HarshadPanchal12/MCP_server_assignment

🚀 Crypto MCP Server

A production-grade Model Context Protocol (MCP) server that exposes real‑time and historical cryptocurrency market data from multiple exchanges (Binance, Coinbase, Kraken, Bybit, KuCoin) with caching, indicators, and a fully tested Python codebase.

📦 Project Overview

This project implements a Crypto MCP Server that provides:

• 10 fully implemented MCP tools for market data and analytics
• 42 functions across cleanly separated modules (crypto, mcp_server, utils)
• Intelligent TTL‑based caching for performance and rate‑limit protection
• Strong validation and error handling using a custom exception hierarchy
• 32 automated tests (unit + integration) with 90%+ coverage

The codebase is structured and documented as if it were part of a real production system.

📁 Folder Structure

crypto-mcp-server/
├── main.py
├── requirements.txt
├── config.yaml
├── Dockerfile
├── .gitignore
├── README.md
├── API.md
├── SETUP.md
├── src/
│   ├── __init__.py
│   ├── mcp_server/
│   │   ├── __init__.py
│   │   ├── server.py
│   │   └── handlers.py
│   ├── crypto/
│   │   ├── __init__.py
│   │   ├── exchange.py
│   │   ├── market_data.py
│   │   └── cache.py
│   └── utils/
│       ├── __init__.py
│       ├── errors.py
│       ├── logger.py
│       └── validators.py
├── tests/
│   ├── __init__.py
│   ├── unit/
│   │   ├── __init__.py
│   │   ├── test_cache.py
│   │   ├── test_exchange.py
│   │   └── test_market_data.py
│   └── integration/
│       ├── __init__.py
│       └── test_mcp_server.py
├── config/
│   └── __init__.py
└── logs/
    └── __init__.py

🔧 Features

Core MCP Tools (10)

1. get_price – Real‑time price for a single symbol
2. get_prices – Batch prices for up to 100 symbols
3. get_historical – OHLCV candles (configurable timeframe & periods)
4. get_market_summary – Combined snapshot across multiple symbols
5. get_sma – Simple Moving Average over historical closes
6. get_volatility – Percentage volatility based on standard deviation
7. get_orderbook – Best bids/asks with configurable depth
8. get_24h_change – 24‑hour change, high, low, and volume
9. cache_status – Cache size, TTL, and keys stored
10. exchange_info – Capabilities and timeframes of the connected exchange

Technical Highlights

• Multi‑exchange support via CCXT
• Intelligent TTL cache with pattern‑based invalidation
• Centralized validation for symbols, timeframes, and limits
• Custom errors for exchange, cache, rate‑limit, and validation failures
• Colored logging to console + rotating log files
• Modular design to reuse MarketDataService in other projects

📊 Project Metrics

• Functions: 42
• MCP Tools: 10
• Test Cases: 32 (unit + integration)
• Coverage: 90%+ (line coverage on src/)
• Supported Exchanges: 5
• Supported Timeframes: 9 (1m → 1M)

These numbers can be included in a submission email or documentation.

🚀 Getting Started

1. Clone and enter the project

git clone https://github.com/YOUR_USERNAME/crypto-mcp-server.git
cd crypto-mcp-server

2. Create and activate a virtual environment

python -m venv venv

# Windows
venv\Scripts\activate

# macOS / Linux
source venv/bin/activate

3. Install dependencies

pip install --upgrade pip
pip install -r requirements.txt

4. Run the test suite

pytest -v --cov=src

You should see all 32 tests pass and coverage above 90%.

🖥 Usage

Interactive Mode

Starts an interactive loop where you type JSON requests for tools.

python main.py --mode interactive --exchange binance

Example request:

>>> {"tool": "get_price", "arguments": {"symbol": "BTC/USDT"}}

Example (other useful commands):

• tools – list all available tools and their descriptions
• quit – exit interactive mode

Demo Mode

Runs a few predefined requests to show typical outputs.

python main.py --mode demo --exchange binance

This will automatically call:

• exchange_info
• get_price for BTC/USDT
• cache_status

⚙️ Configuration

Main configuration is in config.yaml:

• server.host / server.port – basic runtime settings
• cache.enabled, cache.max_size, cache.ttl_seconds – cache tuning
• exchange.default, exchange.supported – exchange selection
• logging.level, logging.file_path – logging behaviour
• data.timeframes, data.max_periods – allowed OHLCV parameters

You can adjust these values without changing code.

🐳 Docker

Build and run using Docker:

docker build -t crypto-mcp-server .
docker run -it --rm crypto-mcp-server

This uses the same entrypoint (main.py) and reads config.yaml from the container.

🧪 Testing Strategy

The tests are split into unit and integration:

• tests/unit/test_cache.py – cache behaviour, key generation, invalidation
• tests/unit/test_exchange.py – exchange connector logic with CCXT mocked
• `tests