agentskills

MCP.Pizza Chef: agentskills

This is the home for the Agent Skills format — an open specification behind reusable instruction packs that many AI apps can load automatically. Instead of retyping the same steps into a chat window every time, you can bundle instructions, reference notes, and templates into a folder, and a compatible assistant pulls it in the moment it recognizes a matching task. The site includes guides, the full format details, and example packs to browse. No account or key is required to read it, though building your own pack takes some technical setup.

Coding
Writing

Use This MCP server To

Browse example skills to see what's possible before building my own Learn the folder format for packaging reusable instructions Check which AI apps already support Agent Skills Read step-by-step guides for writing my first skill Find community-made skills and share mine on Discord

README

Agent Skills

Discord

A standardized way to give AI agents new capabilities and expertise.

What are Agent Skills?

Agent Skills are a lightweight, open format for extending AI agent capabilities with specialized knowledge and workflows.

At its core, a skill is a folder containing a SKILL.md file. This file includes metadata (name and description, at minimum) and instructions that tell an agent how to perform a specific task. Skills can also bundle scripts, reference materials, templates, and other resources.

my-skill/
├── SKILL.md          # Required: metadata + instructions
├── scripts/          # Optional: executable code
├── references/       # Optional: documentation
├── assets/           # Optional: templates, resources
└── ...               # Any additional files or directories

Why Agent Skills?

Agents are increasingly capable, but often don't have the context they need to do real work reliably. Skills solve this by packaging procedural knowledge and company-, team-, and user-specific context into portable, version-controlled folders that agents load on demand. This gives agents:

  • Domain expertise: Capture specialized knowledge — from legal review processes to data analysis pipelines to presentation formatting — as reusable instructions and resources.
  • Repeatable workflows: Turn multi-step tasks into consistent, auditable procedures.
  • Cross-product reuse: Build a skill once and use it across any skills-compatible agent.

How do Agent Skills work?

Agents load skills through progressive disclosure, in three stages:

  1. Discovery: At startup, agents load only the name and description of each available skill, just enough to know when it might be relevant.

  2. Activation: When a task matches a skill's description, the agent reads the full SKILL.md instructions into context.

  3. Execution: The agent follows the instructions, optionally executing bundled code or loading referenced files as needed.

Full instructions load only when a task calls for them, so agents can keep many skills on hand with only a small context footprint.

Where can I use Agent Skills?

agentskills FAQ

Can I use this to make my AI assistant follow the same steps every time?
Yes — you package the steps into a skill folder using the format described here, and a compatible assistant pulls it in automatically when a matching task comes up.
Can I use this to see real skill examples before building my own?
Yes — the Example Skills page links to a public collection you can browse for inspiration.
Do I need an account or an API key?
No sign-up or key is needed to read the documentation or browse examples.
Which AI apps can use these skill packs?
A large number of AI apps and assistants support the format, including many built on Claude, GPT, and Gemini — check the Client Showcase page for the current list.
Do I need to know how to code?
Writing a simple skill mostly means putting plain instructions into a text file, but some skills bundle scripts, so a bit of technical comfort helps.
Is this itself something I install?
Not exactly — this project is the open rulebook for the format, not a single tool. You build or download skill folders separately and load them into whichever assistant you use.
Where can I ask questions or share what I've made?
The project's Discord community is set up for exactly that.
Is there a cost to use this?
The specification and code are free and openly licensed.