binance
Unofficial tools and server implementation for Binance's Model Context Protocol (MCP). Designed to support developers building crypto trading AI Agents.
claude mcp add --transport stdio analyticace-binance-mcp-server uvx binance-mcp-server \ --env BINANCE_API_KEY="your_api_key_here" \ --env BINANCE_TESTNET="true" \ --env BINANCE_API_SECRET="your_api_secret_here"
How to use
This MCP server provides a bridge between AI assistants and your Binance account, exposing a comprehensive set of tools to read market data, manage your account, and perform trading operations through the MCP interface. After installing the server, you can interact with Binance via a standardized set of MCP tools such as get_balance, get_ticker_price, create_order, and get_pnl. For example, you can query real-time prices, check your balances, place market or limit orders, and monitor open positions or PnL. The server supports both market data access and trading operations, making it suitable for building autonomous agents or AI-assisted trading flows.
To use it, ensure your Binance API credentials are configured in the environment variables (BINANCE_API_KEY, BINANCE_API_SECRET) and, if desired, enable the testnet via BINANCE_TESTNET. Integrate the MCP server into your existing MCP configuration (e.g., claude_desktop_config.json or mcp_config.json) by adding the binance server with the appropriate command and arguments. You can then invoke the available tools through your AI agent or MCP client, such as requesting the current ticker price for a symbol, checking your balances, or creating an order.
How to install
Prerequisites:
- Python 3.10 or newer installed on your system
- pip for Python package installation
- Optional: uv (Universal Validator) package manager for Python-based MCP servers
Installation steps:
-
Install the Binance MCP Server package from PyPI: pip install binance-mcp-server
-
Install or enable UV for running MCP servers (if you plan to use uv as shown in the config): pip install uv # or follow your preferred uv installation method
-
Verify installation by running the server directly (example, adjust if you use uvx): uv add binance-mcp-server
or if running via Python module
python -m binance_mcp_server
-
Configure environment variables for Binance access (adjust as needed): export BINANCE_API_KEY="your_api_key_here" export BINANCE_API_SECRET="your_api_secret_here" export BINANCE_TESTNET="true" # optional: use testnet for development
-
Start the MCP server through your MCP manager (as defined in mcp_config).
Note: If you prefer running via a different launcher or container, adapt the commands accordingly to match the environment (e.g., docker, node-based runners, etc.).
Additional notes
Tips and common issues:
- Always test with Binance Testnet first to avoid real funds exposure.
- Ensure your API keys have the necessary permissions (read data, trade if you intend to place orders).
- If you encounter authentication errors, double-check that BINANCE_API_KEY and BINANCE_API_SECRET are correctly exported in the environment where the MCP server runs.
- When using testnet, confirm that the server is pointed to testnet endpoints if the library supports explicit testnet configuration.
- The server exposes a broad set of tools; consult the available tools list (get_balance, get_ticker_price, create_order, etc.) to choose the right operations for your automation.
- If the MCP server is not reachable, verify network access to Binance endpoints and ensure no firewalls are blocking outgoing connections.
Related MCP Servers
shippie
extendable code review and QA agent 🚢
FinanceMCP
这是一个金融领域相关的mcp,本项目通过集成 Tushare API 和 Binance API 为语言模型(如Claude)提供全面的实时金融数据访问能力,支持股票、基金、债券、宏观经济指标、稳定币、虚拟货币等多维度金融数据分析。其中也包含了金融数据查询、财经新闻查询、国家统计局数据查询等
TradingAgents mode
TradingAgents-MCPmode 是一个创新的多智能体交易分析系统,集成了 Model Context Protocol (MCP) 工具,实现了智能化的股票分析和交易决策流程。系统通过多个专业化智能体的协作,提供全面的市场分析、投资建议和风险管理。
semantic-scholar-fastmcp
A FastMCP server implementation for the Semantic Scholar API, providing comprehensive access to academic paper data, author information, and citation networks.
mcp -ccxt
Cryptocurrency Market Data MCP Server
fast -telegram
Telegram MCP Server and HTTP-MTProto bridge | Multi-user auth, intelligent search, file sending, web setup | Docker & PyPI ready