AMAP-ML/SkillClaw

Registry

Summary

SkillClaw is an open-source framework that enables AI agents to evolve their skills collectively through an Agentic Evolver mechanism. Built on Hermes and OpenClaw, it supports continual learning and self-evolving skill acquisition for LLM-based agents.

Key features

• Agentic Evolver for collective skill evolution
• Continual learning and self-evolving capabilities
• Integration with Hermes and OpenClaw ecosystems
• Skill acquisition and management for LLM agents
• Collective intelligence through agent collaboration

Use cases

• Building self-improving AI agents
• Multi-agent systems with shared skill libraries
• Continual learning in dynamic environments
• Automated skill discovery and evolution
• Research in collective intelligence and agentic AI

README excerpt

<div align="center"> <img src="assets/skillclaw_logo.png" alt="SkillClaw" width="150"> # ✨ SkillClaw: Let Skills Evolve Collectively with Agentic Evolver ✨ <h3>AI agent skills that evolve from every real interaction — just talk.<br>Across sessions, agents, devices, and users. Experience compounds. Skills keep growing.</h3> | 🚀 Quick Install | 💬 Just Chat | 🔌 Broad Compatibility | 🧬 Collective Skill Evolution | |:-:|:-:|:-:|:-:| [](https://github.com/NousResearch/hermes-agent) [](https://github.com/openclaw/openclaw) [](https://github.com/AMAP-ML/SkillClaw) [](https://www.python.org/) [](./LICENSE) [](https://arxiv.org/abs/2604.08377) [](https://arxiv.org/pdf/2604.08377) [](https://huggingface.co/papers/2604.08377) [![WeChat](https://img.shields.io/badge/WeChat-Group-07C160?style=flat-sq

Topics