What problem does it solve?
This Skill helps you reliably review Claude Code hook scripts and hook-related settings so you can catch configuration drift, unsafe command paths, and hook logic errors before they impact commit, push, or PR workflows.
Core Features & Use Cases
- Systematic hook audit for safety and determinism: Reviews PreToolUse/PostToolUse/SessionStart hooks and verifies matcher/if dispatch coverage.
- Correct path resolution enforcement: Ensures each
command resolves to a stable absolute path using $CLAUDE_PROJECT_DIR, ${CLAUDE_PLUGIN_ROOT}, or an appropriate user-level prefix rather than brittle relative paths.
- Fail-closed parsing guidance for gate hooks: Validates that hook scripts use the canonical JSON parsing pattern and deny on malformed or missing tool-input data.
- Operational-footprint hardening checklist: Highlights latency and timeout risks for external commands and calls out potential indefinite hangs.
Quick Start
Ask your AI to review your Claude Code hook script .claude/hooks/<name>.sh and your .claude/settings.json hook entries for correct dispatch matching, fail-closed parsing, and safe command path resolution.