finance-agent

finance-agent-mcp-server exposes a wrapper around the FinanceAgent Python package to fetch real-time and public financial data from global markets without requiring API keys. It aggregates stock data from multiple trusted sources and presents information such as symbol, current price, high/low, previous close, update time, market capitalization, and P/E ratio for regions including US, HK, CN Mainland, India, Europe, and Japan. Use the MCP endpoint to spin up the server using the uv runner, then call the provided get_stock_price_global_market utility to request data for a list of symbols in a specified market. The server is designed to surface data from open sources and trusted sites like morningstar, hkex, xueqiu, marketbeat, and others through the FinanceAgent integration.

Prerequisites:

• Python 3.8+ installed on your system
• Access to install Python packages (pip)

Installation steps:

1. Clone the MCP server repository or download the finance-agent-mcp-server package.
2. Install the uv runner (uvx) so you can execute the MCP server via the uv command. Example (pipx):
   • pipx install uvx
3. Ensure the Python environment has the FinanceAgent package installed (the MCP server wraps this package). If needed, install FinanceAgent in the active environment:
   • pip install FinanceAgent
4. Prepare the directory structure and place the server script as expected by the uv command:
   • /path_to_folder/finance-agent-mcp-server/src/finance-agent-mcp-server/server.py
5. Start the MCP server using the uv runner (per the mcp_config example):
   • uv --directory /path_to_folder/finance-agent-mcp-server/src/finance-agent-mcp-server run server.py

Notes:

• Adjust the --directory path to point to where your server.py resides.
• If you prefer npx/docker/node, adapt the mcp_config accordingly, but this server is demonstrated with uv.

Tips and notes:

• The MCP server provides a wrapper to obtain realtime stock data across multiple markets using public data sources; data depend on source availability and site access.
• If you encounter authentication or access issues, verify network access to the source sites and ensure the FinanceAgent package is up to date.
• Environment variables can be used to customize data sources or proxies if supported by FinanceAgent; consider adding VARs for proxy settings or source toggles if needed (e.g., HTTP_PROXY, HTTPS_PROXY).
• The configuration path in the mcp_config entry should reflect your actual installation path. Replace /path_to_folder/finance-agent-mcp-server/src/finance-agent-mcp-server with the real path on your system.