Since this uses SSH, how do you deal with the fact that any user could login to the remote server? Is there a custom daemon running that implements the SSH protocol, or does the client use the real SSH daemon? If it's the latter, someone could easily get a command shell. In that case, there would have to be some sort of sandbox to make sure that user can't do anything dangerous. I'd love to hear how the creators deal with this - great product btw!