Clever. Unfortunately, protection is trivial, and making it work if there is a safeguard would require having access to the recruiter's model and a lot of prompt hacking.
How?
Though I think this sort of prompt injection is actually very difficult to parse for solely because you could also just put it into a body of text.
Your recruiter is only skimming your resume, a quick parenthetical in the middle of a paragraph won’t be noticed.
Will get a chuckle at the interview though, but now you have a foot in the door!