manim-composer_skill

This skill transforms vague video ideas into a detailed scene-by-scene plan for Manim animations, guiding research, questions, and implementation cues.
  • Python

390

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill adithya-s-k/manim_skill --skill manim-composer

  • SKILL.md4.6 KB

Overview

This skill turns a vague math or science explanation idea into a detailed, scene-by-scene plan optimized for Manim-style animations. It researches the topic, identifies the pedagogical hook and common misconceptions, then produces a comprehensive scenes.md with visual, narration, and technical notes. Use this to finalize the narrative and visual design before writing any Manim code.

How this skill works

I first research the concept to find the central insight and likely “aha” moment. Then I ask targeted clarifying questions about audience, length, focus, and style. Finally I output a structured scenes.md: overview, narrative arc, per-scene visual elements, narration cues, technical hints, color palette, math assets, and recommended implementation order.

When to use it

  • You have a rough idea for an educational/explainer video and need a concrete plan
  • You want a 3Blue1Brown-style visualization or explicitly mention "3b1b style" or "explain like 3Blue1Brown"
  • You need a scene-by-scene breakdown before writing ManimCE or ManimGL code
  • You want to ensure pedagogical pacing, visual continuity, and progressive revelation
  • You want to convert a mathematical intuition into sequences of visual metaphors

Best practices

  • Answer a few targeted clarifying questions about audience level, target length, and desired depth before planning
  • Prioritize one clear "hook" and one core insight to avoid scattered narratives
  • Design scenes to transform objects rather than swapping unrelated visuals for continuity
  • Build complexity progressively: start concrete, then generalize with layered visuals
  • Include narration notes and pauses around key insights to allow viewer absorption
  • Specify technical notes for tricky transitions and list needed Manim classes/mobjects

Example use cases

  • Turn a sketchy idea like "explain eigenvectors" into a 10–12 minute scene plan with visual metaphors and equations
  • Plan a short 5–7 minute intuition-focused piece on gradient descent with camera movements and animation beats
  • Create a series outline and per-episode scenes for a multi-part topic (e.g., Fourier series)
  • Produce a scene-level breakdown emphasizing a single aha moment for classroom use
  • Generate a production order and asset list so animators can implement scenes in ManimCE or ManimGL

FAQ

No. I prepare the scene-by-scene specification and technical notes. Use a separate implementation skill for ManimCE or ManimGL code.

What input do you need from me?

Provide the topic, target audience, approximate length, and any style or color preferences; I will ask follow-ups if needed.

Can you handle non-mathematical visualizations?

Yes, as long as the visualization benefits from staged, pedagogical animation and clear visual metaphors.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational