All in all, it seems like some good ideas which suffered in implementation. It's not as much a failure as moo3, but still lacking enough that I'll probably avoid buying it until it drops to the ~$20 range; if then.
No moe work, but I added a little bit of code for my interpreter framework. You can now add an attribute to any type/enumeration and the one-off initialization call will hunt them down and add them as interpretable types. Should save lots of time tracking down bugs which were just me being stupid and forgetting to add a type to the interpreter.
I wanted to like it, but I couldn't play it. Maybe I'm just slow.