T8mars/minimax-h3-prompt-skill-T8 / skills/direct-seedance-music-typography
direct-seedance-music-typography
VerifiedPlan Seedance 2.0 music-video clips where performance, camera, scene changes, spatial lyric typography, and transitions respond to a locked music window. Use for 4–15 second hooks or longer multi-clip MVs with explicit audio/lyric ownership, subject and scene references, beat-led events, and assembly handoffs; not for ordinary caption cleanup or unlicensed lyric copying.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.3 KB |
| agents/openai.yaml | config | 269 B |
| references/summary.md | markdown | 721 B |
| references/template.md | markdown | 1.9 KB |
references/template.md
View on GitHub# Seedance music-typography template
## Master lock
```text
Master audio: 音频1 -> exact music/voice window
Lyrics: [user-supplied / original approved / no lyrics]
Performer@图片1: [identity and wardrobe]
Scene@图片2: [space, light, palette]
Typography style@图片3: [font texture/layout/motion only]
Global look: [grain, color, lighting, lens]
```
## Beat-event plan
Keep external editorial timing in a separate plan. Each event records:
- lyric or musical cue;
- performer action and mouth/breath/gaze behavior;
- one spatial typography event;
- one camera impulse or cut;
- transition handoff.
## Prompt skeleton
```text
音频1提供全片音乐、节奏和人声时序。表演者@图片1保持身份、服装和面部比例,场景@图片2保持空间、光线与色调,图片3只提供文字包装风格。
镜头1:表演者随[opening cue]完成[visible performance],演唱/说出{exact lyric};空间文字【exact lyric】在不遮挡眼睛和嘴部的位置以[one motion]出现,镜头[one coherent behavior]。
镜头2:在[snare/bass/vocal accent]发生[hard cut or match move],动作延续到新构图;表演者完成[next phrase],只出现一个新的【exact visible lyric】事件。
镜头3:[payoff performance and final text state],动势与音乐尾部连续收束。
约束:歌词、口型和可见文字逐字一致;每镜头最多一个主文字事件;人物、服装、颗粒、色调、光向稳定;无普通字幕条、无多余文案Logo水印。
```
## Audit
- Every performed and visible word comes from the locked lyric source.
- Audio, character, scene, and typography references do not cross-contaminate roles.
- Cuts land on meaningful musical or vocal events without breaking an active phrase arbitrarily.
- Multi-clip handoffs preserve master-audio position externally; prompts contain no exact shot times.
- No H3 grammar remains.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |