Veo Use

This lets you generate and edit videos using Google's Veo 2 and Veo 3 models right from the command line. You get text-to-video, image-to-video, reference-based generation, video inpainting for removing or inserting objects, and extension to make clips longer. It's built with portable Python scripts via uv and works with either a Google API key or Vertex AI credentials. The inpainting modes are nice since you can remove watermarks or drop new objects into existing footage with a mask. Resolution goes up to 4K, aspect ratios cover 16:9 and 9:16, and durations range from 4 to 8 seconds per generation.

Install to Claude Code

npx -y skills add cnemri/google-genai-skills --skill veo-use --agent claude-code