Wolfram Alpha MCP Server

# Wolfram Alpha MCP Server [](https://smithery.ai/server/@henryhawke/wolfram-llm-mcp) A Model Context Protocol (MCP) server that provides access to Wolfram Alpha's computational knowledge engine through natural language queries. ## Features - 🧮 **Mathematical Calculations**: Solve equations, perform complex calculations, and work with mathematical formulas - 🔬 **Scientific Computing**: Access physics, chemistry, and scientific data - 🌍 **Geographic Information**: Get information about countries, cities, and geographic features - 📚 **Knowledge Base**: Query facts about history, art, astronomy, and more - 📊 **Data Analysis**: Generate plots, analyze datasets, and perform statistical calculations - 🔄 **Multiple Interpretations**: Handle ambiguous queries with assumption-based clarification ## Tools ### `wolfram_query` Query Wolfram Alpha for computational, mathematical, scientific, and factual information. This has been invaluable to me as LLMs have such a tough time with math computation so Wolfram Alpha comes in so handy. **Parameters:** - `input` (required): The natural language query to send to Wolfram Alpha - `maxchars` (optional): Maximum number of characters in the response (default: 6800) - `assumption` (optional): Assumption to use when Wolfram Alpha provides multiple interpretations - `units` (optional): Unit system preference (e.g., "metric", "imperial") - `currency` (optional): Currency preference for financial calculations - `countrycode` (optional): Country code for localized results - `languagecode` (optional): Language code for results - `location` (optional): Location for location-specific queries - `timezone` (optional): Timezone for time-related calculations ### `wolfram_query_with_assumptions` Query Wolfram Alpha with specific assumptions when the initial query returns multiple interpretations. **Parameters:** - `input` (required): The exact same input from the previous query - `assumption` (required): The assumption value to use from the previous query result - `maxchars` (optional): Maximum number of characters in the response (default: 6800) ## Setup ### Prerequisites 1. **Node.js**: Version 18.0.0 or higher 2. **Wolfram Alpha App ID**: Get one from the [Wolfram Alpha Developer Portal](https://developer.wolframalpha.com/) ### Installing via Smithery To install wolfram-llm-mcp for Claude Desktop automatically via [Smithery](https://smithery.ai/server/@henryhawke/wolfram-llm-mcp): ```bash npx -y @smithery/cli install @henryhawke/wolfram-llm-mcp --client claude ``` ### Installation 1. Clone this repository: ```bash git clone <repository-url> cd wolfram-alpha-mcp-server ``` 2. Install dependencies: ```bash npm install ``` 3. Build the project: ```bash npm run build ``` ### Configuration Set your Wolfram Alpha App ID as an environment variable: ```bash export WOLFRAM_ALPHA_APP_ID=your_app_id_here ``` Or create a `.env` file: ``` WOLFRAM_ALPHA_APP_ID=your_app_id_here ``` ### Getting a Wolfram Alpha App ID 1. Visit the [Wolfram Alpha Developer Portal](https://developer.wolframalpha.com/) 2. Sign up for a Wolfram ID or log in 3. Go to the "My Apps" tab 4. Click "Sign up to get your first AppID" 5. Complete the survey and create your app 6. Copy your App ID for use with this MCP server ## Usage ### Running the Server ```bash npm start ``` ### Example Queries - **Mathematics**: "solve x^2 + 5x + 6 = 0" - **Physics**: "speed of light in vacuum" - **Chemistry**: "molecular weight of caffeine" - **Geography**: "population of Tokyo" - **Astronomy**: "distance to Andromeda galaxy" - **History**: "when was the Eiffel Tower built" - **Unit Conversion**: "convert 100 fahrenheit to celsius" - **Financial**: "apple stock price" ### Handling Multiple Interpretations When Wolfram Alpha provides multiple interpretations of a query, the server will include available assumptions in the response. You can then use the `wolfram_query_with_assumptions` tool to get more specific results. ## Best Practices 1. **Query Optimization**: Convert complex questions to simplified keyword queries when possible - Instead of: "how many people live in France" - Use: "France population" 2. **Language**: Send queries in English only; translate non-English queries before sending 3. **Mathematical Notation**: Use proper mathematical notation - Use `6*10^14` instead of `6e14` - Use single-letter variable names with optional subscripts 4. **Units**: Include spaces between compound units (e.g., "Ω m" for "ohm\*meter") ## Error Handling The server provides comprehensive error handling for common issues: - **Invalid App ID**: Check your `WOLFRAM_ALPHA_APP_ID` environment variable - **Uninterpretable Input**: The query cannot be understood by Wolfram Alpha - **Network Issues**: Connection problems with the Wolfram Alpha API - **Timeouts**: Query too complex or service temporarily unavailable ## Development ### Scripts - `npm run build`: Compile TypeScript to JavaScript - `npm run dev`: Watch mode for development - `npm start`: Run the compiled server ### Project Structure ``` src/ ├── index.ts # Main server implementation package.json # Package configuration and MCP metadata tsconfig.json # TypeScript configuration README.md # This file ``` ## API Reference This server implements the [Model Context Protocol (MCP)](https://modelcontextprotocol.io/) specification and uses the [Wolfram Alpha LLM API](https://products.wolframalpha.com/llm-api/documentation). ## License MIT License ## Contributing Contributions are welcome! Please feel free to submit a Pull Request.