Weren't tunnel diodes really sensitive? Low voltage, a small (low capacitance) part would be fairly vulnerable.
RF parts are the smallest single components you can find. YMMV on ESD or power sensitivity.
Back in the day, 3N138 was supplied with a wire ring shorting the leads together. Not necessarily vulnerable to power, but it didn't take much ESD to kill.
Also, arguably, power in what sense, because a gate has a quite high impedance at low frequencies, so with suitable matching, it won't take much signal power to damage. This is something of a contrived case, as it's hard to make coils with high enough Q at say 1MHz for this to be much of an issue.
Damage from mW? Seems plausible. But again, it's not a very motivating question because one can always make a smaller part and claim it's the most sensitive.
Matter of fact, I think I'll exhale some H2O molecules, claim that they are circuit elements in the EM waves over my bench, and push their hydrogen nuclei into another spin state with a single VLW photon...
Tim