This hooks you into Google's Gemini 3 Pro Image API, the production-grade version of their image generator. It handles the full flow: API setup, basic generation, iterative editing through chat sessions, and character consistency across multiple images. The code examples cover Python, TypeScript, and a Next.js API route, plus configuration for aspect ratios and image sizes up to 4K. Honestly, the text rendering and Google Search grounding features are what make this interesting compared to typical image models. If you're building anything that needs reliable text in images or factually grounded visuals, this is worth the higher cost over the Flash version.
npx skills add https://github.com/hoodini/ai-agents-skills --skill nano-banana-pro