Thats great of you sharing those 🙂 Actually the 1st and 2nd do exist but not as a single node doing exactly that.
The Inverter can be achieved with the Remapper Decorator (remaping Success to Failure and Failure to Success).
The Timeout Interruptor, can be achieved with the Interrutor Decorator by assigning it the TimeOut Condition Task.
But I can see that having them self-contained nodes may speed up workflow 🙂
So in practise, the FirstSuccessOrFailure parallel policy will break out as soon as any child node returns anything but Running with the parallel returning that status. I can’t think of a use case out of my head right now, but it certainly seems interesting. I will add that as well.