This document provides guidance for AI agents working with the skills repository.
This repository contains shareable Agent Skills that can be installed across multiple AI coding agents using the npx skills CLI.
Skills in this repository can be installed using the vercel-labs/skills CLI tool.
# List available skills
npx skills add ronniegeraghty/skills --list
# Install specific skills
npx skills add ronniegeraghty/skills --skill azure-sdk-mcp-adoption
# Install to specific agents
npx skills add ronniegeraghty/skills -a github-copilot -a claude-code
# Install globally (available across all projects)
npx skills add ronniegeraghty/skills -g
# Non-interactive CI/CD installation
npx skills add ronniegeraghty/skills --skill azure-sdk-mcp-adoption -g -a github-copilot -y# List installed skills
npx skills list
# Search for skills
npx skills find azure
# Check for updates
npx skills check
# Update installed skills
npx skills update
# Remove skills
npx skills remove azure-sdk-mcp-adoption
# Create a new skill template
npx skills init my-new-skillSkills can be installed to 35+ agents including:
- GitHub Copilot (
.github/skills/) - Claude Code (
.claude/skills/) - Cursor (
.cursor/skills/) - Codex (
.codex/skills/) - And many more...
See vercel-labs/skills README for the full list.
skills/
├── .claude-plugin/
│ └── marketplace.json # Claude Code plugin marketplace config
├── skills/
│ ├── azure-sdk-mcp-adoption/
│ │ ├── SKILL.md # Required: skill instructions
│ │ ├── package.json # Dependencies for scripts
│ │ └── scripts/ # Executable pipeline scripts
│ ├── copilot-pr-analysis/
│ │ ├── SKILL.md
│ │ ├── package.json
│ │ └── scripts/
│ └── template/
│ └── SKILL.md
├── package.json # Workspace configuration
├── pnpm-workspace.yaml # pnpm workspace packages
└── README.md
- Copy
skills/template/toskills/<skill-name>/ - Edit
SKILL.mdfrontmatter withnameanddescription - Add instructions in the Markdown body
- Optionally add
scripts/,references/, orassets/directories
---
name: my-skill
description: Clear description of what the skill does and when to use it.
---
# My Skill
Instructions for the agent...name- Lowercase alphanumeric with hyphens, max 64 chars, must match directory namedescription- Max 1024 chars, should include WHEN to use the skill
| Directory | Purpose |
|---|---|
scripts/ |
Executable code (Python, Bash, Node.js) |
references/ |
Documentation loaded into context as needed |
assets/ |
Templates, images, data files for output |
Skills with executable scripts should:
- Use
scripts/directory (notsrc/) - Include a
package.jsonwith"type": "module"for ESM - Make scripts executable with shebang (
#!/usr/bin/env node) - Include a main entry point (e.g.,
run.js) - Document usage in SKILL.md with examples
- Handle errors gracefully with clear messages
- Agent Skills Specification
- Skills CLI (npx skills)
- Skills Directory (skills.sh)
- Anthropic Skills Repository
MIT