@Drand
Speaking for myself, I appreciate the clarification and short explanation, as otherwise this can be perceived quite simply as an attack with no true proper reason.

One question, though. Yes, I acknowledge that information on my instance is public and I do not intend to intensely restrict access to it (as there is quite simply no good reason or way to do so without extensive compromises), but this does not explain why the bot needs access to either the public timelines or specific user accounts if it is just searching the vaguely-identified "content moderation policies." For convenience, I'm not going to question the relevance or possibility of moderation in a federated network, as it's quite simply irrelevant right now.
There is a lot of ambiguity in that regard so rather than making assumptions, I'd rather ask: what's up with that? How is the information there relevant? (and also, why are single-user instances, like mine, that only have very few users, relevant to this as well, and is there a way to opt out?)

edit note (in response to the parent post's recent edit):
I'm more or less sure if this answers some concerns both expressed on my end and on the O.P.'s end (natalie), perhaps I am missing an implication or misinterpreting certain things.
Anywho, this is not my debate to start or partake in, so I'll digress. Blocking the IP remains a "good enough" opt-out mechanism, if a bit archaic, for my needs.