Can I generate images with the Gemini Web skill?

Yes, the skill supports full image generation from text prompts and can also create variations based on reference images provided by the user.

How are conversation sessions stored?

Sessions are stored locally as JSON files in a dedicated data directory, containing chat metadata, message history, and timestamps for multi-turn continuity.

Is user consent required for using this skill?

Yes, because this skill uses a reverse-engineered API, it includes a mandatory consent flow and disclaimer that must be accepted upon first run.

Which browsers are supported for the login process?

The skill automatically detects and supports Chrome, Chrome Canary, Chromium, and Microsoft Edge for session and cookie management.

Does this skill require a Google Gemini API key?

No, this skill uses a reverse-engineered web client that authenticates through your local browser's Google account session rather than a standard API key.

Gemini Web Content Engine

Name: Gemini Web Content Engine
Author: Jst-Well-Dan

byJst-Well-Dan

•

Content Management

Integrates Gemini Web API capabilities for advanced text and image generation directly into the Claude Code environment.

This skill provides a robust bridge to Google's Gemini Web interface, allowing Claude to perform multimodal tasks such as high-quality image generation, vision-based analysis, and complex text completion. By utilizing a reverse-engineered API client, it enables developers to leverage Gemini's Pro and Flash models for multi-turn conversations and prompt-based asset creation. It is particularly useful for workflows requiring visual AI capabilities or for bypassing traditional API constraints during the prototyping phase of development.

Key Features

01High-fidelity image generation from text prompts

02Automatic browser-based authentication for Google account sessions

036 GitHub stars

04Persistent session management for multi-turn AI conversations

05Support for multiple model tiers including Gemini 3 Pro and 2.5 Flash

06Vision input support for analyzing and describing reference images

Use Cases

01Generating UI assets and placeholder graphics directly within the development workflow

02Implementing complex multi-step content pipelines that require multimodal AI processing

03Using vision-capable AI to describe or debug frontend components from screenshots

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add jst-well-dan/skill-box baoyu-danger-gemini-web

For use in Claude.ai and ChatGPT

Download Skill