I also want that. You can repurpose the existing filters, basically generating the best workout given the filter constraints you have set. So if you specify between 60 and 90 minutes and a max difficulty at 3, the generator will generate the workout with the best score (so it will be in the top of the list) that is no more than 90 minutes at difficulty 3, while the unconstrained proposal might have been at 95 min at difficulty 3.5. So the proposed workout will then be a bit short in the various XXS, but you accept that because of the constraints you set.