Seedance 2.0 Video Prompt Writer
Turns an idea into a timecoded multi-shot Seedance 2.0 prompt with camera moves, continuity, reference roles, and synced sound cues.
Before & after
Same prompt, same agent — the only difference is the skill.
Here's an enhanced video prompt for a watch floating in water: "A luxury watch floating gracefully in crystal clear water, sunlight filtering through, bubbles rising, slow motion, cinematic, high quality, 4k" You could add more detail about: - Camera angle and movement - Lighting style - The mood or atmosphere - Specific colors More detail generally produces better results. Want me to try a few variations?
Rich, structured prompt for Seedance 2.0 — multi-angle, multi-shot, with camera movement and sound design, so one idea becomes a studio-quality shot instead of a flat description. READY TO PASTE "Luxury dive watch suspended in still, backlit water. SHOT 1 (0-3s): macro push-in on the dial, shallow focus, a single air bubble drifts past the crystal. Light rakes from upper-left, caustics dancing across the bezel. SHOT 2 (3-6s): slow orbit around the case, water clarity revealing the crown and lug detail, fine particulate suspended in the beam. SHOT 3 (6-9s): pull back to reveal the watch turning slowly, refracted light fanning below it. Cool teal water grade, warm key light on the metal. SOUND: muffled underwater ambience, a low resonant tone rising as the reveal lands, one soft chime on the final frame." WHY THE STRUCTURE MULTI-SHOT — three framed beats, not one static description. Seedance renders sequence, so giving it a sequence is what produces a shot that moves. CAMERA MOVEMENT named per shot — push-in, orbit, pull-back. "Cinematic" means nothing; a named move means something. SOUND DESIGN — the audio arc (ambience → rising tone → chime) is what makes it feel produced rather than generated. Give me the idea and I'll write the full multi-shot prompt with sound.
About this skill
Turn a one-line idea into a Seedance 2.0 prompt that comes back as a coherent short piece: clear shots, readable camera moves, a product that does not morph, and sound that lands on the action.
What it does
- Structures the prompt: global style line, a reusable continuity block, timecoded shots, transitions, and audio
- Keeps shot count realistic for the length and limits each shot to one action and one camera move
- Assigns a clear role to each reference file (exact product, character, camera motion) with syntax marked to verify
- Ties ambience, foley, music changes, and short dialogue to specific moments
- Diagnoses failed generations and changes one variable at a time
What you get
A ready-to-paste prompt, the settings to check on your Seedance surface, and the first fix to try if it fails.
Best for
Creators and marketers making AI product ads, short animations, and social clips with Seedance.
SKILL.md preview
---
name: seedance-video-prompts
description: Turn a rough idea into a structured, ready-to-paste prompt for ByteDance's Seedance 2.0 video model - timecoded multi-shot breakdown, one camera move per shot, repeated continuity descriptors, lighting, reference-file roles, and audio cues (ambience, foley, music, short dialogue) tied to exact moments - plus a fix list when a generation comes back wrong. Use when the user says "write a Seedance prompt", "turn this idea into a Seedance video", "15-second product ad in Seedance", "my Seedance shots don't match", "add camera moves and sound to this prompt", "multi-shot prompt for Dreamina", or "why does my Seedance video look off".
---
… (sign up to view the full skill)More creator skills
View all Creator skills →Explainer B-Roll Prompt Pack for AI Video
Turns a VO transcript into a style-locked AI video prompt pack: coverage map, P1-P3 clips timed to the VO, exact data routed to code.
Seedance Scene Prompts: Action, Dialogue and Blockouts
Seedance prompts for scenes with people: weighty action, living performances, timed dialogue, multi-clip continuity, and blockout v2v.
Faceless 3D Animated Shorts Pipeline
Concept, timed narration, style bible, and per-scene image, motion, VO, and SFX prompts for faceless 3D shorts that stay consistent.
Timestamped Voice-Over Image Pack Generator
Turn a timed voice-over transcript into style-locked images, one per beat, generated and named in order, with a hold-time manifest.