Seedance 2.0 Video Prompt Guide
Paste the following prompt into your AI chat to install this skill:
Please install @user_9cf3307c/seedance-video-prompt-zh according to https://skillhub.cn/install/skillhub.md.
About this skill
Problem
When generating videos with Seedance 2.0, the main friction is often prompt ambiguity, not model capability. A bare @video1 does not say whether to copy camera motion, action, effects, or rhythm. It is also easy to exceed the 12 file limit, and clear realistic human faces are blocked. This is especially true when reusing assets, recreating camera moves, or syncing visuals to a beat.
How it works
The skill turns the prompt into a checkable structure:
- Asset role: every @reference states a purpose, such as image 1 as first frame or video 1 camera motion.
- Timing: videos over 8 seconds use timed segments for subject, action, and scene.
- Camera and audio: add push, orbit, Hitchcock zoom, BGM, SFX, and narration.
The workflow is to confirm the video type, inventory assets, assign each file a role, write 4-15s segments, add camera language, then validate duration, file count, and face restrictions.
Boundaries
It fits ad clips, short dramas, dance cuts, explainers, and 360-degree product shots. It does not support clear realistic human face assets. When extending a video, select the added duration, not the total; reference video may increase cost; do not pack too many scenes into 4-5s.
Use Cases
- Create a 15-second e-commerce ad prompt with first frame, product images, camera moves, 360-degree shots, and BGM sync.
- Extend an existing 10-second video by 5 seconds, describing only the new action, scene, and audio bridge.
- Write an explainer animation prompt with timed shots, medical CGI style, narration tone, and sound effects.
- Replicate reference video fight choreography and transitions, using @images to set character, first frame, and last frame.
Best For
- Content producer using Seedance 2.0 to turn image, video, and audio assets into executable prompts.
- Creative lead making ad or short-drama shots that need camera moves, action, effects, and beat sync.
- Multimedia designer creating product clips that need product appearance, first/last frames, outfit, and BGM.
- Video editor producing explainers that need medical CGI style, narration tone, and sound-effect cues.
Related Skills
A ComfyUI image-generation skill that uses a five-step dialog to collect prompts, prefill templates, confirm parameters, submit API jobs, and return results.
A design-system guidance skill for Impeccable that produces token-based rules, component states, accessibility criteria, and QA checklists.
Director-level video lapian that diagnoses material precision, then produces frame evidence, director analysis, a style bible, and a showcase video.
Plan and brief Amazon MAIN, Listing, and A+ image sets from verified product facts, then return plan and image QA status.