What problem does it solve? Generating images with the Nano Banana 2 model requires knowing its prompting patterns, resolution tiers, aspect ratio options, and when to route to sibling models; this Skill packages that knowledge so prompts produce stronger output than naive usage. ## Core Features & Use Cases - Text-to-image generation: Calls runcomfy run google/nano-banana-2/text-to-image with a documented input schema covering prompt, num_images, seed, aspect_ratio, resolution, output_format, safety_tolerance, and enable_web_search. - Prompting guidance: Documents subject-first grammar, quoted in-image typography, seed locking for reproducibility, and anti-patterns to avoid. - Model routing advice: Explains when to use Nano Banana 2 versus Nano Banana Pro, Nano Banana Edit, GPT Image 2, Flux 2, or Seedream. - Use Case: Generate a batch of four vertical 9:16 social thumbnails at 0.5K resolution for ideation, then promote the winner to a 2K final render with a locked seed. ## Quick Start Ask the agent to generate an image with Nano Banana 2, for example: "Use nano banana 2 to generate a 16:9 product card image of a ceramic mug with the headline 'Brewed Quietly'."
