skills.rest
jcdendrite
jcdendriteCOMMUNITY
Explore all skills by this author →
GitHub

skill-review

Audit Claude Code SKILL.md files for trigger accuracy and frontmatter consistency.

4|1|Updated Mar 11, 2026
Search Tags:#code review#Claude Code#YAML frontmatter#skill audit#trigger design#misfire prevention#behavior equivalence
One-click install
npx skills add https://github.com/jcdendrite/claude-config --skill skill-review-jcdendrite
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-review
Source: https://github.com/jcdendrite/claude-config/tree/main/plugins/skill-management/skills/skill-review
Command: npx skills add https://github.com/jcdendrite/claude-config --skill skill-review-jcdendrite

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Prevents Claude Code skills from misfiring or drifting out of spec by auditing each SKILL.md for frontmatter quality, trigger accuracy, voice/structure consistency, and behavior-preserving edits.

Core Features & Use Cases

  • Frontmatter validation: Confirms name and description exist and that the description contains properly structured TRIGGER when: and DO NOT TRIGGER when: blocks.
  • Trigger reliability review: Checks that triggers are specific (e.g., globs and action verbs) and that DO NOT TRIGGER when: blocks name overlapping adjacent skill surfaces to prevent dual-fires.
  • Style, length, and operational correctness: Enforces imperative second-person voice, structured checklist formatting, and keeps bodies within length targets.
  • Change safety checks: Requires a behavioral-equivalence audit for compression/removal diffs and validates platform-genericness (keeps engine tokens out of project skills).
  • Audit completion marker: Guides recording completion via a repo-specific marker command when the review is clean.

Quick Start

Review the skill at claude/.claude/skills/your-skill-name/SKILL.md against the checklist to ensure its triggers and exclusions match its actual behavior.

Frequently Asked Questions about skill-review

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I prevent Claude Code skills from misfiring when triggers overlap?▼

To prevent Claude Code skills from misfiring, you can audit the SKILL.md file to ensure that the DO NOT TRIGGER when: block explicitly names adjacent skill surfaces, preventing dual-fires and trigger overlaps.

How do I validate YAML frontmatter in Claude Code skill files?▼

Validating YAML frontmatter in Claude Code skill files involves checking that the required name and description fields exist, and that the description contains properly structured TRIGGER when: and DO NOT TRIGGER when: blocks.

What is behavior-preserving compression for skill audits?▼

Behavior-preserving compression for skill audits is a change safety check that requires a behavioral-equivalence audit for any compression or removal diffs, ensuring edits do not alter the original skill behavior.

How do I review trigger accuracy across a growing set of Claude Code skills?▼

Reviewing trigger accuracy across a growing set of Claude Code skills involves auditing each SKILL.md file to confirm triggers use specific action verbs and globs while keeping descriptions consistent with actual skill behavior.

What are the structural requirements for a Claude Code SKILL.md file?▼

Structural requirements for a Claude Code SKILL.md file include using imperative second-person voice, structured checklist formatting, keeping bodies within length targets, and maintaining platform-genericness to keep engine tokens out.

When should I not use automated skill audits for Claude Code files?▼

You should not rely on automated skill audits when you need to record completion via a repo-specific marker command, as this requires the review to be completely clean and validated against all structural and trigger rules first.

Related Skills

Explore More Software Engineering Skills→
mkgs-databricks-demosmkgs-databricks-demos
1

databricks-mlflow-evaluation

Locate root SKILL.md files and extract YAML frontmatter metadata into YAML documents.

Official
Advanced
JunHyuk-KangJunHyuk-Kang

orchestrator-developer

Create files, apply design tokens, run builds, and report completion.

Community
Advanced
iamasters-academyiamasters-academy
20

tool-output-verifier

Validate text outputs for brand consistency, readability, length, and factual accuracy.

Official
Advanced
mistakeknotmistakeknot

runtime-diagnostics

Scaffold a diagnostic HTTP server and CLI for Tauri and Electron apps.

Community
Advanced