Crypto Price Server

Python License Stars MCP GitHub

A Model Context Protocol (MCP) server for real-time cryptocurrency price data — prices, market data, trending coins, historical charts, and currency conversion. Powered by the free CoinGecko API — no API key required.

Built for AI agents. Works with Hermes Agent, Claude Code, Cursor, Windsurf, and any MCP-compatible client.

✨ Features (9 Tools)

💵 Price & Conversion (2 Tools)

Tool	Description
get_price	Current price of any crypto in any currency (USD, EUR, JPY, BTC, etc.)
convert_currency	Convert an amount of one crypto to another currency in real time

📊 Market Data (2 Tools)

Tool	Description
get_market_data	Full market snapshot: rank, market cap, supply, ATH, 24h/7d/30d change
get_global_stats	Overall crypto market: total cap ($2.25T+), 24h volume, BTC/ETH dominance

🔥 Discovery (3 Tools)

Tool	Description
get_trending	Top 15 trending coins on CoinGecko right now
get_top_gainers	Top gainers/losers in the last 24 hours
search_coins	Search coins by name or symbol

📈 Historical (1 Tool)

Tool	Description
get_historical_prices	Historical daily OHLC prices (1d, 7d, 30d, 90d, 365d, max)

ℹ️ Info (1 Tool)

Tool	Description
get_coin_info	Detailed coin info: description, links, categories, genesis date

🚀 Quick Start

Install

# From GitHub
pip install git+https://github.com/ceeyang-ai/crypto-price-mcp-server.git

Run

# Start as MCP server (stdio mode)
crypto-price-mcp-server

🔌 Usage with AI Agents

Hermes Agent

Add to ~/.hermes/config.yaml:

mcp_servers:
  crypto-price:
    command: "crypto-price-mcp-server"

Restart → use mcp_crypto_price_get_price, mcp_crypto_price_get_trending, etc.

Claude Code / Cursor / Any MCP Client

Add to your MCP config:

{
  "mcpServers": {
    "crypto-price": {
      "command": "crypto-price-mcp-server"
    }
  }
}

📖 Examples

Check Bitcoin Price

get_price(coin_id="bitcoin", vs_currency="usd")

Returns:

**BITCOIN** (USD)
• Price: 62,807

What's Trending?

get_trending()

Returns top 15 trending coins with rank, BTC price, and score.

Full Market Analysis

get_market_data(coin_id="solana", vs_currency="usd")

Returns rank, price, market cap, volume, supply, ATH, 24h/7d/30d change percentages.

Historical Data

get_historical_prices(coin_id="bitcoin", vs_currency="usd", days=7)

Returns open, close, high, low, change percentage, and recent data points.

Convert Crypto

convert_currency(coin_id="bitcoin", amount=0.5, vs_currency="usd")

Returns:

💱 **0.5 BITCOIN** = **31,417.50 USD**
   Rate: 1 BITCOIN = 62,835 USD

🛠 Requirements

• Python 3.10+
• mcp ≥ 1.0
• httpx ≥ 0.27
• Internet access (CoinGecko API)
• No API key required (uses free public CoinGecko API)

🌐 API Dependencies

This server uses the CoinGecko free API:

• Rate limit: ~30 calls/minute (free tier)
• No registration needed — works out of the box
• No API key required for public endpoints

If you need higher rate limits, consider using a CoinGecko API key with the x_cg_demo_api_key header (modify the _request function in __init__.py).

📦 Project Structure

crypto-price-mcp-server/
├── src/crypto_price_mcp_server/
│   └── __init__.py        # All 9 tools + CoinGecko API client
├── pyproject.toml          # Build config (setuptools)
├── README.md
└── LICENSE (MIT)

👨‍💻 Development

git clone https://github.com/ceeyang-ai/crypto-price-mcp-server.git
cd crypto-price-mcp-server
pip install -e .
python -c "from crypto_price_mcp_server import get_price; print(get_price('bitcoin'))"

Running Tests

python tests/test_server.py

Tests hit the real CoinGecko API — be mindful of the 30 calls/min rate limit.

📄 License

MIT — free for personal and commercial use.