nano-banana-ui

Generate Nano Banana images and assemble them into frontend interfaces.

Updated Feb 24, 2026
One-click install
npx skills add https://github.com/ms-moar/ccc-media --skill nano-banana-ui
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nano-banana-ui
Source: https://github.com/ms-moar/ccc-media/tree/main/ailabspro/files/extracted/v42-nano-banana-ui-skill/nano-banana-ui
Command: npx skills add https://github.com/ms-moar/ccc-media --skill nano-banana-ui

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

Create cohesive, production-ready frontends that use AI-generated Nano Banana images as first-class design assets, removing the guesswork of prompts, validation, cropping, and integration so designs don't look like pasted-on demos.

Core Features & Use Cases

  • Two-generation modes: Anti-Gravity manual prompt mode for guided desktop generation and Gemini API mode for automated image production.
  • Operational workflow: Project scaffolding, asset manifest creation, prompt generation following strict prompting rules, technical validation, visual review, safe cropping, color extraction, and frontend assembly.
  • Built-in tooling: Scripts for image generation, validation, and center-cropping to convert square Nano Banana outputs into usable UI ratios.
  • Use cases: Building landing pages with hero imagery, dashboards with textured backgrounds, portfolio sites with custom illustrations, and any UI that needs cohesive, non-generic visual assets.

Quick Start

Choose Anti-Gravity or API mode, generate the required square images into public/images/, run the validation script, crop any backgrounds as needed, extract dominant colors for CSS variables, and then build the UI integrating the images.

Frequently Asked Questions about nano-banana-ui

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

FAQPage Schema
How do I use AI-generated images for frontend UI design without them looking pasted on?

Nano Banana image generation creates square AI assets that you must center-crop into usable UI ratios for frontend integration. This process removes the guesswork from design-to-code workflows by applying technical validation and color extraction for consistent CSS variables.

How do I generate Nano Banana image assets and assemble them into a production frontend?

You generate Nano Banana image assets by choosing a generation mode, creating an asset manifest, and running scripts for image generation, validation, and center-cropping. After extracting dominant colors for CSS variables, you assemble the images into your frontend UI.

Can I use the Gemini API for automated image generation in frontend design workflows?

Yes, you can use the Gemini API for automated image generation by setting the GEMINI_API_KEY environment variable. This automated mode runs alongside an Anti-Gravity manual prompt mode for guided desktop generation, both requiring Python 3.10+ and the google-genai dependency.

Do I need Python and Pillow to crop and validate square AI images for web layouts?

Yes, you need Python 3.10+ and the Pillow library to validate and center-crop square AI images into usable UI ratios. These dependencies handle the image processing required to convert raw generation outputs into safe, production-ready frontend assets.

What's the best way to build landing pages with AI hero imagery and textured dashboard backgrounds?

The best way to build landing pages with AI hero imagery is to follow an operational workflow that includes project scaffolding, strict prompt generation, technical validation, and safe cropping. This ensures cohesive visual assets for dashboards and portfolio interfaces.

What are the limitations of using AI generated images for production frontend interfaces?

Limitations of using AI generated images for production frontends include the strict requirement to crop square outputs into proper UI ratios and the need for continuous visual review. You must also manually extract dominant colors to maintain consistent design-to-code integration.