> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openbrowser.me/llms.txt
> Use this file to discover all available pages before exploring further.

# Documentation MCP

> Add OpenBrowser documentation context to Claude Code and other MCP clients

## Overview

The OpenBrowser documentation MCP server provides read-only access to OpenBrowser documentation for Claude Code and other MCP-compatible clients. This gives AI assistants deep context about the OpenBrowser library when helping you write code.

<Note> Looking to give an assistant OpenBrowser capabilities? Check out our <a href="/customize/integrations/mcp-server"> Browser Automation MCP.</a> </Note>

## Quick Start

Add the documentation server to your coding agent:

<Tabs>
  <Tab title="Claude Code">
    ```bash theme={null}
    claude mcp add openbrowser-docs --command "uvx openbrowser-ai mcp"
    ```
  </Tab>

  <Tab title="Cursor">
    Add to `~/.cursor/mcp.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "openbrowser-docs": {
          "command": "uvx",
          "args": ["openbrowser-ai", "mcp"]
        }
      }
    }
    ```
  </Tab>

  <Tab title="Windsurf">
    Add to `~/.codeium/windsurf/mcp_config.json`:

    ```json theme={null}
    {
      "mcpServers": {
        "openbrowser-docs": {
          "command": "uvx",
          "args": ["openbrowser-ai", "mcp"]
        }
      }
    }
    ```
  </Tab>
</Tabs>

This enables your AI coding assistant to access OpenBrowser documentation when answering questions or helping with implementation.

## What This Provides

The documentation MCP server gives AI assistants access to:

* API reference and usage patterns
* Configuration options and parameters
* Best practices and examples
* Troubleshooting guides
* Architecture explanations

**Example interactions:**

```
"How do I configure custom tools in OpenBrowser?"

"What are the available agent parameters?"

"Show me how to use remote browsers via CDP."
```

Claude Code can now answer these questions using up-to-date documentation context.

## How It Works

The MCP server provides browser automation capabilities:

* Runs OpenBrowser locally via the MCP protocol
* Provides browser automation tools to AI assistants
* No external API keys required (uses your local LLM keys)

## Next Steps

* Start coding with [Agent Basics](/customize/agent/basics)