financial-reports

This MCP server provides access to the Financial Reports API, exposing a set of tools to search for companies, retrieve company details, access the latest filings, and explore industry classifications. Core tools include: search_companies to find companies by name, country, or sector; get_company_detail to fetch detailed company information; get_latest_filings to retrieve recent filings; get_filing_detail for specifics about a filing; and a suite of get_industry and list_* helpers to explore GICS sectors, industries, and groups. The server is designed to be run via Docker or locally with Python, and can be consumed by clients using the MCP protocol through stdio transport. To run with Docker, you’ll typically pass API credentials and the API base URL as environment variables. When using Claude Desktop or Smithery CLI, you can embed these secrets in the transport arguments to ensure they’re available to the container at runtime. The available tools are all defined in the server’s main entrypoint and can be invoked by an MCP client to fetch structured data or OpenAPI schemas.

Prerequisites:\n- Docker (recommended) or a Python 3.11+ environment.\n- If running locally: Python virtual environment tooling (optional but recommended).\n\nOption A: Docker (recommended)\n1) Build the Docker image (once):\nbash\ndocker build -t financial-reports-mcp .\n\n2) Run the MCP server with required environment variables:\nbash\ndocker run --rm -i \\\n -e API_KEY=your_api_key_here \\\n -e API_BASE_URL=https://api.financialreports.eu/ \\\n -e MCP_TRANSPORT=stdio \\\n financial-reports-mcp:latest\n\n\nOption B: Local Python (development/testing)\n1) Create and activate a virtual environment:\nbash\npython -m venv venv\n# Windows\venv\Scripts\activate\n# macOS/Linux\nsource venv/bin/activate\n\n2) Install dependencies:\nbash\npip install -r requirements.txt\n\n3) Run the server: (choose uv or plain python)\nbash\n# Plain Python entrypoint\npython -m src.financial_reports_mcp\n\n# With uv if available\nuv run src/financial_reports_mcp.py\n\nTip: If you use uv, it will automatically load a .env file from the project root.\n

Environment and configuration tips:\n- Create a .env file to manage API keys and base URL locally, then reference these values in your docker run or Python startup.\n- The MCP_TRANSPORT variable should be set to stdio when using Claude Desktop or other clients that rely on standard input/output.\n- Secrets should be passed securely when using Claude Desktop; the recommended approach in this repo is to pass secrets via -e arguments in the Docker run configuration to ensure Docker receives them reliably across clients.\n- If you see ModuleNotFoundError, ensure dependencies are installed (pip install -r requirements.txt) and that the Docker image is built with the correct context.\n- The examples/ directory contains ready-to-use configs for Claude Desktop and Smithery CLI, which can help you bootstrap client configurations quickly.\n- For debugging, check logs emitted to the console when running locally, or use docker logs <container> when running in Docker.\n