Thomaszhou22

Self Discover Skill — AI skill for Claude Code

AI community

SELF-DISCOVER: AI agents self-compose reasoning structures.

How to install Self Discover Skill

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Thomaszhou22/self-discover-skill and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Self Discover Skill does

SELF-DISCOVER: AI agents self-compose reasoning structures. Based on Zhou et al. (2024), NeurIPS.

Alternatives in AI

WeKnora — Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, an 26.9k ★
Deepreasoning — A high-performance LLM inference API and Chat UI that integrates DeepSeek R1's CoT reasoning traces with Anthr 5.4k ★
Discover — Discovery phase - Multi-AI research and exploration 2.8k ★

README

🔍 Self-Discover Skill

**Self-composing reasoning structures for AI agents.**

*Let AI discover the optimal reasoning strategy — before solving the problem.*

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE) [![OpenClaw](https://img.shields.io/badge/OpenClaw-compatible-blue)](https://openclaw.ai) [![Claude Code](https://img.shields.io/badge/Claude%20Code-compatible-purple)](https://claude.ai) [![Cursor](https://img.shields.io/badge/Cursor-compatible-orange)](https://cursor.sh) [![GitHub Copilot](https://img.shields.io/badge/Copilot-compatible-lightblue)](https://github.com/features/copilot) [![Codex CLI](https://img.shields.io/badge/Codex%20CLI-compatible-green)](https://github.com/openai/codex) [![Gemini CLI](https://img.shields.io/badge/Gemini%20CLI-compatible-4285F4)](https://ai.google.dev) [![Windsurf](https://img.shields.io/badge/Windsurf-compatible-00D4AA)](https://codeium.com/windsurf) [![JetBrains](https://img.shields.io/badge/JetBrains-compatible-000)](https://www.jetbrains.com/ai/) [![Benchmarked](https://img.shields.io/badge/benchmarked-25%25%20improvement-green)](BENCHMARK.md)

What It Does

Most AI reasoning uses a single strategy (usually "think step by step"). This skill lets the AI **discover the optimal reasoning structure** for each specific task:

SELECT → ADAPT → IMPLEMENT → SOLVE

Based on peer-reviewed research showing **up to 32% improvement** over Chain-of-Thought and **>20% over Self-Consistency** with 10-40x less compute (Zhou et al., 2024, ICML).

**One file. Zero dependencies. Zero API cost.** The host AI composes its own reasoning structures.