Skip to content

chore: add adversarial-review-loop skill - #12

Merged
tmack8001 merged 1 commit into
mainfrom
add-adversarial-review-loop-skill
Oct 22, 2026
Merged

tmack8001 merged 1 commit into
mainfrom
add-adversarial-review-loop-skill

Conversation

@tmack8001

@tmack8001 tmack8001 commented Oct 22, 2026 •

Copy link
Copy Markdown
Owner

Pull Request

Description

Adds a Claude Code skill, adversarial-review-loop, under .claude/skills/. It describes a method for running iterative adversarial code reviews to consensus before merging risky work, including how to scope each round, how to treat a reviewer's proposed fix, when the loop is finished, and a mutation-testing step to check test adequacy. This is a tooling/documentation addition with no change to the CLI source or its behavior.

Type of Change

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds functionality)
  • Breaking change (fix or feature that would cause existing functionality to not work as expected)
  • Documentation update
  • Code refactoring
  • Performance improvement
  • Test addition or update

Changes Made

  • Add .claude/skills/adversarial-review-loop/SKILL.md: the skill itself (the loop, how to instruct a reviewer, verifying fixes, convergence criteria, mutation testing, common traps, and a pre-flight checklist). It is covered by the repository's MIT license; the skill has no separate LICENSE file.

Platform(s) Affected

  • BCOEM
  • Reggie
  • BAP
  • All platforms
  • Core functionality
  • Documentation only

Testing

No tests apply: this change adds only a skill definition (Markdown), with no executable code. No tests were run.

  • Existing tests pass (npm test)
  • Added new tests for new functionality
  • Manually tested with real competition URLs
  • Tested with config files
  • Tested both JSON and CSV output

Test Commands

N/A

Test URLs

N/A

Checklist

  • My code follows the style guidelines of this project
  • I have performed a self-review of my own code
  • I have commented my code, particularly in hard-to-understand areas
  • I have made corresponding changes to the documentation
  • My changes generate no new warnings
  • I have added tests that prove my fix is effective or that my feature works
  • New and existing unit tests pass locally with my changes
  • Any dependent changes have been merged and published

Documentation

  • Updated README.md (if needed)
  • Updated relevant docs in docs/ folder
  • Updated CHANGELOG.md
  • Added JSDoc comments to new functions/classes

Screenshots (if applicable)

N/A

Additional Notes

The skill is under the repository's MIT license, the same as the rest of the repo.


Generated by Claude Code

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01W9Y5xJdDbjzZFbNfVFKS5b
@tmack8001
tmack8001 force-pushed the add-adversarial-review-loop-skill branch from f98db62 to b031e82 Compare October 22, 2026 17:05
@tmack8001
tmack8001 merged commit 2ef6cd5 into main Oct 22, 2026
10 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant