Manim MCP

Manim MCP turns natural language prompts into mathematical animations using the Manim library, the same engine behind 3Blue1Brown's YouTube videos. You describe what you want in plain English, and it writes the Manim script, executes it, and hands you back an MP4. I tested it with "show a sine wave morphing into a cosine wave" and got a clean, render-ready animation in under a minute. The real win here is skipping the Manim learning curve. Manim's Python API is powerful but fiddly, full of coordinate systems and timing functions that take hours to get right. This MCP handles that translation layer, so you can iterate on concepts without reading docs. It's genuinely useful for maths educators who want to prototype visual explanations quickly, or researchers who need a figure animated for a talk. The output quality matches what you'd get writing Manim by hand, because it is writing Manim by hand. Quirks: it's only as good as the LLM's understanding of your prompt. Vague requests get vague animations. You'll often need to refine your language or ask for specific camera angles, colours, or timing adjustments. The server doesn't handle complex multi-scene narratives well in one go, you're better off building those scene by scene. Also, rendering can be slow for high-resolution outputs, that's a Manim limitation, not this tool's fault. Who shouldn't bother: if you already know Manim well, this adds a layer of indirection you don't need. If you want drag-and-drop animation software, look elsewhere, this still requires you to think in terms of mathematical objects and transformations. But if you're comfortable describing what you want and tweaking prompts, it's a solid shortcut to publication-quality maths animations.