creature

I think it is important and necessary that serif staff create a detailed manual as soon as possible on how to manually transfer all personal settings from the V1 to the V2. I also mean assets brushes and so on.
 

Here my workflow to straighten skewed rectangular objects:
create a rectangular shape showing the target shape as helper create a curve of 4 nodes matching the skewed source corners add perspective filter.  set to source mode match the nodes to curve helper layer (snapping does not work) set mode to destination  match nodes to rectangle (snapping to other nodes of perspective filter x/y positions is available) This is faster and more reliable that trying to directly match nodes to the source image, or using destination mode.
The main issue with perspective is that the nodes interact with each other. You can’t find a final position of one node, the next, as every node can interact with already positioned nodes. This makes it useless in many cases.
Second issue is missing snapping to grid / layer geometry.

Sure, here's a handful of examples. These are all almost, but not quite square. 

 
I often find myself wishing for a "mirror mode" – if I drag, say, the bottom right node down, I want the bottom left node to move up an equal distance. But really that's just a stopgap – letting the software figure it out is my ideal mode. 
Thanks for the tip about drawing a guiding rectangle, and then distorting it to match the skewed rectangle – I hadn't thought of that, and I can definitely imagine it being useful! 

The perspective tool is far to clumsy in certain use cases.
often I want to correct a skewed image with help of a known rectangular object.
all I want to do
source mode: mark the 4 corners of a known rectangular area (which is perspectively distorted) switch to destination mode Photo creates a default perspective layer based on minimum or maximum of corner positions  Photo shows the result i can adjust the edges in dest mode to taste to correct overall aspect ratio affinity shows the size (in pixel and selectable measurement units) and aspect ratio of the resulting area in info panel while I adjust  

Hi @creature,
Can you you try going into Preferences and on the Performance section there is an option for Enable Mental Compute.  If thats ticked, untick it, close Preferences and restart Affinity and then see if you can replicate the issue.

As a fellow Fuji shooter, I'd love to see this supported in Affinity Photo too. 

May Im the only one who need this in RAW editing for my fuji x-series and canon pictures, but is it on the roadmap for the near future? I can't find any thread about it :)
attached is an example of how adobe did this in lightroom