mcp

This MCP server provides a modular, plugin-based architecture for hosting multiple services under the Model Context Protocol (MCP). It exposes a central gateway that can route standard MCP requests to registered services or use the built-in local tools for quick information retrieval. Core endpoints allow you to inspect server status, health, and registered services, while the MCP endpoints enable interaction via POST or SSE, depending on the transport. The server also includes a rich set of tools for DeFi, crypto markets, NFT data, portfolio tracking, market analysis, and search utilities, all designed to be accessible through the MCP protocol or via dedicated HTTP endpoints. You can manage services dynamically (load, add, reload, enable/disable, delete) and query detailed service statistics to monitor health and performance. The included tools are designed to work with real-world data sources and provide robust error handling with sensible fallbacks when external APIs are unavailable.

Prerequisites:

• Python 3.8+ installed on your system
• Internet access to install Python dependencies

1. Clone the repository or download the code to your server.
2. Create a Python virtual environment (optional but recommended):
   • python -m venv venv
   • source venv/bin/activate # On Windows: venv\Scripts\activate
3. Install dependencies:
   • pip install -r requirements.txt
4. Prepare environment variables (optional, defaults shown):
   • MCP_HOST=0.0.0.0
   • MCP_PORT=3000
   • MCP_DEBUG=false
5. Run the server:
   • python run.py
6. Verify the server is running by hitting the root endpoint or /health:
   • curl http://localhost:3000/
   • curl http://localhost:3000/health

Tips and common considerations:

• The server supports dynamic service management; use the provided endpoints to load, add, reload, or remove services as needed.
• If using a reverse proxy or load balancer, ensure MCP_PORT is properly exposed and TLS is terminated at the proxy if required.
• Environment variables MCP_HOST, MCP_PORT, and MCP_DEBUG can tune binding, port, and debugging output for your environment.
• When integrating with external APIs, enable robust error handling in your services as the built-in tools include fallbacks for unavailable endpoints.
• For local development, you can use the VSCode integration snippet to quickly test HTTP/SSE MCP connections by configuring the server URL and type.