An agentic skills framework & software development methodology that works.
Manim skills for Claude Code, Cursor, and other agents
This repository packages Manim guidance as reusable agent skills for two incompatible versions: Manim Community Edition and ManimGL. Each skill includes best practices, code examples, and rules files that help an agent choose the right API, scene type, animation, and rendering workflow.
Builders who use Claude Code, Cursor, Copilot, or other agents to create Manim animations.
You can generate more accurate Manim scenes with the right version, commands, and patterns from the start.
What it does
Manim Community Edition skill
Guides for `from manim import *`, the `manim` CLI, scene structure, mobjects, animations, text, and LaTeX.
ManimGL skill
Guides for `from manimlib import *`, the `manimgl` CLI, InteractiveScene, camera control, 3D scenes, and interactive workflows.
Version-specific rules files
Separate `rules/` markdown files cover topics like animations, scenes, camera, 3D, and interactive use.
Tested examples
The `tests/manimce` and `tests/manimgl` folders verify that the code examples in the skills still work.
Skills.sh install flow
The repo is organized so agents can install the skills directly with `npx skills add`.
How to get it
- 1macOS
brew install ffmpeg
- 2Ubuntu/Debian
sudo apt update sudo apt install ffmpeg
- 3macOS
brew install mactex
- 4Ubuntu/Debian
sudo apt install texlive-full
- 5Run
# Using pip pip install manim # Using uv (recommended for this project) uv pip install manim # Verify installation manim --version
- 6Run
# Using pip pip install manimgl # Using uv (recommended for this project) uv pip install manimgl # Verify installation manimgl --version
README
Manim Skills Repository
β‘ Quick Start: Add both Manim skills to your AI agent instantly:
npx skills add adithya-s-k/manim_skill
A comprehensive collection of best practices, patterns, and examples for both Manim Community Edition and ManimGL (3Blue1Brown's version). This repository provides battle-tested code examples and guidelines for creating mathematical animations.
https://github.com/user-attachments/assets/3cd398b7-7cc6-43c1-a6e9-20077be6b009
π About the Two Versions
Manim Community Edition (manim)
- Repository: https://github.com/ManimCommunity/manim
- Focus: Community-maintained, stable, well-documented
- Best For: Production use, educational content, collaborative projects
- Command:
manimCLI - Import:
from manim import *
ManimGL (manimgl)
- Repository: https://github.com/3b1b/manim
- Focus: Grant Sanderson's (3Blue1Brown) original version with OpenGL rendering
- Best For: Interactive development, 3D scenes, rapid prototyping
- Command:
manimglCLI - Import:
from manimlib import *
Important: These are separate, incompatible frameworks. Code written for one will not work with the other without modifications.
π Installation
Prerequisites (Both Versions)
- Python 3.7+ - Required
- FFmpeg - For video encoding
- LaTeX - For mathematical typesetting (TeX Live, MiKTeX, or MacTeX)
Install FFmpeg
macOS:
brew install ffmpeg
Ubuntu/Debian:
sudo apt update
sudo apt install ffmpeg
Windows: Download from https://ffmpeg.org/download.html and add to PATH
Install LaTeX
macOS:
brew install mactex
Ubuntu/Debian:
sudo apt install texlive-full
Windows: Install MiKTeX from https://miktex.org/download
Installing Manim Community Edition
# Using pip
pip install manim
# Using uv (recommended for this project)
uv pip install manim
# Verify installation
manim --version
Documentation: https://docs.manim.community/
Installing ManimGL
# Using pip
pip install manimgl
# Using uv (recommended for this project)
uv pip install manimgl
# Verify installation
manimgl --version
Additional macOS (ARM) requirement:
arch -arm64 brew install pkg-config cairo
π Skills.sh Integration
This repository provides two AI Agent Skills that can be installed with a single command using skills.sh:
Install with npx (Recommended)
# Install Manim Community Edition best practices
npx skills add adithya-s-k/manim_skill/skills/manimce-best-practices
# Install ManimGL best practices
npx skills add adithya-s-k/manim_skill/skills/manimgl-best-practices
# Or install both
npx skills add adithya-s-k/manim_skill/skills/manimce-best-practices adithya-s-k/manim_skill/skills/manimgl-best-practices
What are Skills?
Skills are reusable capabilities for AI coding agents. Once installed, your AI assistant (like Claude, GitHub Copilot, or Cursor) automatically gains access to:
- β Domain-specific best practices
- β Working code examples
- β Common patterns and anti-patterns
- β Framework-specific knowledge
The skills follow the Agent Skills open standard and work across multiple AI tools.
When Skills Activate
manimce-best-practices - Automatically loads when:
- You import
from manim import * - You use the
manimCLI command - You work with Scene classes, mathematical animations, or LaTeX rendering
- You create educational videos or visual explanations with Manim Community
manimgl-best-practices - Automatically loads when:
- You import
from manimlib import * - You use the
manimglCLI command - You work with InteractiveScene, 3D rendering, or camera frame control
- You use interactive mode with
.embed()orcheckpoint_paste()
π Using This Repository
Repository Structure
manim_skill/
βββ skills/
β βββ manimce-best-practices/ # Manim Community Edition skills
β β βββ SKILL.md # Skill metadata
β β βββ rules/ # Individual best practice guides
β β βββ animations.md
β β βββ scenes.md
β β βββ mobjects.md
β β βββ ...
β β
β βββ manimgl-best-practices/ # ManimGL skills
β βββ SKILL.md
β βββ rules/
β βββ animations.md
β βββ 3d.md
β βββ camera.md
β βββ ...
β
βββ tests/
βββ manimce/ # Tests for Community Edition
βββ manimgl/ # Tests for ManimGL
What's Inside Each Skill File?
Each .md file contains:
- Best practices for specific Manim features
- Working code examples (all tested!)
- Common patterns and use cases
- Pitfalls to avoid
- API differences between versions
π― Quick Start Examples
Manim Community Edition
from manim import *
class BasicExample(Scene):
def construct(self):
circle = Circle()
circle.set_fill(BLUE, opacity=0.5)
circle.set_stroke(BLUE_E, width=4)
self.play(Create(circle))
self.wait()
Run it:
manim -pql scene.py BasicExample
# -p: preview after rendering
# -q: quality (l=low, m=medium, h=high)
ManimGL
from manimlib import *
class BasicExample(InteractiveScene):
def construct(self):
circle = Circle()
circle.set_fill(BLUE, opacity=0.5)
circle.set_stroke(BLUE_E, width=4)
self.play(ShowCreation(circle))
self.wait()
Run it:
manimgl scene.py BasicExample --write_file
# --write_file: save video output
# -s: skip to last frame
# -w: write file without opening
π§ͺ Running Tests
This repository includes comprehensive tests to ensure all code examples work correctly.
Test Manim Community Edition Skills
# Test all files
uv run python tests/manimce/test_all_skills.py
# Test specific file
uv run python tests/manimce/test_all_skills.py animations.md
# Run with multiple workers (faster)
uv run python tests/manimce/test_all_skills.py -j 4
Test ManimGL Skills
# Test all files
uv run python tests/manimgl/test_all_skills.py
# Test specific file
uv run python tests/manimgl/test_all_skills.py 3d.md
# Run with multiple workers (use caution - can cause OOM)
uv run python tests/manimgl/test_all_skills.py -j 4
Note: Parallel testing with many workers can cause out-of-memory errors. Use 4-6 workers max, or test files individually.
π Key Differences Between Versions
| Feature | Manim Community | ManimGL |
|---|---|---|
| Import | from manim import * | from manimlib import * |
| CLI Command | manim | manimgl |
| Scene Base Class | Scene, MovingCameraScene | Scene, InteractiveScene |
| Creation Animation | Create() | ShowCreation() |
| Text Class | Text(), MathTex() | Text(), Tex() |
| 3D Rendering | Limited | Full OpenGL support |
| Interactive Mode | No | Yes (-se flag, .embed()) |
| Camera Control | MovingCameraScene | self.camera.frame |
| Configuration | Python config | YAML files |
| Color Constants | Same | Same + variations (e.g., BLUE_A, BLUE_E) |
π Exploring Skills
For Manim Community Edition:
Start with these guides in skills/manimce-best-practices/rules/:
- scenes.md - Scene structure and lifecycle
- animations.md - Basic animation patterns
- mobjects.md - Creating and manipulating objects
- colors.md - Color systems and styling
- text.md - Text and LaTeX rendering
For ManimGL:
Start with these guides in skills/manimgl-best-practices/rules/:
- scenes.md - Scene types and InteractiveScene
- animations.md - Animation fundamentals
- camera.md - Camera movement and 3D orientation
- 3d.md - 3D object creation and rendering
- interactive.md - Interactive development workflow
π€ Contributing
Found an issue with an example? Want to add a new best practice?
- Ensure your code example works with the target Manim version
- Add it to the appropriate skill file
- Run the tests to verify:
uv run python tests/<version>/test_all_skills.py <filename> - Submit a pull request
π License
This repository is licensed under the MIT License - see the LICENSE file for details.
Note: This license applies to the educational materials and code examples in this repository. The underlying Manim frameworks (Manim Community Edition and ManimGL) have their own respective licenses.
π Resources
Manim Community Edition
- Documentation: https://docs.manim.community/
- Discord: https://www.manim.community/discord/
- GitHub: https://github.com/ManimCommunity/manim
ManimGL
- Documentation: https://3b1b.github.io/manim/
- GitHub: https://github.com/3b1b/manim
- Tutorial Videos: Grant's YouTube channel
General
- 3Blue1Brown: https://www.youtube.com/@3blue1brown
- Manim Community: https://www.manim.community/
β οΈ Troubleshooting
Common Issues
"Command not found: manim/manimgl"
- Verify installation:
pip list | grep manim - Check PATH configuration
- Try:
python -m manimorpython -m manimlib
LaTeX errors
- Install full LaTeX distribution (not basic)
- ManimCE: Try
manim --tex_template <template>with different templates - ManimGL: Check
custom_defaults.ymlfor LaTeX configuration
Video won't play
- Install media codecs for your OS
- Try different quality settings (
-ql,-qm,-qh) - Check FFmpeg installation:
ffmpeg -version
Out of Memory (parallel tests)
- Reduce worker count:
-j 2or-j 4 - Test files individually
- Close other applications
Import errors / wrong version
- Check you're importing the right version:
from manim import *β Manim Communityfrom manimlib import *β ManimGL
- Uninstall conflicting versions:
pip uninstall manim manimgl manimlib - Reinstall the version you need
π Acknowledgments
This repository exists thanks to the incredible work of:
Grant Sanderson (3Blue1Brown)
Creator of the original Manim animation engine and the 3Blue1Brown YouTube channel. Grant's pioneering work in mathematical visualization has inspired millions of learners worldwide and created an entirely new paradigm for explaining complex concepts through programmatic animation. His commitment to open-source education and visual storytelling has fundamentally changed how mathematics is taught and understood.
Website: https://www.3blue1brown.com/ YouTube: https://www.youtube.com/@3blue1brown Manim (ManimGL): https://github.com/3b1b/manim
The Manim Community
The dedicated team and contributors who maintain Manim Community Edition, ensuring the framework remains accessible, well-documented, and actively developed. Their tireless efforts in creating comprehensive documentation, managing community support, and continuously improving the codebase have made mathematical animation accessible to educators, students, and creators everywhere.
Website: https://www.manim.community/ GitHub: https://github.com/ManimCommunity/manim Discord: https://www.manim.community/discord/
Standing on the Shoulders of Giants
Both frameworks represent countless hours of development, documentation, community support, and creative problem-solving. This repository simply aims to organize and share knowledge about these powerful tools. All credit for the underlying technology goes to Grant Sanderson and the Manim Community contributors.
Thank you for making mathematical beauty programmable and accessible to all. πβ¨
Files in the repo
- skills
- tests
- .gitignore
- LICENSE
- README.md
Discussion (0)
Ask about usage, or say what you built with itSign in to join the discussion.
No comments yet. Be the first to say what this is good for.
More skills

Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
Topic in, narrated explainer video out. A Claude Code / Codex skill that turns any topic into a black-canvas motion-graphics explainer video with TTS voiceover, subtitles and a chapter progress bar. Chinese or English; every frame drawn in code with Remotion.
Public repository for Agent Skills
Open-source AI job search: scan job portals, evaluate listings into a structured A-H report with a global 1-5 score, tailor your CV, track applications β runs locally in your AI coding CLI (Claude Code, Codex, OpenCode, Antigravityβ¦)

Production-grade engineering skills for AI coding agents.