With Nyquist in Audacity it’s sufficient to give te sound object as a return-value back to Audacity, see Nyquist and Audacity → Return Values

As edgar said, you don’t usually need to use (cue …), but in the code that we were discussing recently involving (at …), (cue … ) was necessary because (at …) acts on behaviours, which are different to sounds.

There is a section in the manual about sounds vs behaviours