my-mcp-server

What is my-mcp-server?

my-mcp-server is a high-performance cryptocurrency market data and analysis platform built on the Model Context Protocol (MCP). It integrates with multiple cryptocurrency exchanges via CCXT, stores time-series data in TimescaleDB, and aggregates crypto news from RSS feeds. It provides real-time market data, historical analysis, LLM-powered insights, and news sentiment analysis, enabling seamless integration with AI applications.

How to use my-mcp-server?

Install Python 3.9+, Docker Compose, and the uv package manager. Clone the repository, set up a virtual environment with uv venv and uv pip sync, start TimescaleDB via docker-compose up -d timescaledb, configure .env and config/settings.py, then run the server with uv run python main.py. Interact using any MCP client to call tools, read resources, or use prompts.

Key features of my-mcp-server

• Real-time market data from multiple cryptocurrency exchanges
• Historical price and volume data analysis with TimescaleDB
• LLM integration for market trend insights and predictions
• Crypto news aggregation with custom RSS feed subscriptions
• AI-powered news sentiment analysis
• MCP protocol support for easy integration with AI applications

Use cases of my-mcp-server

• Provide live crypto price feeds and historical data to AI trading agents
• Aggregate and analyze crypto news sentiment for market prediction
• Build multi-exchange portfolio tracking dashboards
• Backtest trading strategies using time-series historical data
• Power LLM-based investment research and trend reports

FAQ from my-mcp-server

What exchanges does my-mcp-server support?

It supports multiple exchanges via the CCXT library; specific exchange configurations can be added in services/exchange_service.py.

What are the runtime dependencies?

Python 3.9+, Docker Compose (for TimescaleDB), the uv package manager, and TimescaleDB. No external API keys are required for basic operation, but exchange API keys may be needed for private endpoints.

Where is historical data stored?

All time-series data is stored in a TimescaleDB instance (PostgreSQL extension) that must be running before the server starts. Data is organized into hypertables for efficient querying.

Can I add custom RSS feeds?

Yes. Edit the RSS_FEEDS list in config/settings.py to add or modify news sources. The system parses feeds with feedparser and supports custom parsing logic in services/news_service.py.

How does my-mcp-server communicate with clients?

It uses the Model Context Protocol (MCP) over HTTP/SSE by default. Clients connect as MCP applications to call tools, read resources, and invoke prompts. No authentication is mentioned in the README.