment. Do not comment on other users' games." When reaching the tested state is not obvious from the criteria alone, the description is also where you spell out that setup path (how to get there) — the criteria then stay focused on what to verify once there. Keep it a path, not choreography: name the state to reach, not every tap.
359
359
 
360
- **Acceptance criteria** — each criterion is a **job to be done**, not a micro-step or a single interaction. Mini acts like a user, so it can figure out HOW to accomplish a goal up to a point — as long as it's given enough context grounded in how the app actually works. The criteria say WHAT to accomplish and WHAT the expected outcome looks like; the description carries the grounding — the feature/product-area anchor so Mini can orient itself, and the big UX path (a path, not choreography). No Given/When/Then. No exact interaction sequences. No embedded reasoning. Aim for 5-15 criteria per story.
360
+ **Acceptance criteria** — each criterion is a **job to be done**, not a micro-step or a single interaction. Mini acts like a user, so it can figure out HOW to accomplish a goal up to a point — as long as it's given enough context grounded in how the app actually works. The criteria say WHAT to accomplish and WHAT the expected outcome looks like; the description carries the grounding — the feature/product-area anchor so Mini can orient itself, and the big UX path (a path, not choreography). No Given/When/Then. No exact interaction sequences. No embedded reasoning. Aim for 5-15 criteria per scenario.
361
361
 
362
362
  Each criterion **must** be:
363
363
 
364
364
  - **Visually verifiable** — Mini (the tester agent) judges only what the running app shows, even when it can read the app's source. Write assertions about what is visible: text, UI elements, navigation state.
365
365
  - Good: "A success toast or confirmation message is displayed"
366
366
  - Bad: "The backend creates a new user record in the database"
367
- - The one exception is backend state the app knowledge or an app skill documents a read call for: a criterion may then assert it ("the API shows the order as paid"), and Mini judges it on that call's response. Write it next to the criterion for what the user sees, never in place of it.
367
+ - The one exception is backend state that an app skill or the app's Backend API (its OpenAPI document) offers a read call for: a criterion may then assert it ("the API shows the order as paid"), and Mini judges it on that call's response. Write it next to the criterion for what the user sees, never in place of it.
368
368
  - **Specific and unambiguous** — avoid vague assertions.
369
369
  - Good: "The cart badge shows the updated item count"
370
370
  - Bad: "The cart works correctly"
@@ -385,11 +385,11 @@ Each criterion **must NOT**:
385
385
  - Hardcode verbatim marketing copy or long strings from the app. Reference content by purpose ("the onboarding slide explains the app's core feature") not by literal text.
386
386
  - Require anything an agent on a real device or browser cannot check. No server-side checks or DB inspection beyond a documented read call (see "Visually verifiable").
387
387
 
388
- **Never assert a volatile value.** Cart contents, unread counts, badges, streak days, playback position, and state-dependent labels such as "Add to cart" versus "Go to cart" all change as earlier stories run — and change again on the next run. Assert the capability, not the number that happened to be on screen when the story was written: "the cart control is present and opens the cart", not "the cart badge shows 3". A criterion has to be as true next month as it was the day it was authored.
388
+ **Never assert a volatile value.** Cart contents, unread counts, badges, streak days, playback position, and state-dependent labels such as "Add to cart" versus "Go to cart" all change as earlier scenarios run — and change again on the next run. Assert the capability, not the number that happened to be on screen when the scenario was written: "the cart control is present and opens the cart", not "the cart badge shows 3". A criterion has to be as true next month as it was the day it was authored.
389
389
 
390
390
  **Customer-specified test data is part of the WHAT, not the HOW.** When a customer's test case names the value under test — a search term, product, coupon code, amount, address or option — or a precise expected outcome, keep it verbatim in the criterion, quoted in its original language: "Search for 'nutella' and confirm no 'nutella' product appears in the results". The rules against choreography and verbatim copy govern how Mini reaches a goal and the app's own copy, not the data the customer chose to test. Never replace such a value with "e.g.", a category, or a description of it: a criterion that reads "search for a product expected to be unavailable" lets Mini pick its own term, and the customer's case never runs. If the value will go stale (a price, a stock count), keep it and flag the risk to the customer instead of generalising it.
391
391
 
392
- **Never inline credentials or profile-specific data into criteria.** A story's login/account details come from its linked test profile, which the customer can edit at any time. Hardcoding the email, password, or other profile values into a criterion duplicates that data and goes stale the moment the profile changes. Refer to the persona abstractly instead, e.g. "Sign in with the profile's credentials and confirm the home screen loads". At run time the backend also rejects a criterion report whose evidence, summary, failure reason or RCA text contains a persona's password, static OTP code, test-card number or CVC, so never write a criterion that asks Mini to state or confirm one — phrase it as "signs in successfully" instead.
392
+ **Never inline credentials or test-account data into criteria.** A scenario's sign-in details come from its bound test account, which the customer can edit at any time. Hardcoding the email, password or other account values into a criterion duplicates that data and goes stale the moment the account changes. Refer to the account abstractly instead: "Sign in with the test account's credentials and confirm the home screen loads". At run time the backend also blocks report text that echoes a test account's password, static OTP code or test card, or an app skill's secret, so never write a criterion that asks Mini to state or confirm one; phrase it as "signs in successfully" instead.
393
393
 
394
394
  **How a criterion is judged.** Each criterion gets exactly one verdict per run and platform: `success`, `failed` (app behaviour Mini watched go wrong), `unprocessable` (Mini could not verify it — a harness limit or missing setup; the reason names who must act), or `skipped` (downstream of an unprocessable criterion, same root cause). A criterion outside Mini's capabilities comes back `unprocessable`, not `failed`.
395
395
 
@@ -425,33 +425,34 @@ Good (goal-oriented):
425
425
 
426
426
  **Tags** are labels on scenarios, shared by every app of the tenant. The UI groups and filters scenarios by tag, and a tag's description is shown to the tester on every scenario carrying it — so tags affect runs, not just grouping.
427
427
 
428
- 1. **List the existing tags first** (`minitest tags list`) and reuse them. Match case-insensitively: never create `checkout` next to an existing `Checkout`.
428
+ 1. **List the existing tags first** (`minitest tags list`) and reuse them. Names are case-insensitive: creating `checkout` when `Checkout` exists returns `Checkout`.
429
429
  2. **Give each scenario 1–3 generic tags** naming its feature area or team (`Checkout`, `Auth`, `Search`, `Onboarding`, `Payments`). A tag is meant to group many scenarios, so **never use the scenario's title as a tag**.
430
- 3. **Create a tag only when nothing existing fits**: `minitest tags create --name <name> [--color <color>] [--description "<what the tester should know>"]`, or pass `--tag <name>` (repeatable) to `minitest scenario create` / `update` — unknown names are created. On `update`, any `--tag` replaces the scenario's whole tag set.
430
+ 3. **Create a tag only when nothing existing fits**: `minitest tags create --name <name> [--color <color>] [--description "<what the tester should know>"]`, or pass `--tag <name>` (repeatable) to `minitest scenario create` / `update` — unknown names are created. On `update`, any `--tag` replaces the scenario's whole tag set, and `--clear-tags` removes them all.
431
431
  4. Tag names are at most 40 characters and contain no comma.
432
- 5. There are **no flow types, story types or categories** anymore: never pass `--type`, never use `minitest flow-types`.
432
+ 5. There are **no flow types, scenario types or categories** anymore: never pass `--type`, never use `minitest flow-types`.
433
433
 
434
434
 
435
- A scenario's **device count** decides how many devices one run drives at once (Android and iOS only — web runs use one browser). A scenario stored without an explicit count uses **auto**: one device per bound persona, minimum one — a story binding zero or one persona runs on a single device, a story binding several personas gets one device per persona at run time. An explicit integer overrides auto in either direction. Both are capped at `min(3, the tenant's device quota, the app's device-concurrency limit)` — a story binding more personas than that cap runs on the cap. Omitting the field when *creating* a story leaves it on auto; on *updates*, omission leaves the current value unchanged — how to set or reset it is documented on each create/update call. The count is decoupled from the personas the story binds — see "Personas vs. devices" in the personas guidance.
435
+ A scenario's **device count** decides how many devices one run drives at once. On web each device is a tab of the one browser: tabs share cookies and storage, so two test accounts cannot be signed in to the same site at once. A scenario stored without an explicit count uses **auto**: one device per bound test account, minimum one. An explicit integer overrides auto in either direction; one above `min(3, the tenant's device quota, the app's device-concurrency limit)` is rejected, and auto never asks for more than `min(3, the tenant's device quota)`. Omitting the field when *creating* a scenario leaves it on auto; on *updates*, omission leaves the current value unchanged and `--device-count auto` resets it. The count is decoupled from the test accounts the scenario binds — see "Test accounts vs. devices" in the test accounts guidance.
436
436
 
437
- **Extra devices are only for simultaneity.** Go above one device **only** for flows that verify **real-time cross-account or cross-device behavior**: live chat or a call between two users, a push or notification one account triggers for another, presence/typing indicators, or cross-account visibility (A blocks B, and B can no longer see A). A journey where one identity acts and another checks the result **sequentially** does not need extra devices — the tester signs in and out on one. Because auto gives a multi-persona story one device per persona, **explicitly set the device count to 1 on every sequential multi-persona story**; never rely on auto to "cover more personas". Most multi-persona suites still run one device per story.
437
+ **Extra devices are only for simultaneity.** Go above one device **only** for flows that verify **real-time cross-account or cross-device behavior**: live chat or a call between two users, a push or notification one account triggers for another, presence/typing indicators, or cross-account visibility (A blocks B, and B can no longer see A). A journey where one identity acts and another checks the result **sequentially** does not need extra devices — the tester signs in and out on one. Because auto gives a scenario one device per bound test account, **explicitly set the device count to 1 on every sequential scenario that binds several test accounts**; never rely on auto to "cover more test accounts". Most suites with several test accounts still run one device per scenario.
438
438
 
439
- Multi-device is the deliberate exception, not a default to reach for. For an ordinary app with no simultaneous-identity flow the correct counts are: **unset on single-persona stories** (auto already means one device) and **an explicit 1 on multi-persona ones** — anything higher changes cost and behavior for nothing.
439
+ Multi-device is the deliberate exception, not a default to reach for. For an ordinary app with no simultaneous-identity flow the correct counts are: **unset on scenarios that bind one test account** (auto already means one device) and **an explicit 1 on those that bind several** — anything higher changes cost and behavior for nothing.
440
440
 
441
441
 
442
- **Cover offline and connectivity scenarios when the app warrants it.** If the app shows offline support, cached content, or sync mechanisms, create dedicated stories to test those flows. Example criterion: "Go offline (airplane mode), confirm the app shows cached content, then go back online and verify data syncs."
442
+ **Cover offline and connectivity scenarios when the app warrants it.** If the app shows offline support, cached content, or sync mechanisms, create dedicated scenarios to test those flows. Example criterion: "Go offline (airplane mode), confirm the app shows cached content, then go back online and verify data syncs."
443
443
 
444
444
  - On Android, name the transition the criterion needs: **"Offline (airplane mode)"** for a fully offline device, **"Wi-Fi off"** only when the app must react to losing Wi-Fi specifically. The tester restores connectivity before it reports.
445
445
  - On web, the tester takes the page offline and back online, so word it as **"Offline"** rather than "Wi-Fi off" or "airplane mode".
446
446
  - **iOS** runs cannot change connectivity — do not write network-toggle criteria for iOS; such a scenario is screened as a capability gap there.
447
447
 
448
448
 
449
- **Bind a test file to every story that picks one.** Test devices start with empty storage: an Android device has no photos, videos, documents or audio at all, and an iOS simulator has only a few stock photos, never the file the story needs. So whenever a story picks something from the device — a photo or video from the gallery, a document from the file picker, an audio clip, anything it attaches, uploads or imports — bind a test file of that kind when you create the story. Do not open the picker first to check: it is always empty until a file is bound.
449
+ **Bind a test file to every scenario that picks one.** Test devices start with empty storage: an Android device has no photos, videos, documents or audio at all, and an iOS simulator has only a few stock photos, never the file the scenario needs. So whenever a scenario picks something from the device — a photo or video from the gallery, a document from the file picker, an audio clip, anything it attaches, uploads or imports — bind a test file of that kind when you create the scenario. Do not open the picker first to check: the file the scenario needs is never there until it is bound.
450
450
 
451
451
  - **Generic file** (any photo, a sample PDF, a short audio clip): bind one yourself. Reuse a test file the app already has when it fits; otherwise upload a stock one, then bind it. Never ask the customer for a generic file.
452
- - **Specific real-world file** (an ID document, a real invoice, a printed QR code the app must recognise): only the customer has it. Ask them for it, and bind it once they upload it.
453
- - Binding replaces the story's whole file set: pass every file the story should keep, not only the new one.
454
- - Bound files are pushed to the device before each run, so the story can simply say "Choose the test photo from the library" — never describe how the file gets onto the device.
452
+ - **Specific real-world file** (an ID document, a real invoice): only the customer has it. Ask them for it, and bind it once they upload it.
453
+ - **Something the app scans with the camera** (a QR code, a document held to the lens) is not a bound file: set it as the scenario's camera feed (`--camera-media`, an image or MP4).
454
+ - Binding replaces the scenario's whole file set: pass every file the scenario should keep, not only the new one.
455
+ - Bound files are pushed to the device before each run, so the scenario can simply say "Choose the test photo from the library" — never describe how the file gets onto the device.
455
456
 
456
457
 
457
458
  Design rules on top of that doctrine:
@@ -478,19 +479,19 @@ your own mechanical invariant pass over `suite.yaml` (full list in
478
479
  `minitest reference suite-schemas`), then the suite self-check below. Skipping the fresh-subagent
479
480
  verification or the mechanical pass is a defect.
480
481
 
481
- **Before the suite is final, audit it as if someone else built it.** You wrote these stories knowing what you intended, which is exactly why you'll miss where they don't hold up. Take one cold pass over the finished set and fix what it surfaces — a few quiet corrections here save the user from a suite that looks complete but doesn't run.
482
+ **Before the suite is final, audit it as if someone else built it.** You wrote these scenarios knowing what you intended, which is exactly why you'll miss where they don't hold up. Take one cold pass over the finished set and fix what it surfaces — a few quiet corrections here save the user from a suite that looks complete but doesn't run.
482
483
 
483
- For each story, ask the question a stranger would: **"Given how this app actually works, can this persona reach these screens and observe every criterion?"** If a criterion checks something this tier can't see, a screen this role can't open, or a state the story never set up, correct the story — retarget the persona, inline the missing setup, or fix the criterion.
484
+ For each scenario, ask the question a stranger would: **"Given how this app actually works, can this test account reach these screens and observe every criterion?"** If a criterion checks something this tier can't see, a screen this role can't open, or a state the scenario never set up, correct the scenario — retarget the test account, inline the missing setup, or fix the criterion.
484
485
 
485
486
  Then check the structure mechanically:
486
487
 
487
- - Every chain is rooted in its persona's sign-in story, and every story in a chain shares that one persona (no cross-persona edges).
488
- - No dependency cycles, and chains stay shallow — a story buried under many prerequisites usually just needs to set up its own cheap state instead.
489
- - Every persona you created is used by a story, and every persona a story needs exists.
490
- - Every distinct capability and account-differentiator you found is covered by at least one story.
491
- - Every non-gate edge names the state its parent produces and the child reads; a record only a commit makes comes from a producer on the same persona.
492
- - Every story that binds several personas but switches between them sequentially has an explicit device count of 1.
493
- - Every acceptance criterion is checkable on-screen — no database, server-log or analytics assertions. A criterion may compare the screen against an API response the app itself received. A criterion about backend state stands only when the app knowledge or an app skill documents the read call that answers it, and only next to a criterion for what the user sees.
488
+ - Every chain is rooted in its test account's sign-in scenario, and every scenario in a chain shares that one test account.
489
+ - No dependency cycles, and chains stay shallow — a scenario buried under many prerequisites usually just needs to set up its own cheap state instead.
490
+ - Every test account you created is used by a scenario, and every test account a scenario needs exists.
491
+ - Every distinct capability and account-differentiator you found is covered by at least one scenario.
492
+ - Every non-gate edge names the state its parent produces and the child reads; a record only a commit makes comes from a producer on the same test account.
493
+ - Every scenario that binds several test accounts but switches between them sequentially has an explicit device count of 1.
494
+ - Every acceptance criterion is checkable on-screen — no database, server-log or analytics assertions. A criterion may compare the screen against an API response the app itself received. A criterion about backend state stands only when an app skill or the app's Backend API documents the read call that answers it, and only next to a criterion for what the user sees.
494
495
 
495
496
  Fix issues in place; don't redesign a working suite around one bad edge.
496
497
 
@@ -507,7 +508,7 @@ unsure of a flag.
507
508
  1. **Distill app knowledge.** Fold the recon + capability findings the tester
508
509
  needs at runtime (paywall hard/soft classification, remote-flag assumptions,
509
510
  provisioning notes) into an app-knowledge update:
510
- `minitest app-knowledge update <app id> --content "<distilled knowledge>"`.
511
+ `minitest app-knowledge update --app <app id> --content "<distilled knowledge>"`.
511
512
  App-level assumptions live here; scenario-local assumptions go in that
512
513
  story's description.
513
514
  2. **Create personas.** One `minitest test-profile create` per persona in
@@ -1,4 +1,4 @@
1
- """Batch commands: list, get, cancel."""
1
+ """Batch commands: list, get, cancel, delete."""
2
2
 
3
3
  import math
4
4
  from typing import Annotated
@@ -6,6 +6,7 @@ from typing import Annotated
6
6
  import typer
7
7
 
8
8
  from minitest_cli.api.client import ApiClient
9
+ from minitest_cli.commands.batch_delete import delete
9
10
  from minitest_cli.commands.batch_helpers import batches_base_path
10
11
  from minitest_cli.commands.run_asks_helpers import count_waiting, warn_waiting
11
12
  from minitest_cli.commands.deprecated_alias import deprecated_option, merge_deprecated_option
@@ -19,7 +20,8 @@ from minitest_cli.commands.run_helpers import (
19
20
  from minitest_cli.models.batch import BatchListItem, BatchListResponse, BatchResponse
20
21
  from minitest_cli.utils.output import output, print_info, print_success, print_table
21
22
 
22
- app = typer.Typer(name="batch", help="List, inspect and cancel runs (Runs in the webapp).")
23
+ app = typer.Typer(name="batch", help="List, inspect, cancel and delete runs (Runs in the webapp).")
24
+ app.command(name="delete")(delete)
23
25
 
24
26
  BATCH_TABLE_HEADERS = ["ID", "Status", "Source", "Commit", "Tag", "Targets", "Created"]
25
27
 
@@ -0,0 +1,77 @@
1
+ """``minitest batch delete``: delete finished runs."""
2
+
3
+ from dataclasses import asdict, dataclass
4
+ from typing import Annotated
5
+
6
+ import httpx
7
+ import typer
8
+
9
+ from minitest_cli.api.client import ApiClient
10
+ from minitest_cli.api.errors import format_network_error
11
+ from minitest_cli.commands._response_errors import EXIT_NETWORK_ERROR, extract_detail
12
+ from minitest_cli.commands.batch_helpers import batches_base_path
13
+ from minitest_cli.commands.run_helpers import (
14
+ EXIT_NOT_FOUND,
15
+ ensure_uuid,
16
+ resolve_app,
17
+ run_api_call,
18
+ )
19
+ from minitest_cli.utils.confirm import confirm_or_exit
20
+ from minitest_cli.utils.output import output, print_error, print_success
21
+
22
+
23
+ @dataclass(frozen=True)
24
+ class RunDeletion:
25
+ id: str
26
+ deleted: bool
27
+ status: int | None = None
28
+ error: str | None = None
29
+
30
+
31
+ async def _delete_run(client: ApiClient, app_id: str, batch_id: str) -> RunDeletion:
32
+ """A failure is this run's result, so the other runs still get their turn."""
33
+ try:
34
+ resp = await client.delete(f"{batches_base_path(app_id)}/{batch_id}")
35
+ except httpx.HTTPError as exc:
36
+ return RunDeletion(id=batch_id, deleted=False, error=format_network_error(exc))
37
+ if resp.status_code >= 400:
38
+ error = extract_detail(resp) or f"API error: {resp.status_code}"
39
+ return RunDeletion(id=batch_id, deleted=False, status=resp.status_code, error=error)
40
+ return RunDeletion(id=batch_id, deleted=True, status=resp.status_code)
41
+
42
+
43
+ def _exit_code(deletions: list[RunDeletion]) -> int:
44
+ failures = [deletion for deletion in deletions if not deletion.deleted]
45
+ if not failures:
46
+ return 0
47
+ if all(deletion.status == 404 for deletion in failures):
48
+ return EXIT_NOT_FOUND
49
+ return EXIT_NETWORK_ERROR
50
+
51
+
52
+ def delete(
53
+ batch_ids: Annotated[list[str], typer.Argument(help="Run (batch) IDs to delete.")],
54
+ yes: Annotated[bool, typer.Option("--yes", help="Confirm deleting the runs.")] = False,
55
+ ) -> None:
56
+ """Delete finished runs: they leave the runs list, scenario history and health for everyone
57
+ in the workspace, and their share links stop working. An unfinished run is refused."""
58
+ settings, app_id, json_mode = resolve_app()
59
+ ids = list(dict.fromkeys(ensure_uuid(batch_id, kind="batch id") for batch_id in batch_ids))
60
+ confirm_or_exit(yes, f"Deleting {len(ids)} run(s)")
61
+
62
+ async def _delete_all() -> list[RunDeletion]:
63
+ async with ApiClient(settings) as client:
64
+ return [await _delete_run(client, app_id, batch_id) for batch_id in ids]
65
+
66
+ deletions = run_api_call(_delete_all())
67
+ if json_mode:
68
+ output({"results": [asdict(deletion) for deletion in deletions]}, json_mode=True)
69
+ else:
70
+ for deletion in deletions:
71
+ if deletion.deleted:
72
+ print_success(f"Run deleted: {deletion.id}")
73
+ else:
74
+ print_error(f"{deletion.id}: {deletion.error}")
75
+ exit_code = _exit_code(deletions)
76
+ if exit_code:
77
+ raise typer.Exit(code=exit_code)