the-focus-ai/nano-banana-cli

Overview

This skill generates and edits images using Google Gemini image models via a CLI tool called Nano Banana. It supports text-to-image, image editing, style transfer, and multi-image composition with customizable prompts and output options. Use it to batch-process images, create marketing visuals, or iterate on design concepts quickly.

How this skill works

Provide a descriptive prompt or an edit instruction and the CLI calls Gemini image models to produce or modify an image. You can pass an input file for edits, choose a specific model, set output paths, and run batch jobs. Iteration is supported by refining prompts, editing outputs, or switching models for different styles or quality levels.

When to use it

• Create a new image from a text description (text-to-image).
• Edit or enhance an existing photo by uploading a file.
• Batch-generate slides, product photos, or social assets with consistent style.
• Produce alternate versions or variations for design exploration.
• Quickly prototype logos, icons, or hero headers with style constraints.

Best practices

• Start prompts with composition and focal subject before adding style details.
• Include lighting, camera/framing terms, and negative guidance to avoid unwanted elements.
• Use consistent style descriptors across related images for visual cohesion.
• Iterate: make one change at a time and preserve elements that already work.
• Specify aspect ratio or output size when creating non-square assets.

Example use cases

• Generate a 16:9 hero image for a blog using clear composition and color palette constraints.
• Edit a product photo to change the background to clean white and adjust lighting for e-commerce.
• Create a set of consistent social media images by reusing a prompt style guide.
• Make a minimalist app icon or logo draft, optimized for small sizes.
• Apply a watercolor or Studio Ghibli–inspired style transfer to an illustration.