So I haven't honestly botted in a very long time on account of all the sniffing programs, but it just occurred with cheap hardware like the RaspberryPi wouldn't the best solution be to make it fully isolated in an entirely other device and only communicate through mouse/keyboard outputs the work exactly the same as a human, and use a visual recognition program?
This would make it entirely untraceable because you would only be feeding video output into a device, and feeding mouse/keyboard commands via usb?
Honestly it would be pretty much exactly how Yeti Wild Catch worked from my understand. It would be pretty basic although I do remember a lot of really older bots used only visual cues to run anyways.
Mind you I don't know if it would be good for complicated tasks, but you could at least use it for fishing, and basic mob grinding.
Outside of the complications of writing the programming could someone tell me why this hasn't been done yet? Like this would also be entirely immune to any changes in the games actually software because it would use visual translation.