generate-video

To generate video from text prompts or images using Veo 3.1, you provide a prompt or starting image, and the Skill automates the generation via Gemini Veo APIs or Replicate Veo backends. It handles prompt rewriting and runs generation as a background task.

Yes, you can use a reference image to maintain subject consistency in text-to-video generation. The Skill supports image-to-video workflows and style application, allowing you to use reference images and last frames for subject-consistent animation.

Veo 3.1 video generation supports adjustable aspect ratios, resolutions, durations, and optional audio generation. You can validate these output parameters before running the generation task to ensure your finished video meets specific formatting requirements.

The best way to automate style-driven storytelling is using Veo 3.1 text-to-video workflows. This Skill removes manual frame-by-frame overhead by applying consistent styling through reference images, enabling rapid creative exploration for promotional clips and storyboards.

Yes, you need the Google GenAI dependency to run Veo 3.1 video generation as a background task. The Skill coordinates with Gemini Veo APIs or Replicate Veo backends through this dependency to perform parameter validation and automated prompt rewriting.