MCP Wolfram Alpha (Client + Server)

Seamlessly integrate Wolfram Alpha into your chat applications.

This project implements an MCP (Model Context Protocol) server designed to interface with the Wolfram Alpha API. It enables chat-based applications to perform computational queries and retrieve structured knowledge, facilitating advanced conversational capabilities.

Included is an MCP-Client example utilizing Gemini via LangChain, demonstrating how to connect large language models to the MCP server for real-time interactions with Wolfram Alpha’s knowledge engine.

---

Features

• Wolfram|Alpha Integration for math, science, and data queries.

• Modular Architecture Easily extendable to support additional APIs and functionalities.

• Multi-Client Support Seamlessly handle interactions from multiple clients or interfaces.

• MCP-Client example using Gemini (via LangChain).

---

Installation

Clone the Repo

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

cd mcp-wolframalpha

Set Up Environment Variables

Create a .env file based on the example:

• WOLFRAM_API_KEY=your_wolframalpha_appid

• GeminiAPI=your_google_gemini_api_key (Optional if using Client method below.)

Install Requirements

pip install -r requirements.txt

Configuration

To use with the VSCode MCP Server:

1. Create a configuration file at .vscode/mcp.json in your project root.
2. Use the example provided in configs/vscode_mcp.json as a template.
3. For more details, refer to the VSCode MCP Server Guide.

To use with Claude Desktop:

{
  "mcpServers": {
    "WolframAlphaServer": {
      "command": "python3",
      "args": [
        "/path/to/src/core/server.py"
      ]
    }
  }
}

Client Usage Example

This project includes an LLM client that communicates with the MCP server.

Run as CLI Tool

• Required: GeminiAPI
• To run the client directly from the command line:

python main.py

Docker

To build and run the client inside a Docker container:

docker build -t wolframalpha -f .devops/llm.Dockerfile .

docker run -it wolframalpha