fmp_mcp_server

This MCP server exposes a single endpoint that bridges AI tools with Financial Modeling Prep data. It offers access to company fundamentals, ratios, price data, earnings transcripts, macro indicators, news, and insider trades through a structured MCP interface. With the default STDIO mode, you can interact directly from a CLI or ChatGPT’s MCP integration; you can also run it in streamable HTTP mode to expose a remote endpoint for cloud-based clients. Available tools include company_profile, income_statement, balance_sheet, cash_flow, financial_ratios, historical_price_eod_full, earnings_call_transcript, economic_indicators, economic_calendar, stock_news_latest, stock_news_search, insider_trading_latest, ping, and when_should_i_use_fmp. Use the streamable-http option to serve an HTTP API at /mcp/ and connect via a tunnel or reverse proxy for external access.

Prerequisites:

• Python 3.9+ (recommended 3.10+)
• Access to install Python packages
• A valid FMP API key (set FMP_API_KEY)

Install steps:

1. Clone the repository: git clone https://github.com/damianboh/fmp_mcp_server.git cd fmp_mcp_server

2. Install dependencies: pip install -r requirements.txt

3. Set your FMP API key (example): export FMP_API_KEY=your_fmp_api_key_here # macOS/Linux setx FMP_API_KEY your_fmp_api_key_here # Windows

4. Run the server (default STDIO mode): python fmp_mcp_server.py

Optionally, run in HTTP mode to expose remotely: python fmp_mcp_server.py --transport streamable-http --host 0.0.0.0 --port 8000

Test health endpoint (if running in HTTP mode): curl http://127.0.0.1:8000/health

Environment and configuration tips:

• Ensure FMP_API_KEY is set; without it, some endpoints may be rate-limited or unavailable.
• In HTTP mode, the default path prefix is /mcp/; you can customize via --path.
• If you encounter rate limits with the free FMP key, consider upgrading to a paid API key.
• For local development with AI tools, STDIO mode is simplest; for cloud deployment, use streamable-http and an external tunnel or Cloudflare Ngrok alternative.
• If you plan to expose the server publicly, secure the endpoint and review rate limits and access controls provided by your hosting or tunneling solution.