Get the FREE Ultimate OpenClaw Setup Guide →

ccxt

CCXT MCP Server bridges the gap between AI models and cryptocurrency trading by providing a standardized interface through the Model Context Protocol. Created to empower automated trading strategies, this tool allows AI assistants like Claude and GPT to directly interact with over 100 cryptocurrency exchanges without requiring users to write comple

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio obinox04-ccxt-mcp node path/to/server.js \
  --env LOG_LEVEL="info|debug|error (default: info)" \
  --env CCXT_API_KEY="Your CCXT API key (if required by exchanges)" \
  --env CCXT_EXCHANGES="Comma-separated list of CCXT exchanges to preload (e.g., binance, kraken)" \
  --env CCXT_API_SECRET="Your CCXT API secret (if required by exchanges)" \
  --env CCXT_API_PASSWORD="Optional API password (if required by some exchanges)"

How to use

The CCXT MCP Server provides a standardized Model Context Protocol interface that lets AI assistants like Claude or GPT interact with more than 100 cryptocurrency exchanges through the CCXT library. This enables actions such as retrieving market data (tickers, order books, trades, OHLCV histories) and executing trading operations (placing, modifying, or canceling orders) in a unified, exchange-agnostic way. By abstracting away exchange-specific quirks, the server empowers AI trading strategies to reason about markets, test signals, and execute decisions across multiple venues with minimal integration effort. Once running, you can query capabilities such as available markets, supported timeframes, and account balance, and you can submit trading intents that the server will translate into CCXT calls to the chosen exchange.

To use the server's tools, provide the MCP payloads that specify the desired action (e.g., fetch price, get order book, place a limit order). The server will handle the CCXT call routing, error handling, and response formatting so that your AI model receives consistent results regardless of the target exchange. Typical workflows include fetching real-time quotes, analyzing liquidity and spreads, backtesting signals against historical OHLCV data, and executing orders when the model's confidence meets a threshold. Remember to supply necessary API keys for any exchange that requires authentication and to monitor rate limits and error responses through the server logs.

How to install

Prerequisites:

  • Node.js (v14+ recommended) and npm installed on your machine
  • Git to clone the repository
  • Optional: API keys for one or more CCXT-supported exchanges if you plan to place live orders

Install instructions:

  1. Clone the repository git clone https://github.com/Obinox04/ccxt-mcp.git cd ccxt-mcp

  2. Install dependencies npm install

  3. Configure environment variables

    • Create a .env file or export variables in your shell CCXT_EXCHANGES=binance,kraken,coinbasepro CCXT_API_KEY=your_api_key CCXT_API_SECRET=your_api_secret CCXT_API_PASSWORD=your_api_password (if required) LOG_LEVEL=info
  4. Start the MCP server npm run start

    or if you have a specific entry script configured

    node path/to/server.js

  5. Verify the server is running by sending a test MCP request or checking logs for a successful startup message.

Additional notes

Notes and tips:

  • Ensure API keys are correctly configured for exchanges you intend to trade on; some exchanges require IP whitelisting or additional permissions.
  • Be mindful of rate limits. The MCP server relies on CCXT's rate-limit handling, but aggressive queries or frequent orders may hit limits on some exchanges.
  • Use a sandbox/test environment when possible to avoid unintended live trades during development.
  • The environment variable CCXT_EXCHANGES controls which exchanges are loaded at startup; add or remove exchanges as needed.
  • If you encounter connection or authentication errors, check that your API keys are valid and that the server has network access to the exchange endpoints.
  • For debugging, set LOG_LEVEL=debug to get verbose logs of CCXT calls and MCP routing.

Related MCP Servers

Sponsor this space

Reach thousands of developers