Direct Seedance Papercraft Explainer

SkillMedia

Design a Seedance 2.0 science, education, or general-knowledge explainer using layered handmade papercraft, miniature diorama depth, physical paper mechanisms, clear knowledge beats, restrained miniature camera, tactile sound, and stable reference roles. Use for one 4–15 second clip or a longer multi-clip plan; not for flat vector cartoons, live action, or paper-texture-only styling.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Direct Seedance Papercraft Explainer skill

What this skill tells your AI

The instructions your AI receives, as published by t8mars/minimax-h3-prompt-skill-t8 in skills/direct-seedance-papercraft-explainer/SKILL.md and read by ahel’s review.

Make an abstract mechanism feel physically built and teachable. This is a T8-authored companion to the MiniMax repository-bundled papercraft-stop-motion-explainer; use the upstream Skill for H3.

Workflow

  1. Lock the topic, audience, core learning sentence, ratio, total duration, and requested deliverable.
  2. Choose one visual metaphor: pop-up journey, paper lab, layered cross-section, miniature nature theater, or mechanism board.
  3. Design only the characters that improve explanation. Define their paper construction and limited gesture system.
  4. Build the stage from foreground, midground, background, and far-background planes. Give every moving element a physical paper mechanism.
  5. Split knowledge into causal beats. A 15-second clip normally supports about three or four readable changes; longer work becomes multiple clips of at most 15 seconds.
  6. Assign reference images narrow character, scene, mechanism, or style roles and compile with references/template.md.

Seedance rules

  • Bind every recurring character/model to its declared image when supplied.
  • Describe cardboard thickness, fibers, folds, seams, tabs, brads, sliders, rotating discs, pull-tabs, layer shadows, and settling.
  • Use 镜头N only for new knowledge states; do not add exact per-shot time ranges.
  • Prefer locked, macro, top-down, slow push, or lateral parallax camera behavior—one coherent move per shot.
  • Put exact requested labels in 【...】, speech in {...}, tactile sound in <...>, and music in (...).
  • No H3 fields, H3 labels, timestamp alignment, plastic CG, liquid morphing, or physics-free motion.

Boundaries

Do not generate media before the plan and references are approved. Do not import this upstream entry into ComfyUI.

Deliver

Return the learning/metaphor lock, stage and mechanism plan, clip breakdown, Seedance prompt per clip, sound plan, and educational/material/continuity audit.

Signals

GitHub stars
217
Forks
21
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
direct-seedance-papercraft-explainer
Source
github.com/t8mars/minimax-h3-prompt-skill-t8