Would it be safe to use a monitoring prism (GMP104) as a backsight from several different station setups?
I would like to put it on a building to save setting up a tripod backsight on revisits.
I would assume that the nodalness of the crosshairs would not matter as the position is arbitrary and not above a nail?
Thanks
William

You mentioned the nodal point. That's the only problem I foresee. If the prism isn't facing your instrument exactly then there is going to be some error. Since you mention monitoring that error could be significant.
If you use an appropriate prism constant then the prism will be nodal. I think this is a 25mm diameter prism so it should be -17.5. This is the same as +17.0 on a Leica.
So using this constant, your autolock and EDM should give you the same point regardless of the direction from which you view it (within reason).
I am not sure how you can check this is the nodal constant though as rotating it on the bolt is not necessarily rotating it through the nodal point. On this prism there is a difference between these points.
We've been using these prisms set for monitoring purposes as back sights like you want to and had pretty good results. They aren't moving so they work well. I usually get about 0.01-0.02 different results. I am setting the temperature and atmospheric pressure in the gun before using them. Close enough for me.
The biggest problem that I have(which you may not have) is TopSURV's way of handling prism offsets. Changing the prism constant routinely seems to confuse the heck out of TopSurv and then it doesn't apply the correct prism offset anymore. My solution is to have a file strictly for resectioning, and watch the results CLOSELY. When the program starts to incorrectly apply the wrong prism factor, time to start a new file and start fresh. When I change the offset constant from Topcon's A7 prism to -.30mm and see no change in the distance, Or backsight and get about a 0.07-0.10' difference, I know it's time...
Short answer is yes you can use it. I have done so many times. Ensure that the mark is secured on a sturdy structure. From extensive experience with these prisms the accuracy degrades the further you move away from a perpendicular location, so be aware.
Also unscrewing the prism for rotation is not a good idea. It never goes as planned.
I have found the following alternative but have yet to try it
http://www.meterriss.de/en/mini-prisma-rotatable.html
Cheers