Browser Automation Agent for Azure VMs
The Hyda Droplet Agent is an intelligent browser automation system that uses AI to perform complex web tasks. It combines OpenAI for task planning and Hyda Vision API for UI element detection to execute browser actions autonomously.
Uses OpenAI GPT-4o to break down high-level commands into actionable tasks
Leverages Hyda vision model to identify and interact with UI elements
Runs on Mac, Linux, and Windows. Test locally, deploy to Azure
Full REST API for task management, monitoring, and control