Show HN: Run Gemini & ChatGPT UI with Python (github.com)

A new tool called Hermex has been introduced, enabling users to automate interactions with the ChatGPT and Gemini interfaces using Python, without the need for API keys or associated costs. Hermex essentially mimics human behavior by controlling a real Chrome browser, allowing users to perform tasks such as OCR, file uploads, and quick queries directly through the platform's web UI. This feature is particularly significant for developers and researchers looking for cost-effective alternatives to official APIs, especially for simple automation tasks that do not require extensive operational capabilities. Hermex employs undetected-chromedriver to avoid bot detection and maintains a persistent browser profile for seamless user experiences across sessions. The tool supports various formats including images and PDFs, and facilitates request handling with straightforward Python commands. While primarily designed for hobbyist and research use, its reliance on browser interactions may make it less robust compared to dedicated APIs, and users are cautioned about potential limitations or changes from the platforms' UIs. This development underscores the growing interest in flexible, accessible automation tools within the AI/ML community, democratizing access to powerful AI functionalities.