skill-judge

Score SKILL.md files across eight dimensions and generate a structured evaluation report.

Updated Feb 2, 2026
One-click install
npx skills add https://github.com/irwins/platform-agent-kit --skill skill-judge-irwins
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-judge
Source: https://github.com/irwins/platform-agent-kit/tree/main/.github/skills/skill-judge
Command: npx skills add https://github.com/irwins/platform-agent-kit --skill skill-judge-irwins

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a rigorous framework to evaluate Skill units (SKILL.md) and generate a structured, actionable report that helps ensure quality, safety, and consistency across Skill packages.

Core Features & Use Cases

  • Dimensional scoring across eight evaluation pillars (Knowledge Delta, Mindset, Anti-Pattern Quality, Specification Compliance, Progressive Disclosure, Freedom Calibration, Pattern Recognition, and Practical Usability) to quantify Skill quality.
  • Pattern-aware analysis that identifies common failure modes (Tutorial, Dump, Orphan References, etc.) and prescribes concrete improvements.
  • Output: a complete Skill Evaluation Report with total score, grade, loading discipline, pattern alignment, knowledge ratio, and prioritized improvements, suitable for product-quality reviews, audits, and training.

Quick Start

Evaluate the provided SKILL.md file to generate a structured evaluation report.

Frequently Asked Questions about skill-judge

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

FAQPage Schema
How do I evaluate SKILL.md documentation quality and score it?

To evaluate SKILL.md documentation quality, score the file across eight dimensions including frontmatter validity, content structure, and loading triggers to produce a structured report with a 0-120 point score.

What is progressive disclosure in skill evaluation and how is it measured?

Progressive disclosure in skill evaluation measures how effectively a SKILL.md file layers information for loading triggers. It is scored as one of eight dimensions to ensure optimal knowledge delivery without overwhelming the system.

How do I identify anti-patterns in SKILL.md files?

Identify anti-patterns in SKILL.md files through pattern-aware analysis that detects common failure modes like Tutorial, Dump, and Orphan References, then prescribes concrete improvements within the evaluation report.

Can I assess skill loading triggers and frontmatter validity without external dependencies?

Yes, you can assess loading triggers and frontmatter validity without external dependencies. The evaluation requires only the SKILL.md file, referenced files in references/, and optional resources to inform the scoring rubric.

What is the best way to audit skill units for specification compliance?

The best way to audit skill units for specification compliance is applying a fixed rubric across eight evaluation pillars, generating a report with a total score, grade, and prioritized improvements for quality reviews.

Why does my skill evaluation report show a low knowledge ratio score?

A low knowledge ratio score in skill evaluation indicates insufficient knowledge delta between the skill and baseline understanding. The report diagnoses this metric alongside mindset and practical usability to recommend targeted enhancements.