tanvi-argade/mcp-crypto-server

📘 Crypto MCP Server

🚀 Overview

This project is a Python-based MCP (Market Context Protocol) server that provides real-time and historical cryptocurrency market data using the CCXT library.

Key features:

• 📈 Real-time crypto prices
• 🕒 Historical OHLCV (1-hour candles) data
• ⚡ In-memory caching to reduce API calls
• 🧪 Automated tests using PyTest
• 🛡 Error handling for invalid symbols or API issues

📂 Project Structure

mcp-crypto-server/
│── src/
│   ├── server.py
│   ├── handlers/
│   │   ├── prices.py
│   │   └── historical.py
│   ├── utils/
│   │   └── cache.py
│── tests/
│   ├── test_prices.py
│── requirements.txt
│── README.md

🛠 Tech Stack

🔧 Installation

1. Clone the repo:

git clone https://github.com/<tanvi-argade>/mcp-crypto-server.git
cd mcp-crypto-server

2. Create virtual environment:

python -m venv venv
# Activate venv
# Windows
venv\Scripts\activate
# Linux/Mac
source venv/bin/activate

3. Install dependencies:

pip install -r requirements.txt

▶️ Run the MCP Server

uvicorn src.server:app --reload

Server runs at:

http://127.0.0.1:8000

🔌 API Endpoints

1️⃣ Real-time Prices

GET /prices/current?symbol=BTC/USDT

Example Response:

{
  "symbol": "BTC/USDT",
  "price": 97852.15,
  "timestamp": 1731666406266,
  "cached": false
}

2️⃣ Historical OHLCV Data

GET /historical?symbol=BTC/USDT&days=1

Example Response:

{
  "symbol": "BTC/USDT",
  "days": 1,
  "cached": false,
  "data": [
    {
      "timestamp": "2025-11-15T12:30:00",
      "open": 96121.27,
      "high": 96372.68,
      "low": 95947.68,
      "close": 96333.86,
      "volume": 618.02
    }
  ]
}

🧠 Caching System

• Cache key: historical:{symbol}:{days}
• Cached responses return "cached": true
• Reduces API calls and respects rate limits
• In-memory cache; resets on server restart

🧪 Running Tests

pytest -v

Expected output:

4 passed in XXs

📌 Assumptions

• Binance is the default crypto exchange
• OHLCV timeframe: 1-hour candles
• No external database; cache stays in memory
• API rate limits are respected

📈 Future Improvements

• Replace in-memory cache with Redis
• Add WebSocket for live prices
• Add technical indicators (RSI, MACD, EMA)
• Support multiple exchanges
• Dockerize the project