I have a suggestion, let the software build in ai text to generate the character, first use the ai to bind the bones, then use the text to make the ai generate different actions. I've used viggle ai so far, and it can already generate different actions for the character using text. They do this by building in a different action prompt, which can be given to the character and then generate different actions, which is a really good idea in my opinion. I think that's a really good way to do it, so when we have a strong prompt library, we can use it for human characters. We'll use different prompt libraries for different characters for objects and animals.

This will make character creation much more efficient!