Hi John, thank you very much for your kind reply and i'll check the example you suggested !
I can confirm that, for now, the batch rendering with separate AOVs is good anyway for compositing in Fusion: metadata for cryptomatte, position pass and other render elements are saved and useful.
I'll need more info also regarding
Context,
Scopes and
Sets. If there are some video tutorials about this topics, are much appreciated. In the meanwhile i'll play with embedded examples in Gaffer.
Thank you again
Cheers