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

# Integrate Para Docs MCP with AI Tools

> Connect the Para Docs Mintlify MCP server to AI tools for direct documentation access and enhanced coding assistance

Connect the Para Docs MCP server to AI tools for seamless access to documentation, code examples, and guides. This integration enables AI assistants to search Para Docs directly via the Model Context Protocol.

## Prerequisites

* Active accounts for target AI tools
* Para Docs MCP server URL: `http://docs.getpara.com/mcp`
* AI tool with remote MCP connection support (may be in beta)

## Installation and Setup

Set up connections by adding the MCP server as a custom connector in each tool's settings.

<Tabs>
  <Tab title="ChatGPT">
    ### Configure ChatGPT Connection

    1. Open ChatGPT settings from your avatar menu
    2. Select **Connectors** in the sidebar
    3. Click **Create** to open the New Connector dialog
    4. Enter the MCP server URL: `http://docs.getpara.com/mcp`
    5. Configure authentication if required (Para Docs MCP uses no security by default)
    6. Save and test the connection
  </Tab>

  <Tab title="Claude Desktop">
    ### Configure Claude Desktop

    1. Navigate to **Settings > Extensions** in Claude Desktop
    2. Click **Advanced settings** and locate the Extension Developer section
    3. Add a custom connector with the remote MCP URL: `http://docs.getpara.com/mcp`
    4. Note: Remote support is in beta; use local STDIO if preferred
    5. Verify the connection in Claude's interface
  </Tab>

  <Tab title="Claude Code">
    ### Configure Claude Code

    1. Run the CLI command: `claude mcp add`
    2. Follow the wizard to input the MCP server URL: `http://docs.getpara.com/mcp`
    3. Select remote MCP support
    4. Integrate tools like search and fetch for Para Docs access
    5. Restart Claude Code to apply changes
  </Tab>

  <Tab title="Cursor">
    ### Configure Cursor

    1. Open Cursor settings and navigate to **Models** or **API Keys**
    2. Disable unnecessary models if needed
    3. Add a custom model or provider, overriding the base URL to `http://docs.getpara.com/mcp`
    4. Use agent mode for MCP interactions (similar to VS Code Copilot)
    5. Verify by testing a documentation query in the editor
  </Tab>
</Tabs>

## Usage

Query the MCP server via your AI tool's interface after setup. Provide search terms to the "SearchParaDocs" tool for relevant results.

<Steps>
  <Step title="Invoke Search Tool">
    Ask your AI: "Search Para Docs for \[topic]"

    The tool returns titles, snippets, and links to relevant documentation.
  </Step>

  <Step title="Handle Responses">
    Review returned contextual content and follow links for full details.
  </Step>

  <Step title="Error Handling">
    Check for connection issues and retry if the server is unreachable.
  </Step>
</Steps>

## Example

### Query Example

```text theme={null}
Search Para Docs for how to sign a basic message
```

### Expected Response

The MCP server returns structured results:

```json Response theme={null}
{
  "results": [
    {
      "title": "Sign Messages with Para",
      "snippet": "Learn how to sign messages using Para's wallet integration...",
      "link": "https://docs.getpara.com/v3/react/guides/web3-operations/sign-with-para"
    },
    {
      "title": "Web3 Operations Guide",
      "snippet": "Complete guide for signing transactions and messages...",
      "link": "https://docs.getpara.com/v3/react/guides/web3-operations"
    }
  ]
}
```