skill-judge

To evaluate SKILL.md frontmatter validity, apply a structured evaluation framework that checks structural compliance, loading triggers, and frontmatter fields. This process identifies anti-patterns and produces a formal Skill Evaluation Report for dissemination.

A knowledge delta models the expert value in content by classifying sections as Expert, Activation, or Redundant. This classification identifies genuine expert knowledge, highlights token waste, and guides progressive disclosure for optimized SKILL.md design.

Audit skill packages for anti-patterns by applying an evaluation framework that scores eight dimensions totaling 120 points. This structured assessment checks frontmatter, loading triggers, and pattern compliance to generate standardized improvement recommendations.

Skill evaluation requires no external dependencies or components to function. The framework operates independently to assess SKILL.md frontmatter validity, structural checks, and pattern compliance without additional environment setup.

The best way to score skill quality is applying an eight-dimension evaluation framework totaling 120 points. This method assesses frontmatter validity, loading triggers, and knowledge delta modeling to produce a structured score and formal evaluation report.

Use progressive disclosure and layered loading in skill design when you need to optimize SKILL.md structure and reduce token waste. This approach is recommended during pre-publish review, content audits, and cross-skill comparison to ensure efficient activation triggers.