Skip to main content

1. Create your library

This only needs to be done once.

2. Install some skills

Pull in open source skills from GitHub:
Or create your own by adding a folder with a SKILL.md file to your library’s skills/ directory.

3. Add skills to a project

From inside a project directory, run reseed to browse your library interactively, or specify skills by name:
This copies skills into .agents/skills/ (or the directory specified with --dir) where your AI agent can find them. Commit them to git and your whole team benefits.

4. Keep things up to date

When upstream skills change, install the latest and sync into your projects: