CCXT

You have a CCXT MCP Server that exposes CCXT library functions as MCP tools for language models. This server lets you fetch market data, manage orders, and access account information through a lightweight, asynchronous MCP interface built on Python. It enables secure, standardized interactions with cryptocurrency exchanges from your LLM workflows.

How to use

Start the server locally and connect your MCP client to it. Use the provided MCP tools to perform public data queries (no authentication required) and private account actions (with authentication in place). The server runs asynchronously, so calls return promptly while your client continues other work.

How to install

Prerequisites you need before starting: Python, a virtual environment tool, and a basic MCP client setup. You will also need network access to reach the server when choosing the http-based connection.

Step by step you can follow to set up locally. Copy these commands exactly as shown to ensure a smooth setup.

# 1) Clone the project (if you haven’t already)
git clone https://github.com/jcwleo/ccxt-mcp-server.git
cd ccxt-mcp-server

# 2) Create and activate a virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows use .venv\Scripts\activate

# 3) Install dependencies
pip install -r requirements.txt

# 4) Start the MCP server using the standard runtime command
uv run uvicorn mcp_server:app

Configuration and startup notes

You can connect MCP clients either via an http URL or via a local stdio command depending on your setup. The http approach targets a running server at a network address, while stdio runs the server as a local process and communicates over standard I/O.

# Example MCP client config (http connection)
{
  "mcpServers": {
    "ccxt_mcp": {
      "type": "http",
      "name": "ccxt_mcp",
      "url": "http://127.0.0.1:8000/mcp/",
      "args": []
    }
  }
}

Additional sections

Security and usage notes: private tools require API credentials. Treat your API keys like secrets and ensure you use secure storage and restricted access in your environment.

Troubleshooting tips: if you see connection errors, verify the server is running (the process should show startup logs similar to a standard FastAPI/UVicorn startup). Confirm the MCP client is pointed at the correct URL and that the port is reachable from your client host.

Available tools

fetch_account_balance

Fetches the current account balance from the linked exchange account.

fetch_deposit_address

Retrieves the deposit address for a specified cryptocurrency.

withdraw_cryptocurrency

Withdraws cryptocurrency to a given destination address.

fetch_open_positions

Fetches open positions, primarily for futures or derivatives markets.

set_trading_leverage

Sets leverage for a trading symbol, typically in futures trading.

create_spot_limit_order

Places a new limit order on the spot market.

create_spot_market_order

Places a new market order on the spot market.

create_futures_limit_order

Places a new limit order on a futures market.

create_futures_market_order

Places a new market order on a futures market.

cancel_order

Cancels an existing open order.

fetch_order_history

Fetches the history of orders (open and closed).

fetch_my_trade_history

Fetches the history of trades executed by the user.

fetch_ohlcv

Fetches historical OHLCV candlestick data for a symbol.

fetch_funding_rate

Fetches the funding rate for a perpetual futures contract.

fetch_long_short_ratio

Fetches the long/short ratio for a symbol, using exchange-specific methods.

fetch_option_contract_data

Fetches market data for options contracts.

fetch_market_ticker

Fetches the latest price ticker for a symbol.

fetch_public_market_trades

Fetches recent public trades for a symbol.

calculate_technical_indicator_tool

Fetches OHLCV data and computes a specified technical indicator (e.g., RSI, SMA, EMA, MACD, Bollinger Bands, Stochastic Oscillator, ATR).