fumadocs-mcp
# Fumadocs MCP Server
An MCP (Model Context Protocol) server that provides AI tools with access to [Fumadocs](https://fumadocs.dev) documentation. This makes it easier for AI assistants to help you integrate Fumadocs into your existing projects.
## Features
- **Search documentation** - Find relevant docs by keyword
- **Browse topics** - Explore documentation sections
- **Fetch pages** - Get full content of any documentation page
- **Setup guides** - Get framework-specific installation instructions
- **Component docs** - Get UI component documentation with props and examples
## Installation
### Using npx (recommended)
```bash
npx fumadocs-mcp
```
### Global installation
```bash
npm install -g fumadocs-mcp
fumadocs-mcp
```
### From source
```bash
git clone https://github.com/k4cper-g/fumadocs-mcp
cd fumadocs-mcp
npm install
npm run build
```
## Configuration
### Claude Desktop
Add to your Claude Desktop configuration file:
**macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
**Windows**: `%APPDATA%\Claude\claude_desktop_config.json`
```json
{
"mcpServers": {
"fumadocs": {
"command": "npx",
"args": ["fumadocs-mcp"]
}
}
}
```
Or if installed from source:
```json
{
"mcpServers": {
"fumadocs": {
"command": "node",
"args": ["/path/to/fumadocs-mcp/build/index.js"]
}
}
}
```
### Claude Code
The MCP will be automatically available when configured in Claude Desktop.
## Available Tools
### `list_topics`
Browse available documentation sections and topics.
```
Parameters:
- section (optional): "all" | "cli" | "headless" | "framework" | "mdx" | "ui"
```
### `search_docs`
Search documentation by keyword.
```
Parameters:
- query (required): Search query string
- section (optional): Filter to a specific section
```
### `get_page`
Fetch the full content of a documentation page.
```
Parameters:
- path (required): Documentation path (e.g., "/docs/manual-installation/next")
```
### `get_setup_guide`
Get a complete setup guide for adding Fumadocs to an existing project.
```
Parameters:
- framework (required): "next" | "react-router" | "tanstack-start" | "waku"
- includeUI (optional): Whether to include UI setup instructions (default: true)
```
### `get_component`
Get documentation for a specific UI component.
```
Parameters:
- component (required): Component name (e.g., "accordion", "tabs", "codeblock")
```
## Example Usage
Once configured, you can ask Claude things like:
- "How do I add Fumadocs to my existing Next.js project?"
- "Show me how to use the Tabs component in Fumadocs"
- "Search for documentation about internationalization"
- "What components are available in Fumadocs UI?"
## Development
```bash
# Install dependencies
npm install
# Build
npm run build
# Run in development mode
npm run dev
# Test with MCP Inspector
npm run inspector
```
## License
MIT
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose: get_component retrieves UI component details, get_page fetches full page content, get_setup_guide provides framework-specific setup instructions, list_topics browses documentation sections, and search_docs performs keyword searches. There is no overlap in functionality, making tool selection straightforward for an agent.
All tool names follow a consistent verb_noun pattern with snake_case: get_component, get_page, get_setup_guide, list_topics, and search_docs. The verbs (get, list, search) are appropriately descriptive and uniform, creating a predictable and readable naming convention throughout the set.
With 5 tools, the server is well-scoped for its purpose of accessing Fumadocs documentation. Each tool serves a specific and necessary function (e.g., retrieving components, pages, setup guides, browsing topics, and searching), with no redundancy or missing core operations, making the count ideal for the domain.
The tool set provides complete coverage for documentation access: it supports retrieving specific components and pages, getting setup instructions, browsing topics, and searching. This covers all typical user workflows (lookup, navigation, and discovery) without obvious gaps, ensuring agents can handle common documentation tasks effectively.