mcp -shioaji

This MCP server exposes Shioaji trading API functionality as MCP tools that AI assistants can call. It provides tools to retrieve real-time stock prices, fetch historical K-bar data, and scan stocks by various criteria, enabling automated trading insights and data-driven decisions. You can interact with the tools by sending standardized MCP tool calls (for example, get_stock_price, get_kbars, and scan_stocks) and receive structured responses with prices, volumes, and market metrics. The server runs with the Shioaji credentials you provide, and exposes endpoints on port 8000 by default.

Prerequisites:

• Python 3.10 or higher
• uv (the Python package manager / uv is used to run the server)

Installation steps:

1. Install Python from the official site or your package manager.
2. Install uv (if not already installed):

pip install uv

3. Install the MCP Shioaji server package (the server is named mcp-server-shioaji on PyPI):

pip install mcp-server-shioaji

4. Start the MCP server using uv:

uv run mcp-server-shioaji

Notes:

• Ensure your SHIOAJI_API_KEY and SHIOAJI_SECRET_KEY are configured (via environment variables or a .env file) before starting the server.

Environment variables:

• SHIOAJI_API_KEY: Your Shioaji API key
• SHIOAJI_SECRET_KEY: Your Shioaji secret key

Configuration tips:

• The server defaults to listening on http://0.0.0.0:8000. If you need to change the port or host, adjust the server configuration or environment as provided by the Shioaji MCP implementation.
• If you use a .env file, ensure it is loaded before starting the server (some setups automatically load it).
• To add more tools, modify the server.py implementation and rebuild/install the package as needed.
• Common issues include invalid credentials, network restrictions, or missing dependencies in your Python environment.