docs: add explicit Threat Model section to README#9
Merged
Conversation
Six enumerated failure modes mirroring the companion paper (agent-closeout-bench paper/uai-2026/main.tex §Threat Model and Limitations): 1. Lexical evasion 2. Hook misconfiguration 3. Runtime bypass 4. In-band manipulation 5. Evidence-marker limitations 6. Coverage and language scope Reviewer-grade limitations were previously documented in PR descriptions and the companion benchmark; this surfaces them at the top of the README so operators considering the suite for safety-critical workflows see them before relying on it. Cross-references the companion paper's benchmark (agent-closeout-bench) as the surface where lexical brittleness is measured rather than hidden. Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
paper/uai-2026/main.tex§Threat Model and Limitations).Why now
Reviewer-grade limitations were previously documented in PR descriptions and in the companion benchmark repo, but never surfaced at the top of this README. Operators considering this suite for safety-critical workflows should see them before relying on it.
Threat model items added
Test plan
agent-closeout-bench.🤖 Generated with Claude Code