Multi-agent iterative code review
audit-projectpluginsetup L3★1,721
ComposioHQ/awesome-claude-plugins ↗What it does
Run iterative code review cycles until zero issues
Best for
Enforcing zero-defect standards on medium codebases where iterative polish beats single-pass review.
Inputs
- · project source code
- · framework context (optional)
Outputs
- · list of reviewed issues
- · remediation status log
- · final zero-issue report
Requires
- · Claude (multi-agent orchestration)
- · git (for diffs)
Preconditions
- · project must be readable in Claude context
- · framework type helps but not required
Failure modes
- · context window exhaustion on large codebases
- · agent loop timeout if issues self-regenerate
- · false negatives if review criteria ambiguous
Trust signals
- · MIT licensed, semantic versioning 3.7.2
- · author has GitHub presence (avifenesh)
- · tagged with multi-agent + framework-specific keywords