self-evolving-skill

Automate skill evolution using predictive coding and value-driven mechanisms.

2|Updated Feb 27, 2026
One-click install
npx skills add https://github.com/jiyangnan/xiaonangua-openclaw-skills --skill self-evolving-skill-jiyangnan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: self-evolving-skill
Source: https://github.com/jiyangnan/xiaonangua-openclaw-skills/tree/main/skills/personality/self-evolving-skill
Command: npx skills add https://github.com/jiyangnan/xiaonangua-openclaw-skills --skill self-evolving-skill-jiyangnan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python, typescript, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the problem of manually evolving skills and automates the process through meta-cognitive mechanisms.

Core Features & Use Cases

  • ResidualPyramid金字塔分解: Quantifies cognitive gaps.
  • 自适应反思触发: Automatically determines when learning is needed based on residual energy.
  • 经验回放: Caches learned patterns to reduce redundant triggers.
  • 价值门控: Accepts variations only if they improve long-term value.
  • 持久化: Automatically saves and loads experiences.
  • MCP工具: Provides a suite of tools for creating, executing, and analyzing skills.

Quick Start

Start the MCP server to enable skill management and analysis.

cd ~/.openclaw/skills/self-evolving-skill
./run_mcp.sh

Frequently Asked Questions about self-evolving-skill

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

FAQPage Schema
How do I automate skill evolution using meta-cognitive mechanisms?

Automating skill evolution requires a meta-cognitive system that uses predictive coding and value-driven mechanisms to quantify cognitive gaps and accept variations only when they improve long-term value.

What is predictive coding for self-evolving skill learning?

Predictive coding in self-evolving systems quantifies cognitive gaps through ResidualPyramid decomposition, automatically triggering adaptive reflections based on residual energy to determine when learning is needed.

Do I need Python and TypeScript to run a meta-cognitive self-learning system?

Yes, you need Python for core functionality and TypeScript for the SDK to run this meta-cognitive self-learning system, which also requires starting an MCP server to enable skill management and analysis.

How does value gating work in automated skill evolution?

Value gating in automated skill evolution evaluates and accepts variations only if they demonstrably improve long-term value, preventing unnecessary updates while caching learned patterns via experience replay to reduce redundant triggers.

Can I use MCP tools to create and analyze skills in OpenClaw?

Yes, this system provides a suite of MCP tools compatible with OpenClaw for creating, executing, and analyzing skills, while automatically saving and loading experiences through built-in persistence mechanisms.