mcp -ccxt
Cryptocurrency Market Data MCP Server
claude mcp add --transport stdio nayshins-mcp-server-ccxt python /path/to/crypto_server.py
How to use
This MCP server provides real-time and historical cryptocurrency market data by integrating with CCXT exchanges. It exposes a set of tools that you can call from an LLM (such as Claude) to fetch prices, market summaries, historical OHLCV data, and trading volumes across multiple supported exchanges. The server is designed to be easily connected to Claude Desktop or other MCP clients by registering the server under an mcpServers entry and pointing the client to the Python script that runs crypto_server.py. Once connected, you can query for current prices, view market summaries, list available exchanges, retrieve historical price data, and analyze price changes and volumes across a range of trading pairs and timeframes. Tools are described in the “Available Tools” section of the README and are intended to be used through natural language prompts to return structured results.
To use the server with Claude Desktop (or any MCP-compatible client), ensure your client configuration points to the python-based server entry script as shown in the example. After connecting, invoke the available tools by their names (e.g., get-price, get-market-summary) with the appropriate parameters like trading pair and exchange. The results will be returned in a consistent, machine-readable format suitable for further analysis or display in your workflow.
How to install
Prerequisites:
- Python 3.9 or higher
- Internet access to install dependencies and reach exchange APIs
- Optional: Smithery CLI for one-click installation (if using Smithery)
Manual installation steps:
- Create and activate a Python environment (recommended):
- macOS/Linux: python -m venv venv && source venv/bin/activate
- Windows: python -m venv venv && .\venv\Scripts\activate
- Install required packages (ccxt and MCP SDK):
- pip install mcp ccxt
- Save the server script (crypto_server.py) at a suitable path on your machine. The server script is expected to implement the MCP interface and expose the tools described in the README.
- Run the server:
- python /path/to/crypto_server.py
Optional Smithery-based installation:
- Install Smithery CLI and install the MCP server package for Claude: npx -y @smithery/cli install mcp-server-ccxt --client claude
- Start (or configure) the server through Smithery as directed by the Smithery UI/commands.
Additional notes
Tips and common issues:
- Ensure Python 3.9+ is used to match the listed dependencies.
- If you encounter connection issues with an exchange, verify that the exchange API keys (if required) are correctly configured and that the exchange is online.
- For rate limiting, consider distributing requests across multiple exchanges or implementing per-exchange delays.
- When troubleshooting data formatting, ensure trading pairs are provided in the standard format like BTC/USDT (not BTCUSDT).
- If running behind a firewall or proxy, verify network access to exchange APIs and MCP endpoints.
- When connecting to Claude Desktop, ensure the path to crypto_server.py in the client config matches your actual file location.
Related MCP Servers
gpt-researcher
An autonomous agent that conducts deep research on any data using any LLM providers.
sonarqube
SonarQube MCP Server
skillz
An MCP server for loading skills (shim for non-claude clients).
beemcp
BeeMCP: an unofficial Model Context Protocol (MCP) server that connects your Bee wearable lifelogger to AI via the Model Context Protocol
mcp -js
MCP server that exposes YepCode processes as callable tools for AI platforms. Securely connect AI assistants to your YepCode workflows, APIs, and automations.
Common_Chronicle
Common Chronicle turns messy context into structured, sourced timelines.