fix(evals): --case list mode and name precheck; case-shard qa-callers; refresh batching and design-with-ui seeds

This commit is contained in:
garrytan committed 2026-09-29 20:28:31 +00:00
1 parent 2dae4bf944
commit 581e603313
6 files changed
+41 -14

No files matched your search

+4 -3
View File
@@ -45,11 +45,12 @@ test('the existing quality and behavior phases retain their complete separate sh
expect(behavior.evalsAll).toBe(true);
expect(qualityFiles).toHaveLength(1);
// 44 files (first-task-scaffold registers no gate case, so the gate lane
// skips it); the six case-sharded files contribute one shard per gate case.
// skips it); the seven case-sharded files contribute one shard per gate case.
expect(new Set(behaviorFiles.map(file => file.split('#')[0])).size).toBe(44);
expect(behaviorFiles).toHaveLength(73);
expect(behaviorFiles).toHaveLength(77);
expect(behaviorFiles).toEqual(expect.arrayContaining([
'test/skill-e2e-qa-callers.test.ts',
...['review-exploratory-small-cli', 'ship-exploratory-small-cli', 'ship-exploratory-unavailable',
'ship-exploratory-plan-checks', 'ship-exploratory-late-input'].map(id => `test/skill-e2e-qa-callers.test.ts#${id}`),
'test/skill-e2e-qa-functional-fix.test.ts',
'test/skill-e2e-qa-functional.test.ts',
'test/skill-e2e-ship-skip.test.ts',