Generates AI images from text prompts using the ListenHub CLI, supporting both Gemini Pro and Flash models with resolutions up to 4K and various aspect ratios. The interaction flow asks one question at a time (model, resolution, aspect ratio, optional reference images) and requires explicit confirmation before generating. Flash model unlocks extreme ratios like 1:8 and 8:1 for panoramic shots. Output can be inline, downloaded to a dated artifact directory, or both. The prompt handling is sensible: it passes your text directly by default and only offers to enrich very short prompts if you haven't asked for verbatim generation. Supports both local files and URLs as style references, up to five per generation.
npx skills add https://github.com/marswaveai/skills --skill image-gen