Seedance Prompt

What it does

The Seedance Prompt skill helps users craft high-quality prompts for generating videos using the Seedance 2.0 platform. It leverages the official Seedance 2.0 user manual to guide prompt creation, focusing on incorporating various elements like image and video references, audio cues, and detailed action descriptions. This skill aims to overcome writer's block and boost brainstorming by providing a structured approach to creating effective prompts for AI-generated videos.

When to use it

This skill is useful in the following scenarios:

• Creating simple image-to-video animations or animations with transitions between frames.
• Generating videos that combine multiple media types (images, video clips, audio).
• Developing prompts requiring specific camera movements and styles.
• Producing product advertisements with detailed visual descriptions.
• Replicating action sequences from existing videos.

Key capabilities

• Provides a structured formula for Seedance 2.0 prompt creation: Material Role Specification + Action/Plot Description + Camera Language + Atmosphere/Sound Effect Instructions.
• Supports specifying material roles (e.g., character, background, prop) using the "@" symbol.
• Enables reference to images, videos, and audio for style, action, or sound effects.
• Facilitates detailed descriptions of actions and emotions within prompts.
• Allows for control over camera movements through text description or video references.
• Supports specifying background sounds, dialogue styles, and overall atmosphere.

Example prompts

• "Reference @Image1's man image, he is in @Image2’s hallway, completely reference all of @Video1’s camera movement effects."
• “@Image1’s woman walks to the mirror / Reference @Image1’s man’s appearance. Quickly reach out and drink a can of cola from the frame, then show a satisfied expression.”
• "0-3 seconds screen: [Scene A description] 3-6 seconds screen: [Scene B description]"

Tips & gotchas

• Material Limits: Be mindful of Seedance 2.0's input limits (≤9 images, ≤3 videos totaling ≤15 seconds, ≤3 audio files totaling ≤15 seconds, and a total of 12 files).
• Specificity is Key: Vague descriptions will lead to unpredictable results. Use detailed, visual language when describing actions and emotions.
• Material Roles: Ensure each material has a clear role specified using the "@" symbol for optimal results.