If you're working with Google's Veo 3.2 video generator and need help turning messy creative input into proper prompts, this does that translation work. It takes your images, videos, audio clips, and rough ideas, then outputs a structured JSON with an optimized prompt using Veo's 5-part framework (shot, subject, environment, camera, style) plus the right reference image attachments and parameters. Built specifically for the Artemis engine's capabilities like native audio generation, multi-reference images, and physics-based rendering. Honestly most useful when you're chaining this into a larger workflow where the next step actually calls the Veo API, since it just prepares the payload rather than generating video itself.
npx skills add https://github.com/pexoai/pexo-skills --skill veo-3.2-prompter