Skillclaw Skill Evolution

If you're running multiple LLM agents and want them to get smarter from each other's experience, this intercepts OpenAI-compatible API calls, records session artifacts, and distills them into reusable SKILL.md files that sync across your agent cluster via OSS/S3. You run a local proxy that all your agents talk through, plus an evolve server (either a fixed three-stage pipeline or an autonomous OpenClaw agent) that watches sessions and writes skills. The architecture is straightforward: proxy records, storage syncs, evolve server learns. It's designed for multi-user setups where you want collective improvement without manually curating every skill, and it ships with WildClawBench integration for benchmarking the evolution loop.

skills.sh installs: 416
GitHub stars: 25
last updated: Jun 3, 2026

View on GitHub

Install

$ npx -y skills add aradotso/trending-skills --skill skillclaw-skill-evolution --agent claude-code

Installs into .claude/skills of the current project.

Files

SkillClaw: Collective Skill Evolution for LLM Agents

Skill by ara.so — Daily 2026 Skills collection.

SkillClaw is a framework that makes LLM agents progressively smarter by evolving reusable skills from real session data and sharing them across a group of agents. It intercepts OpenAI-compatible API calls via a local proxy, records session artifacts, and runs an evolve server that distills experience into SKILL.md files synced via cloud storage (OSS/S3/local).

Architecture

User → OpenClaw Agent → SkillClaw Client Proxy → Upstream LLM API
                              ↓ records sessions
                        Shared Storage (OSS/S3/local)
                              ↑ reads sessions, writes skills
                        Evolve Server (workflow or agent)

Three components share the same storage layer and skill format:

SKILL.md on GitHub

Details

Repository: github.com/aradotso/trending-skills
First seen: Jun 3, 2026