1. I don’t know how you’ve built it so far, but you might want to look into the backdrop-filter css property, which affects whatever happens to be underneath the image. Maybe you could darken the image below with the markings, sort of thing.
2. I guess store that opacity and use it in other places.
3. I guess? 🤷

RE “I want the paws and face to generate as the same opacity as the underbelly.”

At the start of your output items you’d add something like [o1 = "0.{0-99}".evaluateItem, ""] then wherever you write [o1] you’ll get that same opacity. And of course you can have lots of variables like this - e.g. [o1 = "0.{0-99}".evaluateItem, o2 = "0.{0-99}".evaluateItem, ""].

Also instead of "0.{0-99}".evaluateItem (which is a bit hacky-looking) you can use Math.random() which returns a decimal number between 0 and 1.

In general, doing “global coordination” between various aspects of your generator will require creating variables like this “up front” and then using them in various places.

BTW, Imgur has betrayed everyone and started deleting images, so it’s probably a good idea to use perchance.org/upload instead. I’ve moved the note about this to the top of the image layer combiner plugin page. And I’d use webp image format because they’re much smaller (especially for gradient images - like 10x smaller), so they’ll load faster and use up less upload storage space. Webp supports transparency and animations, so they basically replace all other formats - gif, png, and jpg. You can use squoosh.app or photopea.com to convert to webp if needed. Someone should probably make a perchance page that makes it easy for people to mass-convert imgur images using the super fetch plugin and upload plugin. Bascially, paste lists with imgur links in, get lists with perchance upload links out. If someone does that I’ll link it from the image layer combiner plugin page.