siqiliu-tsinghua

siqiliu-tsinghua / mma-mcp

Public

A MCP server that wraps a local Wolfram Engine, enabling AI assistants (Claude, ChatGPT, etc.) to perform symbolic math, numerical analysis, and data visualization via Wolfram Language.

11
1
100% credibility
Found Apr 16, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A server that connects AI assistants like Claude and ChatGPT to a user's local Wolfram Engine for advanced math computations, analysis, and visualizations.

How It Works

1
📰 Discover the Math Booster

You learn about a free tool that lets your AI chat buddy like Claude tackle super tough math problems using Wolfram's powerful engine.

2
📥 Grab the Helper

Download the simple helper program to your computer and check that you have your own Wolfram math software ready.

3
🔧 Quick Setup

Follow the easy steps to link the helper to your Wolfram program and create basic safety rules.

4
🚀 Turn It On

Hit start, and your math bridge comes alive, ready to power up your AI chats.

5
🔗 Connect Your AI

Add one line to your AI chat settings, like Claude or ChatGPT, so it knows about this new math superpower.

6
💬 Chat and Solve

Ask your AI things like 'integrate this function' or 'plot this graph,' and watch it use Wolfram magic behind the scenes.

✨ Perfect Results

Get instant, accurate math solutions, pretty graphs, and explanations that make you feel like a genius.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is mma-mcp?

mma-mcp is a Python MCP server that hooks your local Wolfram Engine into AI assistants like Claude and ChatGPT. It exposes two tools—evaluate for text results and evaluate_image for PNG plots—unlocking symbolic math, numerical analysis, and data viz via Wolfram Language. You supply your own licensed Engine; it runs stdio for local clients or HTTP with OAuth for web UIs.

Why is it gaining traction?

Robust security filters expressions pre-kernel with blacklist/whitelist modes across 29 capability groups, plus per-client RBAC to isolate Claude from ChatGPT on one machine. CLI commands like `mma-mcp init`, `setup`, and `serve` use uv for instant config and kernel pooling. Handles long Claude.ai computes via heartbeats; pairs well with mcp github copilot vscode or mcp server claude setups.

Who should use this?

AI engineers prompting LLMs for exact derivatives, integrals, or plots without hallucinations. Data scientists in mcp github python workflows needing local compute. Researchers evaluating mcp server examples for home assistant or n8n integrations.

Verdict

Solid for Wolfram owners despite 11 stars and 1.0% credibility—great docs, pytest coverage, MIT license. Run `uv sync` and test locally; skip if no Engine license.

(178 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.