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

# Quickstart

> Get Browseagent running in 3 simple steps.

## Get started in three steps

Get your MCP server running by adding Browseagent server config to your AI assistant configuration.

```json theme={null}
{
  "mcpServers": {
    "browseagent": {
      "command": "npx",
      "args": ["@browseagent/mcp@latest"]
    }
  }
}
```

### Step 1: Configure MCP Server

<AccordionGroup>
  <Accordion title="VS Code" icon="code">
    See the complete [VS Code MCP documentation](https://code.visualstudio.com/docs/copilot/customization/mcp-servers) guide here
  </Accordion>

  <Accordion title="Cursor" icon="mouse-pointer-2">
    See the complete [Cursor MCP documentation](https://docs.cursor.com/en/context/mcp) guide here
  </Accordion>

  <Accordion title="Claude Desktop" icon="audio-lines">
    See the complete [Claude MCP documentation](https://modelcontextprotocol.io/quickstart/user) guide here
  </Accordion>

  <Accordion title="Windsurf" icon="sailboat">
    See the complete [Windsurf MCP documentation](https://docs.windsurf.com/windsurf/getting-started) guide here
  </Accordion>
</AccordionGroup>

### Step 2: Install Chrome Extension

1. Open [Chrome Web Store](https://chrome.google.com/webstore/detail/browseagent)
2. Click "Add to Chrome" to install the extension
3. Pin the extension to your toolbar (recommended)

### Step 3: Connect & Start Automating

<AccordionGroup>
  <Accordion title="Connect" icon="unplug">
    1. Open Claude Desktop (or your MCP client)
    2. Click the Browseagent extension icon in Chrome
    3. Click "Connect" in the popup
  </Accordion>

  <Accordion title="Start Automating" icon="sparkles">
    Test with a simple command in Claude (or your MCP client): `Take a screenshot of google.com`
  </Accordion>
</AccordionGroup>

That's it! You should now see Claude navigate to Google and return a screenshot.