Note that, filling in the blanks in the write‐up by analogy with the rules of Compulsion in the original Nemoroth, the zugzwang can in fact be lifted by a ‘saving move’ such as capturing the knight; as such 5. Nb5 would be a blunder: after 5. … c6, 6. c4 is forced (White in any case had only up to 2 free moves), allowing 6. … c4xb5 and white is an N down, which might well be a crippling disadvantage; white can force capture of the Nh6 before it goes on to wreak its own compulsion havoc, but only in exchange for its queen, and of the unmoved knights the black one would seem to have the advantage in terms of enemy pieces' positions (Na6 introduces no new compulsions, is safe from attack, and with the queen gone it poses a credible threat in the white camp, especially if preceded by Qa5 to prevent the white K chasing the N away once it reaches c2).

Also, applying Nemoroth rules, moving into a square under a knight's control is prohibited (4. c3 f6 is exactly forced; the double‐push is illegal), even when coming from another knight‐threatened square (all squares are equally far from the knight and there seems to be no other criterion to favour any above others).

I haven't tried out other openings though as yet; it may well be that there's some other forced win for white. And indeed all this really ought to have been specified in the rules, as with the interactions with check/‐mate.

(I was reminded of this sketch by this discussion; check‐plus‐Nemoroth‐compulsion and check‐plus‐king‐capture present very similar issues in their interactions.)