I'm going to fail in my response too, since it's not my field. :) Starting from the beginning of the search, there are some gases (like oxygen) which are prett

I'm going to fail in my response too, since it's not my field. :) Starting from the beginning of the search, there are some gases (like oxygen) which are pretty volatile and react with many things. If volatile gases and compounds are in the atmosphere, two possibilities exist: 1) It's young, and hasn't burned through all the reactions that its raw materials would present. 2) There's some process actively creating more volatile gases, which could be a sign of something akin to life, because it doesn't drop into an expected low-entropy chemical makeup. People much smarter than I in the field would be able to distinguish more "non-living" sources like geological cycles than less-expected "maybe life" activity. But basically, if there's some life-ish process on the planet, it should affect the entire atmosphere enough to be detectable. Certainly it's only 1 way that chemical activity can be detected, because if things are bound into subterranean caverns or don't involve gases in their activity cycles, it probably wouldn't be visible in its atmospheric composition. However, this sort of search doesn't necessarily target earth-like life, but any "huh, that's strange" observance of chemical makeups, which is good. So we have one additional means of defining "life", in that it goes against the chemical entropy-reducing direction of its planetary environment; there are sustained complex active processes that aren't just raw chemical or physical potential energy releasing itself. If we find something "living", only then the question of whether or not it's "intelligent" becomes meaningful, because the only metric we currently have for that is "thinks like humans" which isn't enough to work with.