nano-banana-2

Generate and edit images with Google Gemini 3.1 Flash via inference.sh CLI.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/j-ching/claw-skills --skill nano-banana-2-j-ching
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: nano-banana-2
Source: https://github.com/j-ching/claw-skills/tree/main/nano-banana-2
Command: npx skills add https://github.com/j-ching/claw-skills --skill nano-banana-2-j-ching

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires infsh.

What problem does it solve?

This skill removes the complexity of interacting with advanced image generation models by providing a streamlined CLI interface for Google Gemini 3.1 Flash, enabling rapid visual content creation.

Core Features & Use Cases

  • Multi-Modal Generation: Supports text-to-image, image editing, and multi-image input processing.
  • Advanced Grounding: Integrates Google Search to ground image generation in real-time information.
  • Use Case: Quickly generate photorealistic marketing assets or visualize real-time data like weather reports by providing a prompt and enabling search grounding.

Quick Start

Use the nano-banana-2 skill to generate a photorealistic image of a futuristic city at sunset.

Frequently Asked Questions about nano-banana-2

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

FAQPage Schema
How do I generate AI images using Google Gemini 3.1 Flash from the command line?

Generate AI images using Google Gemini 3.1 Flash by executing model inference requests through the inference.sh CLI. This skill provides a streamlined CLI interface that handles text-to-image generation, multi-image input processing, and complex image editing tasks via authenticated sessions.

Can I use Google Search to ground AI image generation in real-time information?

Yes, Google Search grounds AI image generation in real-time information. This integration allows you to generate photorealistic marketing assets or visualize real-time data like weather reports by providing a text prompt and enabling search grounding.

Do I need the infsh CLI tool to perform image editing with Gemini 3.1 Flash?

Yes, you need the infsh CLI tool and a valid authenticated session to perform image editing with Gemini 3.1 Flash. The skill requires this environment setup to execute model inference requests and process multi-image inputs.

What is the best way to customize aspect ratios for AI-generated images via CLI?

Customize aspect ratios for AI-generated images by using the aspect ratio customization feature within the inference.sh CLI. This allows you to tailor the dimensions of high-fidelity images generated by the Google Gemini 3.1 Flash model to your specific visual content requirements.

Does multi-image input processing work with the Gemini 3.1 Flash model for visual tasks?

Multi-image input processing works with the Gemini 3.1 Flash model for complex visual tasks. The skill supports multi-modal generation, allowing you to input multiple images simultaneously for advanced editing and high-fidelity image creation.

Why does my Gemini 3.1 Flash image generation request fail during execution?

Gemini 3.1 Flash image generation requests fail during execution if the infsh CLI tool is missing or lacks a valid authenticated session. Ensure the CLI environment is properly configured to successfully execute model inference requests.