Hacker News new | ask | show | jobs
by razfar 825 days ago
I'm curious about the computer vision aspect of this tool. Specifically, how was the model which draws bounding boxes around interactable elements trained? Definitely a step beyond existing browser automation software!
1 comments

It's surprisingly dumber than you think!

I'm always fascinated by how far you can get with heuristics in certain situations. Check out the code here -- https://github.com/Skyvern-AI/skyvern/blob/d0935755963b017ed...