MCP-Enabled Cloudflare Worker for Claude

[MCP-Enabled Cloudflare Worker for Claude] Model Context Protocol Enabled MCP Server Integration for AI Applications

Overview: What is MCP-Enabled Cloudflare Worker for Claude?

The MCP-Enabled Cloudflare Worker for Claude is a bespoke server implementing the Model Context Protocol (MCP), enabling Claude, one of the advanced AI applications, to interface with various external services and APIs seamlessly. MCP ensures a standardized approach that facilitates communication between AI models like Claude Desktop, Continue, Cursor, and other clients through a universal protocol. This document will guide you through setting up, configuring, and using this server for enhanced AI workflows.

🔧 Core Features & MCP Capabilities

The MCP-Enabled Cloudflare Worker for Claude provides the following functionalities:

1. Weather Information: Retrieves current weather data for any city.
2. IP Geolocation: Looks up detailed information about any IP address, including "me" to get client's IP details.
3. Web Search: Performs Google searches with customizable result counts.
4. HTTP Request Proxy: Enables making custom HTTP requests to any endpoint with full control over headers and body data.

These capabilities adhere to the MCP protocol standards, ensuring compatibility across various AI clients that support this protocol. The server is designed to be extensible; you can add new functions by writing custom code in TypeScript within the src/index.ts file.

⚙️ MCP Architecture & Protocol Implementation

The architecture of the MCP-Enabled Cloudflare Worker for Claude follows the MCP core principles, enabling seamless integration with any client that adheres to this protocol. The worker is built using TypeScript and integrates existing APIs for weather data, IP geolocation, web search, and HTTP requests.

1. API Integration

• Weather Information: Uses the OpenWeatherMap API.
• IP Geolocation: Utilizes the IPInfo service.
• Web Search: Leverages Google Custom Search.
• HTTP Requests: Employs the native fetch functionality of JavaScript.

Each function is meticulously documented using JSDoc, ensuring that clients can understand and leverage this server's interface effectively. The workers-mcp framework handles the protocol-level communication, enabling seamless interactions between AI models and external data sources.

🚀 Getting Started with Installation

To install and configure the MCP-Enabled Cloudflare Worker for Claude, follow these steps:

1. Install Wrangler CLI:

   npm install -g wrangler
   

2. Log in to Cloudflare:

   wrangler login
   

3. Clone the Repository and Navigate Inside:

   git clone https://github.com/anishgowda21/cf-mcp-server.git
   cd cf-mcp-server
   

4. Install Dependencies:

   npm install
   

5. Set Up MCP:

   npx workers-mcp setup
   

6. Configure Environment Variables in the Cloudflare Dashboard or wrangler.toml file:

   • OPENWEATHERMAP_API_KEY
   • IPINFO_API_KEY
   • GOOGLE_API_KEY
   • GOOGLE_CX

7. Deployment:

   npm run deploy
   

💡 Key Use Cases in AI Workflows

1. Weather Data for Contextual Conversations

   • Use case: Enhancing Claude's conversations with relevant weather information.
   • Technical Implementation: Integrate getWeatherData(cityName: string) and pass the city name to fetch current weather details.

2. Geolocation for Enhanced Security and Insights

   • Use case: Providing detailed geolocation data on IP addresses in security logs or customer support inquiries.
   • Technical Implementation: Add the function getIpDetails(ipAddr: string) to retrieve accurate location information.

🔌 Integration with MCP Clients

The MCP-Enabled Cloudflare Worker for Claude is designed to be fully compatible with a variety of AI clients, including:

MCP Client	Resources	Tools	Prompts
Claude Desktop	✅	✅	✅
Continue	✅	✅	✅
Cursor	❌	✅	❌

📊 Performance & Compatibility Matrix

The performance and compatibility matrix for this server ensures optimal functioning across multiple clients:

• Claude Desktop: Fully compatible with all features.
• Continue: Fully compatible, though only the tools are utilized. Prompts functionality is not supported.

🛠️ Advanced Configuration & Security

To further customize or secure the worker, follow these steps:

1. Edit Source Code:
   • Modify src/index.ts to add new functions and methods.
2. Add JSDoc Comments: Document new functionalities using JSDoc in src/index.ts.
3. Deploy Changes: Re-run the deployment process with npm run deploy.

❓ Frequently Asked Questions (FAQ)

1. Q: How do I enable web searches?
   • A: Ensure you have set up the GOOGLE_API_KEY and GOOGLE_CX environment variables.
2. Q: Can I customize HTTP requests for specific URLs?
   • A: Yes, use the function makeRequest(url: string, method: string, params?: object) to send custom requests.
3. Q: How do I handle API keys securely?
   • A: It's recommended to store API keys in environment variables through Cloudflare's dashboard for security.
4. Q: What if some functions don't work as expected?
   • A: Double-check your API key setup and the relevant function implementation in src/index.ts.
5. Q: Can I extend this server to support more tools?
   • A: Yes, add new methods to MyWorker class and document them using JSDoc.

👨‍💻 Development & Contribution Guidelines

Contributions are highly welcome! To contribute, follow these steps:

1. Clone the Repository:

   git clone https://github.com/anishgowda21/cf-mcp-server.git
   cd cf-mcp-server
   

2. Make Your Changes: Edit src/index.ts and README.md.
3. Run Tests: Ensure all functions work as expected.
4. Commit and Push:

   git commit -m "Your detailed description of the changes."
   git push origin main
   

🌐 MCP Ecosystem & Resources

For more information on MCP, visit the official Model Context Protocol documentation or join relevant communities to engage with other developers.

MCP Protocol Flow Diagram:

graph TD
    A[AI Application] -->|MCP Client| B[MCP Protocol]
    B --> C[MCP Server]
    C --> D[Data Source/Tool]
    style A fill:#e1f5fe
    style C fill:#f3e5f5
    style D fill:#e8f5e8

MCP Client Compatibility Matrix:

MCP Client	Resources	Tools	Prompts
Claude Desktop	✅	✅	✅
Continue	✅	✅	✅
Cursor	❌	✅	❌

MCP Configuration Code Sample:

{
  "mcpServers": {
    "weatherDataServer": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-weather"],
      "env": {
        "OPENWEATHERMAP_API_KEY": "your-api-key"
      }
    },
    "ipGeolocationServer": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-ip"],
      "env": {
        "IPINFO_API_KEY": "your-api-key"
      }
    },
    "webSearchServer": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-websearch"],
      "env": {
        "GOOGLE_API_KEY": "your-api-key",
        "GOOGLE_CX": "your-custom-search-engine-id"
      }
    },
    "httpRequestServer": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-http"],
      "env": {}
    }
  }
}

This comprehensive documentation ensures a detailed understanding of how to set up, configure, and use the MCP-Enabled Cloudflare Worker for Claude effectively. By adhering to these instructions, developers can significantly enhance the functionality and interoperability of AI applications like Claude with external data sources and tools through MCP.