How to install Claude Code skills

What is a Claude skill?

A skill (Anthropic calls them Agent Skills) is a folder with a SKILL.md file. The file starts with a short name and description, followed by instructions the agent follows when a task matches — plus optional scripts, templates, or reference files. The agent only reads the description up front and loads the rest when it needs it, so you can install dozens of skills without bloating every conversation.

Install skills with one command

The open-source skills CLI installs a whole repository, or one skill from it, into every agent it finds on your machine:

Add -g to install for your user instead of the current project, and -a claude-code to target a single agent. Every skill page on Skill Index has its exact command ready to copy — browse the directory.

Install a skill by hand

Clone the repository and copy the skill's folder into your skills directory:

Or use the Prompt tab on any skill page and paste it into your agent — it will fetch and install the skill for you.

Where Claude Code skills are stored

• Personal skills: ~/.claude/skills/<skill-name>/SKILL.md — available in every project.
• Project skills: .claude/skills/<skill-name>/SKILL.md inside a repository — commit them so your whole team gets them.
• Plugin skills: bundled inside installed Claude Code plugins.

The SKILL.md format

YAML frontmatter with a name and a description, then Markdown instructions. The description decides when the agent uses the skill, so say both what it does and when to use it:

---
name: release-notes
description: Writes release notes from merged pull requests. Use when the user asks for release notes, a changelog, or "what shipped".
---

# Release notes

1. List the pull requests merged since the last tag.
2. Group them into Features, Fixes, and Internal.
3. Write one plain-language line per change.

Keep the main file focused and move long reference material into separate files in the same folder.

Skills vs MCP vs subagents vs plugins

• Skills teach the agent how to do a task — instructions, examples, and scripts loaded on demand.
• MCP servers give the agent access to outside tools and data, such as a database, a browser, or an API.
• Subagents are separate workers with their own context that the main agent delegates a task to.
• Plugins package skills, commands, subagents, hooks, and MCP servers together so they install as one unit.

They combine well: a skill can explain how to use the tools an MCP server exposes.

Using skills in Codex, Cursor and Copilot

SKILL.md is an open format. Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, OpenCode and other agents can load the same skills. Each agent reads from its own folder; the skills CLI detects which agents you have and installs into the right place for each, so the same command works everywhere.

Skill not showing up? Checklist

1. The file is named exactly SKILL.md and sits directly inside the skill's own folder.
2. The folder is in a skills directory the agent reads (see where skills are stored).
3. The frontmatter is valid YAML between two --- lines, with both name and description.
4. The description clearly says when to use the skill — vague descriptions rarely trigger.
5. You started a new session after installing or editing it.

Suggest a repository

Know a good public skill collection we're missing? Email the.english.lab.factory@gmail.com with the GitHub link. We only list repositories with a license that allows redistribution.