nano-banana-pro-3

Generate or edit images with Gemini 3 Pro via OpenRouter API.

Updated Aug 27, 2026

Search Tags:#image #composition #generate #OpenRouter #prompt-engineering #image-editing #Gemini-3-Pro

Ready to run

npx skills add https://github.com/pascalandy/dotfiles --skill nano-banana-pro-3

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nano-banana-pro-3
Source: https://github.com/pascalandy/dotfiles/tree/main/dot_config/opencode/skill/util-nano-banana-pro-3
Command: npx skills add https://github.com/pascalandy/dotfiles --skill nano-banana-pro-3

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires typer, python-dotenv, openai, Pillow, rich, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill enables users to quickly generate and edit images by interfacing with the Gemini 3 Pro Image model via the OpenRouter API, eliminating tedious manual artwork or editing steps.

Core Features & Use Cases

Text-to-image generation using Gemini 3 Pro.
Image editing and composition by supplying input images and a descriptive prompt.
Use Case: Create multiple visual concepts for a product launch by providing prompts and optional input images; iteratively refine results with minimal manual work.

Quick Start

Use the nano-banana-pro-3 skill to generate an image with a prompt, e.g.: uv run .opencode/skill/nano-banana-pro-3/scripts/gen_image.py --prompt "A futuristic city skyline at sunset"

Frequently Asked Questions about nano-banana-pro-3

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

FAQPage Schema

How do I generate images with Gemini 3 Pro via OpenRouter?▼

To generate images with Gemini 3 Pro via OpenRouter, run the provided Python script with a text prompt. The skill handles text-to-image generation and saves the output files directly to your local directory.

Can I edit and compose multiple images using Gemini 3 Pro?▼

Yes, you can edit and compose multiple images using Gemini 3 Pro by supplying up to four input images alongside a descriptive prompt. The skill processes the inputs to transform or combine them as directed.

What image resolutions and aspect ratios does Gemini 3 Pro generation support?▼

Gemini 3 Pro image generation supports configurable image sizes of 1K, 2K, and 4K. You can also specify custom aspect ratios and use a seed for reproducible image outputs.

Do I need an OpenRouter API key to run the image generation script?▼

Yes, you need an OpenRouter API key configured in your environment to authenticate requests. The script uses python-dotenv to load credentials and the OpenAI client library to interface with the API.

How can I test image generation prompts without using API credits?▼

You can test image generation prompts without using API credits by enabling the dry-run mode. This validates your prompt configurations and setup before making actual requests to the OpenRouter API.