Final Testing and Presentation

[SharanyaSD]

1. Screenshot Collection: Puppeteer captures the browser screen and sends it to the FastAPI backend.

2. Vision Model Prediction: The Llama Vision Model processes the screenshot and predicts: Coordinates (x, y) of the element. Action to be performed (e.g., click, scroll).

3. Action Execution: Puppeteer performs the predicted action in the browser.