skills.rest
takimoto-sketch
takimoto-sketchCOMMUNITY
Explore all skills by this author →
GitHub

gemini-image

Generate images from text prompts and edit user-supplied images with JSON metadata output.

Updated Aug 23, 2026
Search Tags:#gemini#image-generation#prompting#image-editing#nano-banana#asset-creation
One-click install
npx skills add https://github.com/takimoto-sketch/medica-agent --skill gemini-image-takimoto-sketch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: gemini-image
Source: https://github.com/takimoto-sketch/medica-agent/tree/main/.claude/skills/gemini-image
Command: npx skills add https://github.com/takimoto-sketch/medica-agent --skill gemini-image-takimoto-sketch

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Converts natural-language image requests into finished visual assets and automates common image editing tasks so users do not need to manually operate design tools or perform repetitive retouching.

Core Features & Use Cases

  • Text-to-image generation: Create new images from descriptive prompts with selectable Nano Banana Gemini models for different speed and fidelity tradeoffs.
  • Image editing and retouching: Modify user-supplied images for background replacement, style transfer, element addition or removal, and format conversion.
  • Flexible output and workflows: Choose aspect ratios, save outputs to reports/generated-images/, and receive JSON metadata for integration into reporting or downstream automation.
  • Use Case: Quickly produce social media thumbnails, presentation headers, or iterate logo concepts by requesting variations and edits directly from the agent.

Quick Start

Ask the agent to generate a watercolor illustration of Mount Fuji at 1:1 aspect ratio and save the image to reports/generated-images/.

Frequently Asked Questions about gemini-image

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

FAQPage Schema
How do I generate images from text prompts using the Gemini Nano Banana model?▼

To generate images from text prompts, you provide a natural language description and select a Nano Banana Gemini model for the desired speed and fidelity. You can also specify an aspect ratio, and the system will save the output image file to a designated directory.

Can I edit existing images and remove backgrounds without manual retouching?▼

Yes, you can edit existing images by supplying the file and natural language edit instructions. This automates background replacement, style transfer, and element removal without manually operating design tools.

Do I need a configured GEMINI_API_KEY to use this image generation and editing tool?▼

Yes, a configured GEMINI_API_KEY is required to authenticate and run the Nano Banana models. Without this key, the tool cannot process text-to-image generation or edit user-supplied image files.

What's the best way to produce marketing assets and batch variations for presentations?▼

The best way to produce marketing assets is to request batch variations and edits directly from the agent using descriptive prompts. It saves outputs as image files with JSON metadata for easy integration into reporting or downstream automation.

How does the output format work for integrating generated images into downstream automation?▼

The tool saves finished visual assets to a specific folder and outputs JSON metadata alongside the image files. This structured JSON metadata allows for seamless integration into reporting pipelines or downstream automation workflows.

Related Skills

Explore More Design & Creative Skills→
dair-aidair-ai
600

Image Generator

Generate and edit images using Google's Gemini Nano Banana Pro model.

Official
Advanced
jame581jame581
538

assets-pipeline

Automate asset import configuration and optimization for Godot projects.

Community
Intermediate
jolequesjoleques

nina-geradora-visual-nano-banana-2

Generate image prompts and PNG files for Nano Banana 2 with preflight validation.

Community
Advanced
doany-aidoany-ai
5

image-inpainting

Perform mask-driven image edits via the RunComfy CLI inpainting models.

Official
Intermediate