agent-sync
One tool to rule them all: Sync, Centralize, and Share AI Agent configurations and skills.
agent-sync solves the fragmentation of the AI agent ecosystem by providing a unified workflow for your CLI tools.
Features
- Unified Skills Hub - Stop duplicating skills across different agents. Centralize everything in
~/.agents/skills/ - Private Backup & Sync - Keep your agent configurations safely backed up in a private GitHub repository
- Share with the World - Effortlessly publish your best custom skills to a public repository
- Extension Support - Automatically detects and backs up skills from agent extensions
Supported Agents
| Agent | Config Files | Skills Path |
|---|---|---|
| claude-code | settings.json |
~/.claude/commands/ |
| gemini-cli | settings.json |
~/.gemini/tools/ |
| opencode | opencode.json |
~/.config/opencode/skills/ |
| pi.dev | settings.json |
~/.pi/agent/skills/ |
| qwen-code | settings.json |
~/.qwen/skills/ |
Quick Start
# Install
pipx install agent-sync
# First machine - create repo
agent-sync init
agent-sync push
# Other machines - link to existing repo
agent-sync link https://github.com/user/repo.git
agent-sync pull
Protocol Support
agent-sync supports two protocols:
- DotAgents Protocol (default) - Vendor-neutral skills hub
- GitAgent Protocol (opt-in) - Comprehensive agent definitions
Learn more about DotAgents Protocol
License
MIT License