> but I find the generative art bits really frustrating, and typing in long prompts to be also kind of not worth the effort

I think this might just be inherent in the domain - the state space for images is so much larger than it is for text, so there's just a lot more ways to interpret a text prompt. Sane "defaults" help, but it might just be inherently true that it takes a longer prompt to get close to what you're seeing in your head.