yesterday someone thought it's a good idea to create an account in my project, let an agent fuck around (it was probing for php2shell), then send me an email with a list of "improvements". Just to understand how valuable these points were one problem it flagged was that you need an account to log in...
I thought about sending the guy an email that his agent is trying to hack my stuff, or when he emailed I was thinking of replying how this was a waste of my time (his robot buddy didn't even activate the account so all he saw was a "pay" button). In the end I just blocked the sender but hoo boy, some people really have some nerve, and lack brain cells too. What did he think will come out of this? Couldn't I have done the same (but better) if I were using LLMs?
The only good thing it resulted in is I now ban agentmail.to e-mail addresses