pmind-grok-mcp

# PMIND Grok MCP Server > ⚠️ **Experimental**: This MCP server is in an experimental state and may have rough edges. Please report any issues you encounter. A comprehensive MCP (Model Context Protocol) server for Grok AI, providing access to chat, reasoning, live search, image understanding, and image generation capabilities through a clean, modular interface. ## Features - 🤖 **Chat Completions** - Basic and history-managed conversations with Grok models - 🧠 **Reasoning** - Step-by-step problem solving with access to thinking traces - 🔍 **Live Search** - Real-time data from web, X (Twitter), news, and RSS feeds - 👁️ **Image Understanding** - Analyze images with custom prompts - 🎨 **Image Generation** - Create images from text descriptions ## Prerequisites - Python 3.12-3.13 - xAI API key from [https://console.x.ai](https://console.x.ai) - uv package manager ## Installation ```bash # Clone the repository git clone https://github.com/yourusername/pmind-grok-mcp.git cd pmind-grok-mcp # Install dependencies with uv uv sync ``` ## Configuration 1. Copy the example environment file: ```bash cp .env.example .env ``` 2. Edit `.env` with your settings: ```bash # Required XAI_API_KEY=your_api_key_here # Optional GROK_MODEL=grok-4 GROK_TIMEOUT=3600 DOWNLOADS_DIR=downloads ``` ## Usage ### Start the Server ```bash uv run pmind-grok-mcp ``` The server will start and be available for MCP clients to connect. ### Using with Claude Code Use the following command to add the server: ```bash claude mcp add pmind-grok-mcp -- uv run --directory /path/to/pmind-grok-mcp pmind-grok-mcp ``` Replace `/path/to/pmind-grok-mcp` with the actual path where you cloned the repository. ## Available Tools The server provides tools for chat completions, reasoning, live search, image understanding, and image generation. Each tool is accessible through the MCP protocol with comprehensive parameter documentation. ## Troubleshooting ### API Key Issues - Ensure your API key is set in the environment or `.env` file - Verify the key at [https://console.x.ai](https://console.x.ai) ### Timeout Errors - Increase `GROK_TIMEOUT` for reasoning models (default: 3600s) - Reasoning models require longer processing time ### Rate Limiting - Monitor your usage at [https://console.x.ai](https://console.x.ai) - Implement exponential backoff for high-volume requests