Can I use images as input for vision-related tasks?

Yes, you can upload local images as reference files, allowing the skill to perform vision tasks like describing images or using them as creative guides.

Which Gemini models are supported?

The skill supports several models including gemini-3-pro (default), gemini-2.5-pro, and the faster gemini-2.5-flash.

Where are the generated images stored?

Generated images are saved locally. You can specify a custom path using the --image [path] option, or it will default to generated.png.

Do I need a Google Gemini API key to use this skill?

No, this skill uses the Gemini web interface. It will prompt you to log in via a Chrome window on the first run to cache authentication cookies.

How do I maintain a conversation across multiple prompts?

Use the --sessionId flag with a unique identifier. This allows the skill to reuse Gemini's chat metadata and maintain context for multi-turn interactions.

Gemini Web Image & Text Generator

Name: Gemini Web Image & Text Generator
Author: proyecto26

byproyecto26

•

Content Management

Generates high-fidelity images, text, and videos using the Google Gemini web interface directly within your terminal.

The Gemini Web skill empowers Claude Code with the full creative suite of Google Gemini's web-based capabilities. It enables seamless text generation, high-quality image creation with automated downloading, and vision-based tasks through local image uploads. Designed for complex workflows, it supports multi-turn conversations via persistent session IDs and experimental video generation. It serves as a powerful content engine for developers needing to automate asset creation for articles, social media, or UI/UX mockups without the overhead of traditional API management.

Key Features

01Vision support via local reference image uploads for visual analysis.

0214 GitHub stars

03Experimental video generation capabilities through Gemini Web.

04High-resolution image generation with automated local saving.

05Support for multiple Gemini models including Pro and Flash variants.

06Persistent multi-turn conversations using unique session IDs.

Use Cases

01Rapidly prototyping marketing copy and visual assets for social media platforms.

02Automating the generation of blog cover images and article illustrations.

03Developing AI agents that require visual context and long-term memory.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add proyecto26/sherlock-ai-plugin genimg-gemini-web

For use in Claude.ai and ChatGPT

Download Skill