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
Configuration
View docs{
"mcpServers": {
"lazy-dinosaur-ccxt-mcp": {
"command": "npx",
"args": [
"-y",
"@lazydino/ccxt-mcp",
"--config",
"/path/to/config.json"
]
}
}
}The CCXT MCP Server lets AI models interact with cryptocurrency exchange APIs through the Model Context Protocol (MCP). By wrapping CCXT’s broad exchange access in MCP, you enable automated trading queries, market data retrieval, and account actions to be driven by AI in a scalable, standardized way.
You connect an MCP client (such as Claude Desktop or another MCP-enabled interface) to the CCXT MCP Server and issue requests that cover market data, trading actions, and analytics. The server translates your requests into CCXT calls to supported exchanges and returns structured responses back through MCP.
Typical workflows you can perform include checking prices across exchanges, querying order books, placing and managing orders, and retrieving historical OHLCV data. You can also run trading analyses like performance metrics and risk metrics, then adjust positions using dynamic sizing and risk controls.
For authentication and account-specific actions, you provide account credentials in a structured configuration (as described in the configuration options). You can embed these accounts directly in the client’s Claude Desktop settings or reference them from a separate configuration file passed to the server via the --config option.
Prerequisites: you need Node.js and npm installed on your system.
Install the CCXT MCP Server package globally so you can run the mcp from anywhere.
# Install the package globally
npm install -g @lazydino/ccxt-mcp
# Run with default settings
ccxt-mcp
# or run without installation
npx @lazydino/ccxt-mcpYou can register the CCXT MCP Server in your MCP client by configuring a server entry named ccxt-mcp. You have two practical options for providing account information and configuration.
Option 1: Include account information directly in your client’s configuration (basic method). This embeds CCXT account details in the client configuration file.
Option 2: Use a separate configuration file that contains accounts and is referenced by the server through the --config option.
Option 1 shows account details embedded in the client configuration. The server is started via a command that uses npx to run the CCXT MCP server and includes the --config path.
{
"mcpServers": {
"ccxt_mcp": {
"command": "npx",
"args": ["-y", "@lazydino/ccxt-mcp", "--config", "/path/to/config.json"]
}
}
}Limit access to your MCP client and secure your configuration files since credentials appear in the accounts section. Consider using a separate configuration file for sensitive data and manage permissions at the file level.
If the server fails to start, verify that Node.js and npm are installed, and that the npx command resolves to the correct package. Ensure the --config path points to a valid JSON file that contains an accounts array when using the separate-configuration approach.
The CCXT MCP Server exposes a broad set of capabilities through MCP, including market data retrieval, trading actions, and comprehensive analytics. You can leverage these features through MCP-enabled AI prompts to perform exchanges operations programmatically.
Retrieve the list of available exchanges supported by CCXT MCP Server.
Query trading pairs, price data, and market info for a specific exchange.
Get real-time price information for a specific symbol across configured exchanges.
Fetch the current order book for a given symbol on a chosen exchange.
Search and retrieve historical OHLCV data for given symbols and timeframes.
Create market or limit orders on a selected exchange account.
Cancel active orders and query their status.
View account balances and trading history.
Compute daily/weekly/monthly performance metrics and win/loss statistics.
Set risk parameters, stop losses, take profit, and position sizing rules.