kingbootoshi

kingbootoshi / nano-banana-2-skill

Public

AI image generation CLI powered by Gemini 3 Pro. Green screen transparency, reference images, style transfer. Also a Claude Code plugin.

101
4
100% credibility
Found Feb 27, 2026 at 59 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

A user-friendly command-line tool for generating AI images powered by Google Gemini models, featuring custom resolutions, aspect ratios, reference image editing, transparent backgrounds, and usage cost tracking.

How It Works

1
🔍 Discover Nano Banana

You find this fun tool that turns your words into amazing pictures using smart AI.

2
📦 Get it ready

You download and set it up on your computer with a few simple steps so it's easy to use anytime.

3
🔗 Connect the AI magic

You link it to Google's free image-creating service so it can bring your ideas to life.

4
🎨 Describe your picture

You type a simple description like 'a sunny beach at sunset' and it generates a beautiful image right away.

5
✨ Make it perfect

You pick sizes from quick sketches to giant posters, shapes like square or movie-wide, blend in your own photos, or remove backgrounds for clean cutouts.

6
📊 Check your progress

You see a summary of how many pictures you've made and what it cost, keeping everything organized.

🎉 Your images are ready!

You now have custom artwork for websites, designs, games, or fun, all created effortlessly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 59 to 101 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is nano-banana-2-skill?

Nano-banana-2-skill is a TypeScript CLI for image generation AI, powered by Gemini models like 3.1 Flash and 3 Pro, letting you create images from prompts at resolutions from 512px to 4K with custom aspect ratios. It handles reference images for editing or style transfer, generates broadcast-grade transparent assets via green screen keying, and tracks costs automatically. Built for Bun runtime, it requires FFmpeg and ImageMagick for transparency, and doubles as a Claude Code skill for seamless AI-assisted workflows.

Why is it gaining traction?

It stands out as a lightweight image generation tool with Gemini's top leaderboard performance at low cost—think $0.05-0.30 per image—plus unique green screen transparency that delivers clean PNGs without manual editing. Devs love the dead-simple CLI for quick prompts, multi-ref support, and Claude integration that auto-builds commands like "generate a robot mascot with transparency." Cost logging and free API keys from Google AI Studio make it a no-brainer for frequent use over heavier image generation models.

Who should use this?

Frontend devs prototyping UI mockups or dashboard screenshots, indie game developers churning out transparent sprites and assets, and marketers needing fast product visuals or hero images. It's ideal for anyone embedding image generation prompts into GitHub workflows, ReadMe PNGs, or profile banners without spinning up complex setups.

Verdict

Grab it if you need quick Gemini-powered image generation free of fluff—40 stars and 1.0% credibility score signal early-stage maturity with thin tests, but polished docs and CLI make it production-ready for personal projects today. Scale up once it hits broader adoption.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.