feat(evals): canonical top failure modes with failureModeDetail (phase 11) - #26
Merged
Conversation
Audit GAP #10: five named top failure modes now each carry >=50 cases at the literal failureMode level. The previous fine-grained slugs are preserved in the optional failureModeDetail field, and the runner/CLI gain a symmetric --failure-mode-detail filter.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Audit GAP #10: five named top failure modes each carry >=50 cases at the literal failureMode level (reminder-atomicity-violation 50, delivery-concurrency-violation 50, idempotency-replay-duplicate 57, temporal-misinterpretation 60, security-boundary-breach 50). Fine-grained slugs preserved in the new optional failureModeDetail field; runner/CLI gain --failure-mode-detail. Suite unchanged: 299 cases, same IDs/contractRefs/expected.
Gates: ruff/mypy clean, pytest 853 passed, eval gate 299/299, coverage 92%, diff-cover 100%.