MCP Service: FinancialModelingPrep (FMP) Proxy

This service acts as a secure and standardized intermediary between Claude and the third-party FinancialModelingPrep (FMP) API. It simplifies the vast FMP API into a set of logical, consistent MCP endpoints, handles authentication, and enforces standard error handling and logging.

Running the Service

1. Install Dependencies:

   pip install -r requirements.txt

2. Configure API Key: Create a .env file in the root directory of this project and add your FMP API key:

   FMP_API_KEY="YOUR_FINANCIAL_MODELING_PREP_API_KEY"
   

3. Start the Service:

   python -m mcp_fmp_proxy

   The service will be available at http://127.0.0.1:8000.

API Documentation

This service uses FastAPI, which automatically generates interactive API documentation. Once the service is running, you can access it at:

• Swagger UI: http://127.0.0.1:8000/docs
• ReDoc: http://127.0.0.1:8000/redoc

This documentation provides a complete and interactive reference for all available endpoints, including their request and response models.

Available Endpoints (Summary)

Below is a summary of the implemented endpoints. For detailed request/response formats, please refer to the auto-generated documentation linked above.

Company Data (/company)

• POST /company/profile

  • Description: Retrieves core profile information for a single company.
  • Request Body: {"symbol": "AAPL"}

• POST /company/quote

  • Description: Retrieves real-time or delayed price quotes for one or more symbols.
  • Request Body: {"symbols": ["AAPL", "MSFT"]}

Financial Statements (/financials)

• POST /financials/statement
  • Description: Retrieves income, balance sheet, or cash flow statements.
  • Request Body: {"symbol": "AAPL", "statement_type": "income", "period": "annual", "limit": 2}
  • statement_type can be income, balance_sheet, or cash_flow.
  • period can be annual or quarter.

Running Tests

To ensure the service is functioning correctly, you can run the test suite:

1. Make sure your FMP_API_KEY is correctly set in the .env file.
2. Run the tests using pytest:

   python -m pytest

Extending the Service

This service is designed to be easily extensible. To add a new set of endpoints (e.g., for "Market Performance"):

1. Create a new router file: Create mcp_fmp_proxy/routers/market_performance.py.

2. Define Models and Endpoints: In the new file, define the Pydantic models for your requests and implement the FastAPI APIRouter with the desired endpoints. Use the get_data_from_fmp function from fmp_client.py to interact with the FMP API.

3. Include the Router: In mcp_fmp_proxy/main.py, import your new router and include it in the FastAPI app:

   from .routers import market_performance
   
   # ...
   app.include_router(
       market_performance.router,
       prefix="/market",
       tags=["Market Performance"]
   )