Skip to content

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Agent Skills

A minimal implementation of agent skills - a lightweight, open format for extending AI agents with specialized capabilities.

Written by Aman, Claude, and Cursor

What Are Agent Skills?

Agent skills are markdown files (SKILL.md) that give AI agents specialized knowledge and workflows. Each skill contains:

  • YAML frontmatter: Name, description, metadata
  • Markdown instructions: Detailed guidelines for the agent

When a user's task matches a skill, the agent automatically activates it to provide expert assistance.

Quick Start

# Install
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

# Configure
cp .env.example .env
# Add your OPENAI_API_KEY to .env

# Run
python -m agent.chat

Or use the run script:

./run.sh

Included Skills

  • code-review: Reviews code for bugs, security issues, and best practices
  • git-helper: Git workflows, commit messages, and troubleshooting
  • file-organizer: Intelligent file organization strategies
  • api-tester: Makes actual HTTP requests to APIs and analyzes responses (GET, POST, PUT, PATCH, DELETE)

How It Works

  1. Discovery: Scans skills/ directory for SKILL.md files
  2. Loading: Parses YAML frontmatter (name, description)
  3. Activation: When user's task matches, loads full skill instructions
  4. Execution: Agent follows skill guidelines to complete task

Creating Your Own Skill

mkdir -p skills/my-skill

Create skills/my-skill/SKILL.md:

---
name: my-skill
description: What this skill does
version: 1.0.0
---

# My Skill

You are an expert at [skill domain].

## Guidelines

1. Step one
2. Step two
3. Step three

## Examples

Show examples of using this skill effectively.

That's it! The skill will be automatically discovered.

Project Structure

agent-skills/
├── agent/
│   ├── skills_manager.py    # Skill discovery & loading
│   ├── llm_client.py         # OpenAI client
│   ├── chat.py               # Terminal interface
│   └── code_executor.py      # Code execution for skills
├── skills/                   # Skills directory
│   ├── code-review/
│   ├── git-helper/
│   ├── file-organizer/
│   └── api-tester/
├── requirements.txt
└── run.sh

Resources

Learn More

Tutorial: Building Agent Skills from Scratch - Detailed guide on how this implementation works

Repository: github.com/onlyoneaman/agent-skills

License

MIT

About

A minimal implementation of agent skills - a lightweight, open format for extending AI agents with specialized capabilities.

Resources

Stars

10 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages