baoyu-danger-gemini-web

Generate text and images via the reverse-engineered Gemini Web API.

2|Updated Apr 6, 2026

Search Tags:#cli #gemini #image-generation #text-generation #vision-input #reverse-engineered #oauth-cookies

Ready to run

npx skills add https://github.com/augustscl/awesome-xiawang-skills --skill baoyu-danger-gemini-web-augustscl

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: baoyu-danger-gemini-web
Source: https://github.com/augustscl/awesome-xiawang-skills/tree/main/baoyu-danger-gemini-web
Command: npx skills add https://github.com/augustscl/awesome-xiawang-skills --skill baoyu-danger-gemini-web-augustscl

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Provides a local CLI client that uses a reverse-engineered Gemini Web API to generate text and images when an official API or integrated vision model is not available, enabling automation and integration into developer workflows.

Core Features & Use Cases

Text generation and multi-turn conversations with session persistence for stateful dialogs.
Image generation from prompts, saving generated images, and using reference images for vision-enabled prompts.
Chrome-based authentication and cookie management (auto-login, cookie refresh, and CDP-based cookie loading) for developers needing programmatic access.
Use Case: Produce prompt-driven concept images and associated textual descriptions for content pipelines, then save images and JSON outputs for downstream tooling.

Quick Start

Run the CLI with a prompt to produce text or an image, for example: npx -y bun skills/baoyu-danger-gemini-web/scripts/main.ts --prompt "A cute cat" --image out.png

Frequently Asked Questions about baoyu-danger-gemini-web

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

FAQPage Schema

How do I generate images and text from a local CLI using the Gemini Web API?▼

You can generate images and text via a local CLI by running a reverse-engineered Gemini Web API script, passing prompts for text or output paths for image generation. This enables automated conversational workflows and prompt-driven concept image creation directly from your terminal.

Do I need Chrome CDP cookies to authenticate the Gemini Web reverse-engineered client?▼

Yes, Chrome CDP cookie loading is required for browser-based authentication. The client handles Chrome-based cookie management, auto-login, and cookie refresh to establish programmatic access to the Gemini Web API for local automation workflows.

Can I maintain multi-turn conversations and session persistence with the Gemini Web CLI?▼

Yes, the Gemini Web CLI supports multi-turn conversations with session persistence. It manages session files to maintain stateful dialogs, allowing you to continue conversational text generation across multiple interactions within your local developer workflows.

Does the Gemini Web CLI support vision input and reference images for prompts?▼

Yes, the CLI supports vision input by uploading reference images for vision-enabled prompts. You can generate images from text prompts, save generated output images, and use existing reference images to guide the Gemini Web text and image generation process.

What is the best way to integrate Gemini Web image generation outputs into downstream tooling?▼

The best way to integrate outputs is by utilizing the CLI's JSON output format alongside saved image files. Running the script with specific output paths captures both the generated concept images and textual descriptions, providing structured JSON data for downstream content pipelines.

Are there limitations to using a reverse-engineered Gemini Web API for local text generation?▼

Using a reverse-engineered Gemini Web API depends on browser-based cookie authentication and is subject to web endpoint stability. It is intended for local CLI workflows where an official API or integrated vision model is not available, meaning it may break if the web interface changes.