Skip to content
Back to skills

Direct Seedance Handdrawn Live Fusion

ASecurity

Create a Seedance 2.0 prompt for a rough hand-drawn luminous entity that physically contacts a live-action space, continuously transforms while remaining traceable, escapes through adjacent geography, and is followed by a slightly delayed handheld camera. Use for playful 4–15 second mixed-media clips; not for horror, polished CG, unrelated cuts, or simple visual filters.

  • 278 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added September 12, 2026
ai-agentsreact

Works with

  • cli

Security analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned September 12, 2026

npx -y skills add T8mars/minimax-h3-prompt-skill-T8 --skill direct-seedance-handdrawn-live-fusion --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Direct Seedance Handdrawn Live Fusion?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Direct Seedance Handdrawn Live Fusion
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/t8mars-direct-seedance-handdrawn-live-fusion/badge)](https://www.skillsdirectory.com/skills/t8mars-direct-seedance-handdrawn-live-fusion)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: direct-seedance-handdrawn-live-fusion
description: Create a Seedance 2.0 prompt for a rough hand-drawn luminous entity that physically contacts a live-action space, continuously transforms while remaining traceable, escapes through adjacent geography, and is followed by a slightly delayed handheld camera. Use for playful 4–15 second mixed-media clips; not for horror, polished CG, unrelated cuts, or simple visual filters.
---

# Direct Seedance Hand-drawn Live Fusion

Design a fresh mixed-media event around physical contact and continuous transformation. This is a T8-authored companion to the MiniMax repository-bundled `handdrawn-live-video-generator`; use the upstream Skill for H3.

## Workflow

1. Choose one everyday live-action space or a set of visibly adjacent areas.
2. Invent a new hand-drawn entity, limited palette, material language, contact object, transformation chain, route, camera reaction, and final spatial payoff.
3. Make the first causal event a readable contact between the drawn entity and a real hand/object.
4. Preserve one trace across transformations: line, tail, smear, contour, color, or rhythmic motion.
5. Let the camera notice and follow slightly late; do not keep the entity perfectly centered.
6. Build a continuous action paragraph with `references/template.md` unless a genuine cut is required.

## Seedance rules

- Describe contact geometry, resistance, deformation, release, inertia, and the real object's response.
- Use rough crayon, chalk, pencil, pastel, or dry-brush texture with uneven fill, edge chatter, and frame-redrawn variation.
- Keep one entity; do not replace it with an unrelated new creature.
- Prefer one continuous handheld move through connected space. Use `镜头N` only for real cuts.
- Use an overall duration only; never add exact per-shot seconds.
- Use `<...>` for physical sound and `(...)` for music. Do not invent speech or visible text.
- Exclude horror anatomy, threatening attack motion, smooth neon vectors, plastic CG, and perfect tracking.

## Boundaries

Do not copy the upstream example's nouns, transformation sequence, room, route, gag, or ending. Do not call generation or import into ComfyUI without a separate request.

## Deliver

Return the new concept variables, final Seedance paragraph, anti-copy list, and contact/trace/camera/medium audit.

Files in this skill

  • SKILL.md2.3 KB
  • agents/openai.yaml292 B
  • references/summary.md778 B
  • references/template.md1.9 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…