Assessing the subjective quality of a thing is in my experience one of the worst ways to use any LLM.
anthropic frontend-design skill does a great job with it.
i'd say this is something that has gotten orders of magnitude better with recent releases than it used to be, fwiw
There's a lot of objective principles and decisions that go into subjective quality; if you don't know the field well, asking LLM for assessment is a good way to discover all that.