Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Generates new videos from text prompts, images, or reference inputs using 165+ AI models via the EachLabs Predictions API. Covers text-to-video, image-to-video, transitions, motion control, talking head, and avatar generation. For editing existing videos, a separate skill handles that.
Access to 165+ specialized models through one consistent API means you can pick the right model for the task without managing separate integrations for each provider.
When to use it
- Animating a product photo into a short social clip
- Creating a talking head video from a portrait and audio file
- Generating cinematic b-roll from a scene description
- Building smooth transitions between two images for a presentation
- Producing avatar videos with lip sync from a script
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Create a 5-second video of a golden retriever running through a meadow at sunset
- 1Check model schema with GET /v1/model?slug=pixverse-v5-6-text-to-video to confirm input parameters
- 2POST to /v1/prediction with model slug, version 0.0.1, prompt, resolution 720p, duration 5, and aspect ratio 16:9
- 3Receive prediction_id from the response
- 4Poll GET /v1/prediction/{prediction_id} until status is success
- 5Extract the output video URL from the response
A URL to the generated 5-second cinematic video clip
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.