feat(analyze): aggregate batch error-analysis tool (closes #159) - #255
Merged
Conversation
clawbench-analyze reads a directory of completed run outputs and produces an aggregate error analysis, reusing the per-run classifier (classify_run): - Stage-1 (interception) and Stage-2 (judged) rates - failure taxonomy (result_category counts: intercepted / infra_failure / zero-action / not-intercepted ...) - per-category Stage-1 breakdown (category from task metadata or task-id segments) - interceptor false-positive check (0-action runs that were intercepted -> should be 0), a core validity signal - self-report gap (agent claimed success but the run actually failed) Markdown or --json output. Self-contained + unit-tested on fixtures (5 tests); ruff + pyright clean; suite green.
Perry2004
requested changes
Jul 11, 2026
Perry2004
left a comment
Perry2004
left a comment
Collaborator
There was a problem hiding this comment.
Please update the READMEs about this new script entrypoint as well.
Perry2004
reviewed
Aug 30, 2026
Perry2004
approved these changes
Aug 30, 2026
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Aggregate batch error-analysis tool (closes #159)
clawbench-analyzereads a directory of completed run outputs (<run>/data/interception.json, optional<run>/reward.json) and produces an aggregate error analysis, reusing the per-run classifier (classify_run):result_categorycounts (intercepted / infra_failure / zero-action / not-intercepted …)Self-contained, unit-tested on fixtures (5 tests); ruff + pyright clean; full suite green. Complements
clawbench-batch/clawbench-rescore.