Repository navigation
docs: README leads with the real-corpus numbers - #17
Merged
Merged
Conversation
Measured now opens with the Stage 2 corpus runs (agent mode): quiet rate 100% / 86%, hunk recall 1/15 / 3/15, rule recall 0/15, gate not ready, with their caveats. The Stage 1 toy suite (precision 0.17, recall 0.92) stays below it, framed as the opposite failure on hand-written snippets. Presets note that focused cannot catch the corpus's R4 defects; roadmap items 5, 8, 9 and 11 reflect what is done and what is next. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01WkPsqEJ4QYWRW5eCFBYRPS
aisona-lab
marked this pull request as ready for review
October 23, 2026 09:15
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
README only. The Measured section used to show only the Stage 1 toy suite (precision 0.17 on 13 hand-written snippets).
Measured now opens with the two Stage 2 corpus runs from #12 and #13:
It also says what the runs suggest: the reviewer is quiet on real code but blind to most defects, most defects aren't visible from a hunk, and the rules overlap. The caveats are stated too: soft isolation, a small corpus with provisional clean labels, and an agent as reviewer rather than the API path. The toy suite stays below, framed as the opposite failure.
Smaller updates:
focuseddoesn't ask R4, which is 8 of the corpus's 15 labelled defects.Every number was cross-checked against
corpus/seed.jsonlanddocs/eval-runs/2026-10-23-agent-mode-corpus*.txt.Generated by Claude Code