This wraps Google's Gemini 3.1 Flash Image Preview API for generating and editing images straight from the command line. You get explicit control over resolution (512px through 4K) and aspect ratios, plus the ability to feed in up to 14 reference images when you need character consistency or multi-source composition. The skill maps natural language like "quick draft" or "hero image" to actual parameters, which saves you from memorizing the API. It's faster than waiting on Midjourney's queue and more scriptable than clicking through web UIs. If you're building workflows that need repeatable image generation with version control and CI integration, this is the straightforward Python wrapper you'd write yourself if you had an afternoon.
npx skills add https://github.com/intellectronica/agent-skills --skill nano-banana-2