Repository navigation
Expose safe diagnostics for invalid evaluator responses - #1
Merged
Merged
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
REV currently collapses malformed evaluator answers into
invalid-response, so callers cannot tell whether an answer is missing, a probability distribution is invalid, or the chosen option disagrees with its probabilities. This has repeatedly blocked aviarytech/tv#21 without actionable diagnostics.Add safe structural diagnostics to the JSON report and action log: the locally generated question identifier, a fixed validation-rule code, and finite numeric aggregates where useful. Preserve fail-closed validation and existing thresholds. Do not record provider strings, raw responses, evidence, or credentials.
Validation: syntax checks and all 34 tests pass. Regression tests cover each structural failure, redaction, report/log propagation, and nonzero exit status on malformed responses.