Back to Skills

manim-composer

Trigger when: (1) User wants to create an educational/explainer video, (2) User has a vague concept they want visualized, (3) User mentions "3b1b style" or "explain like 3Blue1Brown", (4) User wants to plan a Manim video or animation sequence, (5) User asks to "compose" or "plan" a math/science visualization. Transforms vague video ideas into detailed scene-by-scene plans (scenes.md). Conducts research, asks clarifying questions about audience/scope/focus, and outputs comprehensive scene specif

975stars80forksUpdated 7/11/2026

Security Assessment

Safe(100/100)
Security Score100/100

About manim-composer

Before a single line of Manim code is written, this skill turns a vague video idea into a detailed, scene-by-scene plan captured in a scenes.md file. It targets educational and explainer videos, especially in the 3Blue1Brown style, and is meant to run ahead of implementation - the companion manimce-best-practices or manimgl-best-practices skills handle the actual ManimCE or ManimGL code.

The workflow moves through three phases. First, understand the concept: research the topic deeply with web search, identify the key insights and the aha moment that makes it click, note common misconceptions, and find the narrative hook - the question the video answers and the surprising element that makes viewers care. Second, clarify with the user through targeted, adaptive questions rather than a single questionnaire, covering audience and scope (assumed background, target length, standalone or series), focus and depth (what to emphasize or skip, proof-heavy versus intuition-focused, real-world applications), and style preferences (color scheme, narration tone, visual metaphors). Third, produce scenes.md using a comprehensive structure: an overview with topic, hook, audience, length, and key insight; a narrative arc; and per-scene sections detailing duration, purpose, visual elements (mobjects, animations, camera movements), content, narration notes, and technical notes on specific Manim classes and methods. It closes with transitions and flow, a color palette, mathematical content to render, and a suggested implementation order.

Throughout, it applies 3b1b style principles: visual storytelling that shows rather than tells and transforms objects instead of replacing them, deliberate pacing that pauses for insight and varies rhythm, attention to mathematical beauty by connecting algebraic, geometric, and intuitive representations, and engagement techniques like posing questions and celebrating earned insight. Reference files on narrative patterns, visual techniques, and scene examples, plus a blank scenes template, support the process.

FAQ

Does this skill write Manim code?

No. It plans the video and outputs a scenes.md specification; implementation is handled by the manimce-best-practices or manimgl-best-practices skills.

What does it produce?

A comprehensive scenes.md file with an overview, narrative arc, per-scene breakdowns (duration, purpose, visuals, content, narration, technical notes), color palette, mathematical content, and implementation order.

What happens before any questions are asked?

The topic is researched deeply to find the key insights, the aha moment, common misconceptions, and the narrative hook.

What does it ask the user about?

Audience and scope, focus and depth, and style preferences - asked as targeted, adaptive questions rather than all at once.

Does it support both ManimCE and ManimGL?

Yes. The scene specifications are meant to be ready for implementation with either ManimCE or ManimGL.

All Files

5 files
references/narrative-patterns.md3.4 KB
View
templates/scenes-template.md3.4 KB
View
SKILL.md4.6 KB
View
references/visual-techniques.md4.8 KB
View
references/scene-examples.md5.3 KB
View

Install manim-composer

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill