seedance-shotdesign-skills

English | 中文 | 日本語 | 한국어 | Español | Português | Français

Seedance Shot Design turns a rough video idea or an approved source clip into a production-ready Seedance 2.5 plan and copyable prompt. Version 3.0 is a capability-level rewrite for 30-second standard generation, 30–180 second ultra-long work, multimodal references, extension, precision editing, and industrial workflows.

What changed in 3.0

Area	Seedance 2.5 behavior
Standard generation	One continuous 4–30s generation; exactly 30s stays in standard mode by default
Ultra-long	Continuous 30–180s stories with a continuity bible, acts, sequences, and a resolved ending
Extension	Source ≤30s, add 4–30s, final ≤60s; the prompt controls only the added portion
Multimodal input	Up to 30 images, 10 videos, and 10 audios under the documented per-type duration/file constraints
Editing	Smart edit, annotation-based advanced edit, local replacement/removal, and viewpoint reconstruction
Audio	Audio-only references, voice/timbre binding, multilingual dialogue, BGM removal with speech/ambience preservation
Production modes	Creative transfer, green screen, rough/fine white model, seamless transition, multi-panel storyboard
Validation	Mode-aware duration, asset, timeline, contract, preservation, and contradiction checks
Localization	Native directing language; dialogue is explicitly bound to speaker and target language

The skill deliberately removes obsolete assumptions: no forced split above 15 seconds, no 9/3/3/12 mixed-file rule, no universal 500-character ceiling, no forced English for non-Chinese users, and no undocumented 1080p or CLI claims.

Core workflow

flowchart LR
    A[Production brief] --> B{Route by intent}
    B --> C[Generation or ultra-long]
    B --> D[Extension]
    B --> E[Editing or audio]
    B --> F[Industrial workflow]
    C --> G[Asset binding and continuity]
    D --> G
    E --> G
    F --> G
    G --> H[Mode-specific prompt contract]
    H --> I[Mode-aware validation]
    I --> J[Copyable Seedance 2.5 delivery]

The skill:

1. infers the production brief;
2. selects one primary Seedance mode;
3. loads only the relevant reference guides;
4. binds every asset to a role, scope, and exclusion;
5. designs time, action causality, and continuity state;
6. validates and delivers one self-contained prompt.

Supported routes

• standard: 全能参考 / 首尾帧, 4–30s
• ultra_long: 超长视频, 30–180s
• extension: 视频延长
• smart_edit / advanced_edit: 智能编辑 / 高级编辑 / 视频编辑
• viewpoint: 空间视角修改
• bgm: 音轨编辑 / BGM separation
• creative_transfer: 迁移创意
• green_screen: 绿幕编辑
• rough_white_model / fine_white_model: 粗颗粒 / 细颗粒白模
• seamless_transition: 视频无缝转场
• storyboard: 多宫格分镜

Installation

Place this folder in the skills directory used by your agent host. Common examples:

.claude/skills/seedance-shot-design/
~/.codex/skills/seedance-shot-design/
.cursor/skills/seedance-shot-design/

The repository currently remains available at its existing URL:

git clone https://github.com/woodfantasy/Seedance2.0-ShotDesign-Skills.git seedance-shot-design

Invoke it explicitly when desired:

Use $seedance-shot-design to design a continuous 30-second vertical suspense film.

It can also activate implicitly for Seedance prompts, storyboards, extension, editing, white-model rendering, green-screen compositing, and related shot-design requests.

Example requests

Create a 30-second 9:16 product story with three image references and one voice reference.

Continue @Video1 forward by 20 seconds from its last-frame motion, without changing the original clip.

Remove the poster inside the red box from 4–11s, reconstruct the brick wall, then remove BGM while keeping dialogue and footsteps.

Render this 24-second fine white-model animation as a realistic sci-fi hangar while preserving geometry, camera, and collision timing.

Validator

The standard library–only validator can be used locally or in CI:

python3 scripts/validate_prompt.py \
  --mode standard \
  --duration 30 \
  --resolution 720p \
  --prompt-file prompt.txt

For extension, --duration means the added duration:

python3 scripts/validate_prompt.py \
  --mode extension \
  --duration 20 \
  --source-duration 25 \
  --prompt-file extension.txt

Run the regression suite:

python3 -m unittest scripts/test_validate.py

Structure

seedance-shot-design/
├── SKILL.md
├── agents/openai.yaml
├── assets/logo.svg
├── references/
│   ├── seedance-specs.md
│   ├── workflow-router.md
│   ├── prompt-contracts.md
│   ├── long-form-storytelling.md
│   ├── multimodal-references.md
│   ├── video-editing.md
│   ├── industrial-workflows.md
│   ├── cinematography.md
│   ├── quality-anchors.md
│   ├── micro-expressions.md
│   ├── audio-tags.md
│   ├── director-styles.md
│   └── scenarios.md
└── scripts/
    ├── validate_prompt.py
    └── test_validate.py

Platform notes

• The supplied Seedance 2.5 manual documents 480p and 720p output. 720P+ appears as a UI label, not a separately documented parameter. Verify the current UI before promising anything beyond 720p.
• Stability guidance is reported as a warning, not confused with a hard upload limit.
• Prompt syntax uses the exact asset tokens shown by the user's interface, such as @图片1, @视频1, and @音频1.
• Recognizable living-person likeness, brands, copyrighted characters, and sensitive content still require appropriate authorization and platform-policy compliance.
• The skill designs and validates prompts; it does not submit a generation job or consume platform credits.

License

MIT-0