skill-improve

Identify skill gaps and apply test-fix-retest loops with keep or revert decisions.

3|Updated Jun 3, 2026
One-click install
npx skills add https://github.com/Negentropy-Laby/Constitution-Driven-Development --skill skill-improve-negentropy-laby
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-improve
Source: https://github.com/Negentropy-Laby/Constitution-Driven-Development/tree/main/.agents/skills/skill-improve
Command: npx skills add https://github.com/Negentropy-Laby/Constitution-Driven-Development --skill skill-improve-negentropy-laby

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Improve a skill by automating a multi-phase test-fix-retest loop that detects gaps, proposes targeted fixes, rewrites the skill, and re-validates results to decide keep/revert based on score change.

Core Features & Use Cases

  • Static checks identify gaps in the skill's frontmatter, structure, and guidance.
  • Proposes targeted patches, applies rewrites, and re-runs tests to measure impact.
  • Maintains an audit trail and supports keep/revert decisions based on score change.
  • Useful for refining Claude Code-based skills in a standardized, repeatable workflow.

Quick Start

Run the command /skill-improve [skill-name] to start the improvement loop for a specific skill.

Frequently Asked Questions about skill-improve

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

FAQPage Schema
How do I automate skill improvement for Claude Code with a test-fix-retest loop?

Automate skill improvement by running a multi-phase loop that uses static checks to identify gaps, proposes targeted patches, rewrites the skill, and re-runs tests to measure impact. Use the command /skill-improve [skill-name] to start the workflow.

What is the best way to patch gaps in a skill and decide whether to keep or revert changes?

The best way to patch skill gaps is applying proposed fixes, re-running tests, and evaluating score improvements to decide keep/revert outcomes. This automated loop maintains an audit trail and versioned history to document results and guide iterations.

Can I use static checks to identify gaps in my skill's frontmatter and structure?

Yes, static checks identify gaps in your skill's frontmatter, structure, and guidance. The improvement loop leverages these checks to propose targeted patches, applies the rewrites, and validates results to ensure score improvements.

How does the skill testing loop document outcomes and guide future iterations?

The testing loop documents outcomes by leveraging memory_bank, skill_testing catalogs, and versioned history. It maintains an audit trail of applied patches and keep/revert decisions to guide future skill iterations.

Do I need any dependencies to run the automated skill improvement workflow?

No external dependencies are required to run the automated skill improvement workflow. The Skill operates independently to perform static checks, apply patches, and execute retests for refining Claude Code-based skills in a standardized workflow.