What models does Nanobanana support?

It supports gemini-3-pro-image-preview for high-quality production images and gemini-2.5-flash-image for faster, iterative generation.

How do I install the necessary dependencies?

You can install the required Python packages (google-genai, Pillow, and python-dotenv) using the provided requirements.txt file in the skill directory.

What aspect ratios are available?

The skill supports a wide variety of ratios including 1:1 square, 16:9 landscape, 9:16 portrait, and 21:9 ultra-wide, among others.

Do I need an API key to use this?

Yes, a valid Google Gemini API key is required and must be configured in your environment variables or a local .nanobanana.env file.

Can I edit existing images with this skill?

Yes, you can provide one or more input images along with a prompt to perform AI-driven edits, such as adding objects or changing styles.

Nanobanana Image Generator

Name: Nanobanana Image Generator
Author: Microck

byMicrock

•

Design Tools

Generates and edits high-quality images using Google Gemini API models directly within the Claude Code terminal.

About

The Nanobanana Image Generator skill empowers Claude to create or modify visual content using Google's Gemini models. By integrating directly with the Gemini API, it allows users to specify aspect ratios, resolutions up to 4K, and model preferences for tasks ranging from logo design and UI sketching to photo editing. This skill is ideal for developers and designers who want to maintain their creative flow without leaving their coding environment, providing a seamless bridge between text-based instructions and high-fidelity visual assets.

Key Features

Wide range of supported aspect ratios including Portrait, Landscape, and Ultra-wide
Support for specialized models including Gemini 3 Pro and Gemini 2.5 Flash
Advanced image generation from text prompts using Google Gemini models
81 GitHub stars
Scalable resolution options from 1K up to 4K for high-fidelity outputs
Sophisticated image editing by combining existing files with new instructions

Use Cases

Generating custom logos, social media graphics, and blog illustrations
Iteratively editing existing design assets via natural language instructions
Creating UI mockups and placeholder assets during frontend development

About

Key Features

Wide range of supported aspect ratios including Portrait, Landscape, and Ultra-wide
Support for specialized models including Gemini 3 Pro and Gemini 2.5 Flash
Advanced image generation from text prompts using Google Gemini models
81 GitHub stars
Scalable resolution options from 1K up to 4K for high-fidelity outputs
Sophisticated image editing by combining existing files with new instructions

Use Cases

Generating custom logos, social media graphics, and blog illustrations
Iteratively editing existing design assets via natural language instructions
Creating UI mockups and placeholder assets during frontend development