What problem does it solve?
Provide an AI-executor-focused audit of Claude Code plugin artifacts to find logic bugs, trigger failures, and execution feasibility issues that impede reliable runtime behavior and orchestration.
Core Features & Use Cases
- Detects scope from user input (specific files, entire plugin, recent git changes, or full audit) and lists files for confirmation.
- Chooses Strategy A (parallel plugin-dev validators + internal reviewer) if plugin-dev is available, or Strategy B (internal reviewer with supporting docs) otherwise, and merges or presents results accordingly.
- Produces a unified report across nine review dimensions with grouped severities, file:line references, and actionable fixes for Bug-level findings; supports applying fixes on approval.
- Use Case: Run a focused review after modifying skills or agents to catch dispatch loops, trigger conflicts, and execution edge cases before release.
Quick Start
Ask the plugin-review skill to audit my plugin directory "my-plugin" for logic bugs, trigger issues, and execution feasibility.