MCP server for xAI's Grok API with agentic tool calling, image generation, vision, and reasoning models.

• Agentic Tool Calling: Web search, X search, and code execution with multi-step reasoning
• Multiple Grok Models: Access to Grok-4.1-Fast-Reasoning, Grok-4.1-Fast-Non-Reasoning, Grok-4-Fast, Grok-3-Mini, and more
• Image Generation: Create images using Grok's image generation model
• Vision Capabilities: Analyze images with Grok's vision models
• Reasoning Models: Advanced reasoning with extended thinking models (Grok-4.1-Fast-Reasoning, Grok-3-Mini, Grok-4)
• Stateful Conversations: Use this newly released feature to maintain conversation context as id across multiple requests

• Python 3.11 or higher
• xAI API key (Get one here)
• Astral UV

1. Clone the repository:

git clone https://github.com/merterbak/Grok-MCP.git
cd Grok-MCP

2. Create a venv environment:

uv venv
source .venv/bin/activate # macOS/Linux or .venv\Scripts\activate on Windows

3. Install dependencies:

uv sync

Add this to your Claude Desktop configuration file:

{
  "mcpServers": {
    "grok": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/Grok-MCP",
        "run",
        "python",
        "main.py"
      ],
      "env": {
        "XAI_API_KEY": "your_api_key_here"
      }
    }
  }
}

Claude Desktop can't send uploaded images in the chat to an MCP tool. The easiest way to give access to files directly from your computer is official Filesystem MCP server. After setting it up you’ll be able to just write the image’s file path (such as /Users/mert/Desktop/image.png) in chat and Claude can use it with any vision chat tool.

{
  "mcpServers": {
    "filesystem": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-filesystem",
        "/Users/<your-username>/Desktop",
        "/Users/<your-username>/Downloads"
      ]
    }
  }
}

For stdio:

uv run python main.py

Docker:

docker compose up --build

Mcp Inspector:

mcp dev main.py

List all available Grok models.

Standard chat completion.

Analyze images with text.

Returns: Content + usage with prompt_image_tokens

Get detailed reasoning with the response.

Returns: Content, reasoning_content, usage (with reasoning_tokens)

Create images from text.

Agentic web search with autonomous research.

Returns: Content, citations, tool_calls, usage

Agentic X (Twitter) search.

Returns: Content, citations, tool_calls, usage

Combined agentic search with web, X, and code execution.

Execute Python code for calculations and analysis.

Returns: Content, tool_calls, code_outputs, usage

Maintain conversation state across requests.

Returns: Content, response_id, usage

Retrieve a stored conversation.

Delete a stored conversation.

This project is open source and available under the MIT License.