Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Fix: silent polling to prevent message spam during generation
What This Skill Does
Generates AI videos from text prompts or reference images using ByteDance's Seedance 2.0 model via the EvoLink API. Supports durations from 4 to 12 seconds, resolutions up to 1080p, and auto-generates synchronized audio including voice, sound effects, and background music. Realistic human faces are restricted by the model.
Auto-generated audio synced to video content removes the need for a separate audio production step that most text-to-video tools skip entirely.
When to use it
- Generating a short product demo clip from a text description
- Animating a static reference image into a 6-second scene
- Creating vertical 9:16 video content for social media
- Producing a narrated clip with auto-generated background music
- Quickly prototyping a video concept before full production
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Generate a 6-second 720p video of a sunset over mountains with background music
- 1Check that EVOLINK_API_KEY is set
- 2Confirm prompt, duration (6s), resolution (720p), and audio enabled with the user
- 3Run seedance-gen.sh with the confirmed parameters and poll for completion
- 4Suppress intermediate polling output and wait for the final result
- 5Return the video URL to the user with a note that it expires in 24 hours
A URL to a 6-second 720p video with auto-generated background music, ready to download or share
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.