core-audit

Identify and report quality issues in Claude Skills.

Updated Jan 1, 2026
One-click install
npx skills add https://github.com/AirMile/claude-config --skill core-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: core-audit
Source: https://github.com/AirMile/claude-config/tree/main/skills/core-audit
Command: npx skills add https://github.com/AirMile/claude-config --skill core-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyze and refine Claude Skills for clarity, correctness, and effectiveness. Detects redundancy, dead paths, ambiguity, and poor structure. Optional internal walkthrough for deeper insights.

Core Features & Use Cases

  • Quick audits identify ambiguity, dead ends, and misalignment with intended pipelines.
  • Extended walkthroughs offer in-depth analysis of a target skill and suggested improvements.
  • Context-aware loading uses the triggering skill to tailor the analysis.

Quick Start

Run /core-audit to scan already-loaded skills, or specify a target skill like /core-audit dev-build to load and audit.

Frequently Asked Questions about core-audit

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

FAQPage Schema
What is a skill audit and how does it identify workflow quality issues?

A skill audit analyzes and refines Claude Skills to detect workflow quality issues like redundancy, dead paths, ambiguity, and poor structure. It evaluates the target skill's logic and outputs a structured analysis with prioritized findings and actionable refactors.

How do I run a quick audit to check skill clarity and effectiveness?

Run a quick audit by executing /core-audit to scan already-loaded skills, or specify a target like /core-audit dev-build. This performs a quick surface check to identify ambiguity, dead ends, and pipeline misalignment without requiring extended walkthroughs.

Can I perform an extended walkthrough to analyze a specific skill's structure?

Yes, you can perform an extended walkthrough for in-depth analysis by specifying a target skill. This context-aware loading tailors the evaluation to the triggering skill, offering deeper insights into structural issues and suggesting specific improvements for correctness.

Does the audit support quick surface checks for already-loaded skills?

Yes, the audit supports quick surface checks for already-loaded skills. Running the command without arguments scans the active environment, immediately reporting misalignment with intended pipelines and frontmatter health issues.

What are the limitations of automated skill structure analysis?

Automated skill structure analysis focuses on identifying redundancy, dead paths, ambiguity, and structural flaws. It outputs prioritized findings and actionable refactors, but deeper semantic correctness may require manual review despite the extended walkthrough insights.