> For the complete documentation index, see [llms.txt](https://docs.pepeunchained.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.pepeunchained.com/agent-tooling/mcp-and-skills.md).

# MCP Server and Skills (overview)

This is the hub for the PEPU agent tooling: infrastructure that helps coding agents (Claude Code, Cursor, VS Code and similar) build on Pepe Unchained and get PEPU's specific facts right. It is not a claim that PEPU is "an AI chain". It is not one. AI is a feature builders add to their own products. This tooling exists so an agent, or a person using one, does not guess a chain ID, invent an address or use the canonical Multicall3 address that has no code here (see [The Multicall3 Trap](/contracts-and-addresses/multicall3.md)).

![Dev kit map: the SDK at the base; the MCP server with 33 tools, 12 skills, the Claude Code plugin, the create-pepu-app CLI and 6 templates all build on it](https://3849619095-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FKvVTxzL1E48r5cDRL2er%2Fuploads%2Fgit-blob-124a3bf8f925ffa6013a7f99ea2b11897bcc4b11%2Fdevkit-map.png?alt=media)

*Everything builds on the SDK.*

The hosted bundler (`https://bundler.pepeunchained.com`) and paymaster service (`https://paymaster.pepeunchained.com`) are the defaults. Sponsorship needs a project token in the paymaster URL (`PEPU_PAYMASTER_URL`).

## The pieces

| Piece              | What it is                                                                                                                                                       | Page                                                                                                         |
| ------------------ | ---------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------ |
| `pepu` MCP server  | `@pepu-chain/mcp`: 33 tools, 6 prompts and 5 resources for chain facts, probes, explainers, scaffolding, env and Privy checks, session policies and SDK snippets | [MCP Server Overview](/mcp/index.md), [Install per Client](/mcp/install.md), [Tool Reference](/mcp/tools.md) |
| Skills             | 12 `SKILL.md` playbooks, each built on `@pepu-chain/sdk` and the MCP tools                                                                                       | [Skills Catalog](/agent-tooling/skills.md)                                                                   |
| Claude Code plugin | `pepu-devkit`: the MCP server, the 12 skills, 6 slash commands and the `pepu-builder` subagent in one install                                                    | [Claude Code Plugin](/agent-tooling/plugin.md)                                                               |
| Template agent kit | `AGENTS.md`, `CLAUDE.md` and editor configs shipped in every scaffolded app                                                                                      | [AGENTS.md and CLAUDE.md in Templates](/agent-tooling/agents-md.md)                                          |

## Quick start

In Claude Code, install the plugin (recommended):

```
/plugin marketplace add pepu-org/pepu-devkit
/plugin install pepu-devkit@pepu
```

Or add only the MCP server with `claude mcp add pepu -- npx -y @pepu-chain/mcp`. Cursor, VS Code and other clients are covered in [Install per Client](/mcp/install.md).

## How they fit together

* The SDK (`@pepu-chain/sdk`) holds the functions. Skills and tools tell the agent to wire them, not to hand-roll infrastructure.
* The MCP server answers live questions: what is deployed, is sponsorship available, what does this error mean, which env var is missing.
* Skills carry the method: the order of steps, the dashboard settings, the safety rules and the pitfalls.
* Each skill tells the agent to call `pepu_chain_info` or the relevant probe first, so a fact that changed since the skill was written is caught by a live check.

The same line applies to everything here: it makes building on PEPU easier, or it does not belong.

## Related

* [Build with an AI Coding Agent in 3 Minutes](/get-started/ai-assisted.md)
* [create-pepu-app CLI](/templates-and-cli/create-pepu-app.md)
* [SDK Overview and Reference](/sdk/reference.md)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.pepeunchained.com/agent-tooling/mcp-and-skills.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.