skill-evolution-manager

Record runtime errors and improvements into evolution.json and SKILL.md.

2|Updated May 13, 2026
One-click install
npx skills add https://github.com/danger-English/claude-code-intro --skill skill-evolution-manager-danger-english
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-evolution-manager
Source: https://github.com/danger-English/claude-code-intro/tree/main/skills/skill-evolution-manager
Command: npx skills add https://github.com/danger-English/claude-code-intro --skill skill-evolution-manager-danger-english

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Skill evolution manager records and evolves skill performance by capturing runtime errors and improvement points, stores them in evolution.json, and stitches them back into SKILL.md for continual optimization.

Core Features & Use Cases

  • Records errors, improvements, and boundary cases during skill usage to build a growing knowledge base.
  • Saves experiences to ~/.claude/skills/<skill-name>/evolution.json and later merges them into SKILL.md under an "进化经验" section.
  • Enables iterative refinement by reusing past learnings to enhance future executions.

Quick Start

Record an error or improvement during a skill run and save it to evolution.json.

Frequently Asked Questions about skill-evolution-manager

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

FAQPage Schema
How do I record runtime errors and improvement points during a skill execution?

You can record runtime errors and improvement points during skill execution by capturing them interactively and saving the data directly into a local evolution.json file for iterative tracking.

How does logging evolution data in JSON help improve skill performance?

Logging evolution data in JSON captures errors and boundary cases, building a knowledge base that can be stitched back into SKILL.md for continual optimization and iterative refinement.

What is the best way to automate skill improvement from past execution logs?

The best way to automate skill improvement is by merging recorded runtime experiences from evolution.json back into the SKILL.md file under a dedicated evolution section for future reference.

Do I need a specific file path to start logging skill evolution data?

Yes, you need a local evolution.json file located at ~/.claude/skills/<skill-name>/evolution.json to store runtime logs and support patching SKILL.md with improvement feedback.

Can I patch SKILL.md with new evolution feedback without breaking existing content?

Yes, evolution feedback is patched into SKILL.md by appending it specifically under the '进化经验' section, ensuring existing content remains intact while adding new iterative learnings.

Why does my skill keep repeating the same runtime errors across sessions?

Skills repeat runtime errors across sessions when past issues are not logged. Saving errors to evolution.json and stitching them into SKILL.md enables reusing past learnings to prevent recurrence.