It isn't certain that you'd be able to tell WHICH input that the attacker used, at least not with my scheme where you hide who's using what input. Revealing who's using what input might not be optimal if a user want to use inputs already tied to himself AND some inputs that aren't already, and doesn't want the unlinked ones to become linked to him.
Send all inputs to one key first.
I don't see how that solves anything. You just openly linked your own inputs together yourself, then.
Who said do it in the open? I was thinking of a group signature join, where the inputs aren't correlated to the outputs, but then you control all the outputs (they are paying to your anonymous public keys) even where you might not have controlled all the inputs (some of whom are paying to you). Perhaps I am misinterpreting the case you are describing?
Perhaps my prior comment was "off topic" because I was thinking in terms of a solution where we can put the group signature in the blockchain. Thus my rebuttal to you here is I guess "off topic". At least that explains what I was thinking before.