mcp-wolframalpha

This MCP server integrates Wolfram|Alpha capabilities into your chat applications. It exposes an MCP server that interfaces with the Wolfram Alpha API so you can perform computational queries and retrieve structured knowledge within a conversation. The repository also includes an MCP-Client example that demonstrates connecting a Gemini-based LLM (via LangChain) to the MCP server, enabling real-time querying of Wolfram Alpha from within large language model workflows. A Gradio-based UI is provided to offer a user-friendly web interface for interacting with both Google AI (Gemini) and Wolfram Alpha through the MCP server. To use it, configure the MCP server (e.g., via a VSCode MCP Server or Claude Desktop) and run the provided client tooling to issue Wolfram Alpha queries through the MCP protocol. The client UI/CLI examples show how to start the Gradio UI or run the CLI to send queries to Wolfram Alpha through the MCP layer.

Prerequisites:

• Python 3.8+ installed on your system
• Git
• Access to Wolfram Alpha API (WOLFRAM_API_KEY)
• Optional: Google Gemini API key (GeminiAPI) if you plan to use the MCP client with Gemini

1. Clone the repository:

git clone https://github.com/ricocf/mcp-wolframalpha.git
cd mcp-wolframalpha

2. Create and populate environment variables (example):

# .env or export variables in shell
export WOLFRAM_API_KEY=your_wolframalpha_appid
export GeminiAPI=your_google_gemini_api_key

3. Install Python dependencies:

pip install -r requirements.txt

4. (Optional but recommended) Ensure uv is installed if you plan to use UV-based tooling:

uv sync

5. Run the MCP server (example using the provided configuration via VSCode or Claude Desktop):

# Example command from the provided MCP config:
python /path/to/src/core/server.py

6. If you want to use the MCP configuration as in VSCode or Claude Desktop, create or edit the mcp.json accordingly (see README for the WolframAlphaServer example). For VSCode MCP Server, place a file at .vscode/mcp.json that points to the Python server script as shown in the README.

Tips and notes:

• Ensure your Wolfram API key is valid and has permission for the queries you intend to perform.
• If you plan to use the MCP Client with Gemini, provide a valid GeminiAPI key; the client UI will rely on Gradio and Gemini for the frontend/backend interaction.
• The UI support uses Gradio; you can run the UI with the provided client entry points (e.g., python main.py --ui) to interact with Wolfram Alpha through the MCP server.
• For containerized deployments, Docker images are demonstrated for both the UI and the client; adapt image names and Dockerfiles as needed for your environment.
• If you encounter connection issues, verify that the MCP server is reachable from the client, and ensure the correct port and host are configured in your client setup.
• The README mentions optional VSCode MCP Server and Claude Desktop configurations; those workflows expect you to point to the Python server script path in the mcp.json configuration.