seedance-camera
This skill should be used when the user asks for camera movement, shot scale, lens feel, framing, one-take direction, dolly, pan, tilt, push-in, handheld, aerial, macro, or camera-transfer guidance for Seedance 2.0.
- License 许可
- MIT
- 最近更新
- 2026-11-12
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- seedance-camera
- tags
- [ "camera", "cinematography", "seedance-20" ]
- license
- MIT
- metadata
- { "author": "Iamemily2050 (@iamemily2050)", "parent": "seedance-20", "updated": "2026-07-04", "version": "6.6.0", "openclaw": { "emoji": "?", "homepage": "https://github.com/Emily2040/seedance-2.0" }, "repository": "https://github.com/Emily2040/seedance-2.0" }
- description
- This skill should be used when the user asks for camera movement, shot scale, lens feel, framing, one-take direction, dolly, pan, tilt, push-in, handheld, aerial, macro, or camera-transfer guidance for Seedance 2.0.
- user-invocable
- true
03 SKILL.md 正文
seedance-camera
Use one clear camera idea per short clip unless the user asks for a multi-shot sequence. The best camera direction has a start frame, movement, speed, subject relationship, and endpoint. Avoid stacking moves that fight each other, such as drone rise, dolly-in, handheld shake, and orbit in the same five-second shot.
Load [ref:quick-ref] for prompt assembly, [ref:cinematography-shot-language] for professional shot contracts, [ref:directing-engine] to derive the move from the scene's one intention so it reinforces light, performance, and sound instead of competing, and [ref:vocab/zh] or [ref:vocab/ru] when camera wording must be multilingual.
Intent
When a user asks about camera, they are really asking where the viewer's body stands and what the viewer is made to feel from there. Camera grammar is empathy mechanics: a push-in is leaning closer, a locked frame is holding your breath. Choose the move that puts the audience where the user's feeling lives.
Camera Contract
State: shot scale, angle, movement, speed, subject relationship, and endpoint. A prompt-ready camera phrase should be physically possible and tied to the subject's action.
| Need | Strong phrase | Avoid |
|---|---|---|
| Emotional realization | slow dolly-in from medium close-up to tight close-up as Character A lowers the envelope |
dramatic cinematic zoom |
| Product reveal | controlled slider move from silhouette to front three-quarter hero angle, ending on the label |
dynamic product camera |
| Scale | low-angle crane up from boots to skyline, ending behind the character's shoulder |
epic wide moving shot |
| Instability | subtle handheld shoulder camera, small breathing sway, subject kept centered |
shaky chaotic camera everywhere |
| Precision detail | locked macro shot, focus stays on the watch gears while the second hand clicks once |
cool close-up details |
Lens and Framing Anchors
Use lens anchors only when they improve direction: 24mm wide lens for spatial energy, 35mm natural street perspective, 50mm portrait compression, 85mm shallow close-up, or macro lens for material detail. Pair lens words with subject distance and motion; do not stack lens numbers as decoration.
Move Selection
Use locked-off shots for lip-sync, product identity, and delicate VFX. Use dolly-in for discovery or realization. Use tracking for travel, pursuit, and product motion. Use orbit only when the subject can remain clear from all sides. Use crane or drone for scale, arrival, or reveal. Use handheld only when realism matters more than precision.
Continuity Rules
For multi-character scenes, anchor the camera to named tags: camera holds Character A in foreground while Character B crosses behind. For I2V, preserve the image composition unless the user explicitly wants a reframing. For reference video, state whether [Video1] transfers camera movement, action rhythm, or blocking; do not let it transfer identity unless authorized.
For complex camera movement, a video reference often works better than a long verbal stack. Use [Video1] controls camera rhythm only; do not transfer performer, room, logo, or identity.
Conflict Rule
If the user gives several incompatible moves, choose one primary camera move and put the rest into optional variants. If the shot needs multiple beats, recommend splitting into separate clips or a time-segmented prompt.
Sequence State
When sequence state is present, inherit the observed camera phase, screen direction, current clip scope, continuity locks, exact reference tags, and reserved future beats before choosing a move. A continuation camera phrase must begin from the accepted source frame or observed end state; do not restart a pan, focus pull, or tracking move unless an intentional next shot declares the reset.
Output Contract
Return the selected camera phrase, why it fits the shot, conflicts removed, fragile anchors, endpoint, and a prompt-ready integrated sentence.
04 同仓库更多 Skills
create-master
基于佛教经典文献,生成特定高僧大德的 AI 教学角色
writing-dna-skill
从至少 20 篇完整文章中蒸馏可复用的写作 DNA,分析语言、文章结构、选题逻辑、素材策略、认知框架和视觉风格,并生成 Writing-DNA.md;按该 DNA 写作时会读取全部蒸馏产物和 5 篇相关原文。用于中英文作者、账号、品牌或出版物的风格分析与一致性写作。 Distill reusable Writing DNA from at least 20 complete articles for Chinese or English authors, publications, brands, and accounts; use for language, structure, topic logic, source strategy, cognitive-frame, and visual-style analysis.
stop-slop
Remove AI writing patterns from prose. Use when drafting, editing, or reviewing text to eliminate predictable AI tells.
shuorenhua
检查和清理中英文文本里的 AI 套路,适用于“去 AI 味”“说人话”“自然一点”“别像模板”“先标问题”这类改写和审稿需求;按场景控制力度,同时保留事实、术语、语域和责任主体。
naming
Name products, SaaS, brands, open source projects, bots, and apps. Use when the user needs to name something, find a brand name, or pick a product name. Metaphor-driven process that produces memorable, meaningful names and avoids AI slop.
humanizer-zh
去除文本中的 AI 生成痕迹。适用于编辑或审阅文本,使其听起来更自然、更像人类书写。 基于维基百科的"AI 写作特征"综合指南。检测并修复以下模式:夸大的象征意义、 宣传性语言、以 -ing 结尾的肤浅分析、模糊的归因、破折号过度使用、三段式法则、 AI 词汇、否定式排比、过多的连接性短语。