Turns text prompts or reference images into full 8x9 animated spritesheets for Codex pets, complete with idle, running, jumping, waving, and failure states. It handles the whole pipeline: generating base art, producing nine animation rows via the imagegen skill, extracting frames, composing the atlas with transparent cells, running QA contact sheets, and packaging everything into pet.json. The workflow is surprisingly rigid about grounding every row in the generated base reference and enforcing a specific pixel-art house style with thick outlines and no floating effects. If you need a custom digital pet with proper animation rows and you're willing to burn ten image generation calls, this will save you from manually slicing frames and guessing atlas geometry.
npx skills add https://github.com/nexu-io/open-design --skill hatch-pet