Hi Alan,
We do V4 Miniscope imaging through a cranial window all the time. In mice you can leave Dura intact but in larger animals you usually have to remove it. The key here is you need to be able to get the bottom of the Miniscope within its working distance to the cells you want to image (also remember that you generally are limited to imaging through ~300um of tissue before scattering of fluorescence degrades your spatial resolution too much to resolve cell bodies). With the standard V4 Miniscope lenses (
https://github.com/Aharoni-Lab/Miniscope-v4/wiki/Lens-Configurations) this means the bottom of the Miniscope will need to be ~700um from the cells you want to image. In this space you need to fix a cranial window and the cement for the Miniscope baseplate. It can sometimes be challenging to fit this all within ~700um so you might want to swap out one or both of the Miniscope objective lenses to give you a larger working distance.
When baseplating, you will want to first attach the baseplate to the bottom of your Miniscope and watch the live view of the Miniscope while baseplating. You will want to adjust the x,y,z position of the Miniscope + baseplate until you see cells (or other landmarks that denote the layer you want to image) and then cement the baseplate in this exact spot. Generally this will still leave a small gap between the bottom of the Miniscope and the cranial window but this will depend on the lenses used, the thickness of your window, and the depth you are trying to image.