A Docker-based environment for creating mathematical animations with Manim, featuring both a CLI interface and a web API with Model Context Protocol (MCP) support for AI assistants.

This project provides:

1. Containerized Manim Environment: Run Manim in an isolated, reproducible Docker environment
2. Web API: Create and manage Manim animations via HTTP requests
3. MCP Integration: Direct interaction with AI assistants like Claude
4. File Management: Upload scripts and download generated animations

• Docker and Docker Compose installed on your system

Simply pull the prebuilt image from Docker Hub:

docker pull wstcpyt/manim-docker-mcp:latest

Then run it with docker-compose:

docker compose up -d

1. Clone the repository:

   git clone https://github.com/YOUR_USERNAME/manim-docker-mcp.git
   cd manim-docker-mcp

2. Build the Docker images:

   docker compose build

Create a Python file in the animations directory (see example below), then run:

docker compose run manim -pql animations/example.py ExampleScene

Start the API server:

docker compose up -d manim-api

Access the API documentation at http://localhost:8000/docs

Create a file animations/example.py:

from manim import *

class CircleToSquare(Scene):
    def construct(self):
        circle = Circle()
        circle.set_fill(BLUE, opacity=0.5)

        square = Square()
        square.set_fill(RED, opacity=0.5)

        self.play(Create(circle))
        self.wait()
        self.play(Transform(circle, square))
        self.wait()

# CLI mode with preview (-p), low quality (-ql)
docker compose run manim -pql animations/example.py CircleToSquare

# API mode
curl -X POST "http://localhost:8000/run-manim?filepath=/manim/temp/circle_example.py&scene_name=CircleToSquare&quality=low_quality"

manim-docker-mcp/
├── animations/           # Manim animation scripts
├── app/                  # FastAPI application
├── media/                # Generated animations (CLI mode)
├── output/               # Generated animations (API mode)
├── temp/                 # Temporary files
├── uploads/              # Uploaded animation scripts
├── Dockerfile            # Docker image definition
├── docker-compose.yml    # Docker Compose configuration
└── README.md             # This file

• -p: Preview the output file
• -t: Transparent background
• --save_last_frame: Render only the last frame
• -c COLOR: Set background color

GET /list-files?directory=/manim

POST /write-file?filepath=/manim/temp/example.py

POST /run-manim?filepath=/manim/temp/example.py&scene_name=CircleToSquare

GET /download-file?filepath=/media/videos/example/480p15/CircleToSquare.mp4

Full API documentation is available at the /docs endpoint.

This project supports the Model Context Protocol (MCP), enabling AI assistants to:

1. Create Manim scripts based on natural language descriptions
2. Run animations and provide download links
3. Browse and manage generated media files

Example MCP session:

User: Create an animation showing a circle morphing into a square
AI: I'll create that for you...

The container includes a minimal LaTeX installation. Custom LaTeX can be used in animations:

formula = MathTex(r"\int_{a}^{b} f(x) \, dx = F(b) - F(a)")
self.play(Write(formula))

Modify the docker-compose.yml file to mount additional directories:

volumes:
  - ./my_custom_dir:/manim/custom

• Docker not running: Make sure Docker daemon is running
• Permission errors: The container needs write access to mounted volumes
• Missing media: Check the correct output directory (media/ for CLI, output/ for API)

If you encounter issues:

1. Check the Manim documentation
2. Search existing GitHub issues
3. Create a new issue with details about your problem

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

• Manim Community for the amazing animation engine
• FastAPI for the web framework
• Model Context Protocol for AI integration