It is passive. But the problem is not the hot signal in case of many compressors. I have an active pad, so I can decrease the gain to any value, the bigger problem for compressors is that in the style I play, you mostly use the notes between frets 12-24, and especially around 24 the intermodulation occurs. You can influence it with input gain setting, but only to a limited amount. So for example you set it to an ideal level, notes around fret 12 do not distort, but at the same or even lower volume (you pick more lightly) the higher notes distort, if the compressor has non-linearity problems. If you lower the input volume more, the IMD disappears but then the noisefloor will be very high. This happens like that with OTA and some other designs like the one I linked. FET is okay, and also VCA and some opto like the Carl Martin. The Gurus I linked produces IMD however, and also the Demeter Compulator behaved strangely. The latter was a strange exception though, that was distorting on the decay.
In case there is no IMD, I have no problems setting the compressor. I just put the pad after the guitar, set the reduction to 50-20%,depending on what comes after it. But, for comparison: if the pedal is problematic because of IMD, i need to decrease even more, I have no idea how much exactly, but to 10% or maybe more (usually that extra attenuation happens on some of my pedal outputs). This final extra attenuation you cannot do on the pad or the guitar, because you have to decrease so much that you would need a multi turn trimmer to do it. With normal pot or small trimmer you are decreasing, and at one moment the IMD is still there, while in the next moment the sound is gone. Not sensitive enough. But this correlates to the specs of the ICs in the problematic pedals, their linear operation window is that small indeed as I read. That's why when being a beginner I did not understand how the pedals can still distort when I turned the pot back to almost 0. If I'd play below the 18th fret or so most of the time, I would have never noticed this problem.
However the challenges to set compressor comes from the operation of the circuit most of the time. That is an even bigger problem for me. We play at fast tempo, sometimes around 127 BPM, many 16th notes, this challenges the compressor to do Attack/Release on tempo. But yet the DBX rendition in Logic seems to keep up with the situation and is surprisingly good. Very likely because of the dynamic timings.