MCP Grok - AI Communication Server

Enable Claude to communicate with Grok AI through the Model Context Protocol

🚀 Features

• Direct Grok Integration: Send messages to Grok and receive responses
• Model Selection: Uses the latest Grok-4 model
• Customizable Parameters: Control temperature, max tokens, and system prompts
• Simple & Lightweight: Minimal dependencies, easy to set up

📋 Prerequisites

• Node.js 18 or higher
• Grok API Key - Get yours at x.ai/api

🛠️ Installation

Quick Install for Claude Desktop

# Install globally for all projects
claude mcp add grok -s user -e GROK_API_KEY=your_api_key -- npx @chinchillaenterprises/mcp-grok

# Or install for specific project
claude mcp add grok -s project -e GROK_API_KEY=your_api_key -- npx @chinchillaenterprises/mcp-grok

Manual Installation

1. Install the package:

   npm install -g @chinchillaenterprises/mcp-grok

2. Configure Claude Desktop:

   Edit your Claude configuration file:

   • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
   • Windows: %APPDATA%\Claude\claude_desktop_config.json

   {
     "mcpServers": {
       "grok": {
         "command": "npx",
         "args": ["@chinchillaenterprises/mcp-grok"],
         "env": {
           "GROK_API_KEY": "your_grok_api_key_here"
         }
       }
     }
   }

🎯 Quick Start

1. Send a message to Grok:

   Use grok_send_message with message "Explain quantum computing in simple terms"

2. Send a message with custom parameters:

   Use grok_send_message with message "Write a haiku about AI", temperature 1.5, and system_prompt "You are a creative poet"

🔧 Available Tools

grok_send_message

Send a message to Grok AI and get a response.

Parameters:

• message (required): The message to send to Grok
• model (optional): The Grok model to use (defaults to grok-4-0709)
• temperature (optional): Sampling temperature between 0 and 2 (default: 0.7)
• max_tokens (optional): Maximum tokens in response (default: 2048)
• system_prompt (optional): System prompt to set context

Example:

{
  "message": "What's the meaning of life?",
  "model": "grok-4-0709",
  "temperature": 0.8,
  "system_prompt": "You are Grok, a witty AI assistant with a sense of humor"
}

grok_list_models

List all available Grok models and show the currently selected model.

No parameters required.

Returns:

{
  "models": ["grok-4-0709"],
  "current_model": "grok-4-0709"
}

grok_set_model

Set the default Grok model for subsequent messages.

Parameters:

• model (required): The Grok model to use

Example:

{
  "model": "grok-4-0709"
}

📊 Usage Examples

Basic Conversation

You: Use grok_send_message with message "Tell me a joke about programming"

Grok: Why do programmers prefer dark mode? Because light attracts bugs!

Direct Messaging

You: Use grok_send_message with message "What makes you different from other AI assistants?"

Creative Writing with Higher Temperature

You: Use grok_send_message with message "Write a short story about a robot learning to dance", temperature 1.5

Setting Context with System Prompt

You: Use grok_send_message with message "Explain this code: print('Hello, World!')", system_prompt "You are a patient programming teacher for beginners"

🎨 Model Information

| Model | Best For | Context Length | Special Features | |-------|----------|----------------|------------------| | grok-4-0709 | Advanced reasoning | Extended | Most capable model, enhanced performance |

⚙️ Configuration

Environment Variables

| Variable | Description | Required | |----------|-------------|----------| | GROK_API_KEY | Your Grok API key from x.ai | ✅ Yes |

Getting Your API Key

1. Visit x.ai/api
2. Sign in with your X (Twitter) account
3. Navigate to API Keys section
4. Generate a new API key
5. Copy and save it securely

🚨 Error Handling

The server provides clear error messages for common issues:

• Missing API Key: Instructions on how to set GROK_API_KEY
• Invalid API Key: Prompts to check your credentials
• Rate Limiting: Clear message when rate limits are exceeded
• Network Issues: Descriptive connection error messages
• Invalid Model: Lists available models when wrong model is specified

🧪 Development

Local Development Setup

# Clone the repository
git clone https://github.com/ChinchillaEnterprises/ChillMCP.git
cd ChillMCP/mcp-grok

# Install dependencies
npm install

# Build the project
npm run build

# Run in development mode
npm run dev

Testing Locally

# Set your API key
export GROK_API_KEY=your_api_key_here

# Test with Claude Desktop
claude mcp add grok-dev -s user -- node $(pwd)/dist/index.js

🛣️ Roadmap

Version 1.0 (Current)

• ✅ Basic message sending
• ✅ Model selection
• ✅ Custom parameters

Version 2.0 (Planned)

• 🔄 Conversation management
• 🔄 Message history
• 🔄 Multi-account support

Version 3.0 (Future)

• 🔄 Image analysis with vision model
• 🔄 Real-time information access
• 🔄 X/Twitter integration
• 🔄 Personality modes (fun, serious, based)

🤝 Contributing

1. Fork the repository
2. Create your feature branch: git checkout -b feature/amazing-feature
3. Commit your changes: git commit -m 'Add amazing feature'
4. Push to the branch: git push origin feature/amazing-feature
5. Open a pull request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🔗 Links

• NPM Package: @chinchillaenterprises/mcp-grok
• GitHub Repository: ChillMCP
• Grok API Documentation: x.ai/api
• Model Context Protocol: modelcontextprotocol.io

🆘 Support

• Issues: GitHub Issues
• Grok Support: x.ai/support

Built with ❤️ by Chinchilla Enterprises for the AI community