nano-banana

Generate and edit AI images with Google Gemini scripts.

10|1|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/pbc-os/smb-starter-kit --skill nano-banana-pbc-os
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nano-banana
Source: https://github.com/pbc-os/smb-starter-kit/tree/main/skills/tier-4-growth/nano-banana
Command: npx skills add https://github.com/pbc-os/smb-starter-kit --skill nano-banana-pbc-os

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, Pillow, numpy, scipy, python-dotenv, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This skill enables teams to rapidly generate and edit AI-driven images and build web-ready assets using Google Gemini, reducing design cycles.

Core Features & Use Cases

  • AI image generation from prompts for marketing visuals and product shots.
  • Image editing, background removal, and region-targeted inpainting to produce clean transparent assets for web use.
  • Web-asset rendering pipeline guidance and reproducible workflows for consistent asset quality.

Quick Start

Install dependencies, export GEMINI_API_KEY, and run the provided scripts to generate and edit images.

Frequently Asked Questions about nano-banana

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

FAQPage Schema
How do I generate AI images for marketing visuals using Gemini?

To generate AI images for marketing visuals, you run the generate_image.py script using the Google Gemini API and your text prompts. This enables rapid creation of product shots and web assets directly within your Python environment.

Do I need a Gemini API key to remove image backgrounds for web assets?

Yes, a GEMINI_API_KEY is required to remove image backgrounds for web assets. The make_transparent.py script uses Google Gemini to process images and produce clean transparent assets for web deployment.

Can I use Vertex AI for region-targeted inpainting?

Yes, you can use optional Vertex AI for region-targeted inpainting. The inpaint_image.py script edits specific image regions, allowing precise modifications to produce clean assets for web use.

What's the best way to edit existing images with Gemini in a production pipeline?

The best way to edit existing images with Gemini in a production pipeline is using the edit_image.py script. It provides reproducible workflows for image editing, ensuring consistent asset quality for marketing visuals.