Getting started¶

This guide walks through the end-to-end flow without relying on a TTY or interactive prompts.

Prerequisites¶

• Python 3.10+ and uv (or pip)
• A project directory with a pyproject.toml (for dependency scanning)

1. Install skilly¶

uvx skilly --version

2. Create a skill¶

skilly create deployment-checks \
  --description "Validate deployment readiness before a production release." \
  --instructions "# Instructions\n\nRun the deployment checklist." \
  --yes

This creates deployment-checks/SKILL.md in the default destination. Use --directory to choose a different location.

3. See what's installed¶

skilly list

If you created the skill with a destination flag (e.g. --local), use the same flag when listing.

4. Install skills from a dependency scan¶

skilly scan

5. Install from GitHub¶

skilly download https://github.com/example/skills-repo --all

6. Check for updates¶

skilly update

This checks every installed GitHub- and SkillsMP-backed skill for newer versions. Pass --yes to apply all discovered updates without prompting.

Next steps¶

• Installation options — choose your package manager.
• Core concepts — skills, resources, origins, statuses.
• CLI reference — full command documentation.
• Python API — programmatic usage.