Let ClawSweeper judge real behavior proof#48
Merged
pashpashpash merged 2 commits intomainfrom May 5, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
ClawSweeper already records a structured
realBehaviorProofjudgement and uses it to block pass/automerge markers, but it did not yet own the positive label that tells maintainers the evidence was convincing.This makes the proof review explicitly agent-led. The review prompt tells Codex to inspect PR bodies, comments, screenshots, videos, logs, terminal output, and links with its own tools and best judgement. Review runs now give Codex a scratch directory plus a separate read-only inspection token, while the deterministic wrapper keeps the write token for comments and labels. During apply, ClawSweeper syncs
proof: sufficientwhen the structured judgement is sufficient and removes it when the evidence is missing, weak, mock-only, or no longer applicable.