nano-banana

Generate, edit, and compose images with the Gemini 3 Pro Image model.

12|2|Updated Dec 21, 2025

Search Tags:#image #Gemini #edit #generate #compose #aspect-ratio #Pillow

Ready to run

npx skills add https://github.com/devonjones/devon-claude-skills --skill nano-banana-devonjones

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nano-banana
Source: https://github.com/devonjones/devon-claude-skills/tree/main/plugins/nano-banana/skills/nano-banana
Command: npx skills add https://github.com/devonjones/devon-claude-skills --skill nano-banana-devonjones

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, Pillow, and includes scripts (resource) components.

What problem does it solve?

This Skill enables creators to rapidly generate, edit, and assemble images using Gemini 3 Pro Image model, reducing design time and enabling rapid visual iteration.

Core Features & Use Cases

Text-to-image generation with adjustable aspect ratios and resolutions.
Image editing and refinement via natural language prompts.
Multi-image composition (up to 14 reference images) for product shots, logos, social visuals.

Quick Start

Run: python scripts/generate_image.py "A cat wearing a wizard hat" cat.png --aspect 16:9 --size 2K

Frequently Asked Questions about nano-banana

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I generate images with Gemini using a text prompt and specify the aspect ratio?▼

To generate images with Gemini, run the local Python script by passing your text prompt, output filename, and optional flags like --aspect for aspect ratio and --size for resolution.

Can I edit and refine existing images using natural language prompts with Gemini?▼

Yes, you can edit existing images by providing natural language prompts to the Gemini 3 Pro Image model, enabling rapid visual iteration and refinement without manual editing software.

What is multi-image composition and how many reference images can I combine for product shots?▼

Multi-image composition merges visuals for product shots or logos, allowing you to combine up to 14 reference images into a single cohesive output using the Gemini model.

Do I need an API key and Python packages to use the Gemini image generation scripts?▼

Yes, you need a GEMINI_API_KEY along with the Python packages google-genai and Pillow installed to run the local scripts for image generation, editing, and composition.

What are the limitations of using the Gemini 3 Pro Image model for design workflows?▼

Limitations include dependency on a valid GEMINI_API_KEY and local Python environments with Pillow and google-genai, meaning workflows are constrained by API usage limits and script execution capabilities.

Is nano-banana the best way to automate marketing visuals and logo design with Gemini?▼

nano-banana automates marketing visuals, product design, and logos using Gemini 3 Pro Image, offering fast iteration and consistent aesthetics through local Python scripts for rapid visual generation.