I had a look at that site.
First thoughts -
The capacitor might provide some protection but it is an electrolytic with a low 16 volt rating.
It might not work if you get on the wrong side of B+ in a tube circuit.
I think I would use a .1 mfd. 600V mylar type cap of some sort.
Large enough to not roll off low frequencies.
Also, instead of a potentiometer you might consider various precision decade resistors like 1% at 100 ohms 1000 ohms 10K, 100K and 1 meg. (one at a time) in series with the cap on the positive side of the probe. If you build a network with a terminating resistor to ground, then you will know more precisely what you are measuring, like 10V in = 1V measured on the scope etc.
I know that decade (X10, X100) attenuation of signals and a switchable HV DC decoupling cap were features of "real" scopes and their probes.
I will have to see if I can find a schematic of a probe.