Is the statement "Claims demand evidence" always true?

[Mudhammam]

Metaphysical questions are always empty in that sense: if the question is well defined, you probably won't be able to find the answer; and if you leave the question vaguely defined, your answers are more likely to be determined by your chosen semantics than by what is real or true.  Is there a "Sky Daddy"?  No way, that's an incoherent idea.  Is there "some kind of entity or philosophical quantity which is responsible for the existence of the universe?"  Almost for sure yes, but it's up to you whether you'd call it God.

What if I define "mind" as the "ability to process information from one's environment?"  Is there evidence for mind?  Absolutely, but it's arbitrary whether one is willing to call that mind.  So if I ask you, "Show me evidence that Susan has a mind," and you point to her brain activity, what does this mean?  Is it evidence or isn't it?  It is evidence only if we've already accepted it as fact that certain brain activity is evidence of mind-- i.e. it begs the question.

Not sure if I answered you or not.  The short version is that reality itself may be dependent on our stance on it, much as the nature of a photon is dependent on how we interact with it-- it is in fact both / neither / one of a particle and a wave, and the truth "resolves" in response to our semantics and our chosen method of interaction.

"Evidence," then, is a very slippery thing, and whether X is evidence for Y may depend more on how you choose to define terms than on some objective reality underlying them.

I think I agree, though perhaps I would add: there are better or worse descriptions, and some questions might in principle have no "answer," nor demand one.

This possibly, as you said, begs a further question, which is whether or not any description can be objectively better or worse, or if that is a fundamentally subjective determination. But given that I think such a route probably tends towards self-defeating conclusions, or at the very least, extremely awkward ones, I would be inclined to defend the objectivity of rational value judgments.