The Gemini AI skill bridges the gap between Claude and Google's Gemini ecosystem by using browser automation to perform tasks on the Gemini web interface. It allows users to trigger text generation and image creation commands seamlessly, utilizing either a fresh browser instance or an existing Chrome session via Chrome DevTools Protocol (CDP). By reusing existing browser profiles, it maintains authentication states and bypasses the need for manual logins, making it a powerful utility for developers who need to cross-reference AI models or generate visual assets during their coding workflow.
主な機能
01Automated image creation with customizable prompts and output paths
02Seamless text generation and Q&A using the Gemini web interface
03CDP integration to reuse existing Chrome instances and login sessions
04Support for both headless and visible browser modes for debugging
053 GitHub stars
06Automatic dependency and virtual environment management via run.py
ユースケース
01Automating research tasks that require Gemini's real-time web access
02Generating UI assets or image placeholders during frontend development
03Cross-verifying complex logic or code snippets against Gemini's LLM