Grok MCP server for AI agents

The Grok MCP Plugin provides seamless access to Grok AI's powerful capabilities directly from Cline using the Model Context Protocol (MCP). It enables you to use Grok's language models, vision capabilities, and function calling features through a simple interface.

Prerequisites

Node.js (v16 or higher)
A Grok AI API key (obtain from console.x.ai)
Cline with MCP support

Installation

Clone the repository:

git clone https://github.com/Bob-lance/grok-mcp.git
cd grok-mcp

Install dependencies:
```
npm install
```
Build the project:
```
npm run build
```

Add the MCP server to your Cline MCP settings:

For VSCode Cline extension, edit the file at:

~/Library/Application Support/Code/User/globalStorage/saoudrizwan.claude-dev/settings/cline_mcp_settings.json

Add the following configuration:

{
  "mcpServers": {
    "grok-mcp": {
      "command": "node",
      "args": ["/path/to/grok-mcp/build/index.js"],
      "env": {
        "XAI_API_KEY": "your-grok-api-key"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

Replace /path/to/grok-mcp with the actual path to your installation and your-grok-api-key with your Grok AI API key.

Usage

Once installed and configured, you can use three different tools with the Grok MCP plugin in Cline:

Chat Completion

Generate text responses using Grok's language models:

<use_mcp_tool>
<server_name>grok-mcp</server_name>
<tool_name>chat_completion</tool_name>
<arguments>
{
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello, what can you tell me about Grok AI?"
    }
  ],
  "temperature": 0.7
}
</arguments>
</use_mcp_tool>

Image Understanding

Analyze images with Grok's vision capabilities:

<use_mcp_tool>
<server_name>grok-mcp</server_name>
<tool_name>image_understanding</tool_name>
<arguments>
{
  "image_url": "https://example.com/image.jpg",
  "prompt": "What is shown in this image?"
}
</arguments>
</use_mcp_tool>

You can also use base64-encoded images:

<use_mcp_tool>
<server_name>grok-mcp</server_name>
<tool_name>image_understanding</tool_name>
<arguments>
{
  "base64_image": "base64-encoded-image-data",
  "prompt": "What is shown in this image?"
}
</arguments>
</use_mcp_tool>

Function Calling

Use Grok to call functions based on user input:

<use_mcp_tool>
<server_name>grok-mcp</server_name>
<tool_name>function_calling</tool_name>
<arguments>
{
  "messages": [
    {
      "role": "user",
      "content": "What's the weather like in San Francisco?"
    }
  ],
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "get_weather",
        "description": "Get the current weather in a given location",
        "parameters": {
          "type": "object",
          "properties": {
            "location": {
              "type": "string",
              "description": "The city and state, e.g. San Francisco, CA"
            },
            "unit": {
              "type": "string",
              "enum": ["celsius", "fahrenheit"],
              "description": "The unit of temperature to use"
            }
          },
          "required": ["location"]
        }
      }
    }
  ]
}
</arguments>
</use_mcp_tool>

API Reference

Chat Completion Parameters

messages (required): Array of message objects with role and content
model (optional): Grok model to use (defaults to grok-2-latest)
temperature (optional): Sampling temperature (0-2, defaults to 1)
max_tokens (optional): Maximum number of tokens to generate (defaults to 16384)

Image Understanding Parameters

prompt (required): Text prompt to accompany the image
image_url (optional): URL of the image to analyze
base64_image (optional): Base64-encoded image data (without the data:image prefix)
model (optional): Grok vision model to use (defaults to grok-2-vision-latest)

Note: Either image_url or base64_image must be provided.

Function Calling Parameters

messages (required): Array of message objects with role and content
tools (required): Array of tool objects with type, function name, description, and parameters
tool_choice (optional): Tool choice mode (auto, required, none, defaults to auto)
model (optional): Grok model to use (defaults to grok-2-latest)

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.