But it can be equally malicious to just drop generic potential shortcomings of scientific studies, without any attempt to figure out whether they apply to the specific study under discussion.
I don't agree with this comment at all. Skepticism isn't denying the claim or advocating an alternative one, it is promoting thoroughness and responsibility. The uninspected claim is quite a bit more dangerous.
This particular flavor of skepticism does not promote thoroughness and responsibility, in fact it promotes laziness: "Why read (or even skim) through the article when I can just state these five ways in which the article might be bad and therefore its argument void?"
If you think broad skepticism or criticism of a scientific study based on class characteristics is admissible (some statistical investigations are poorly conducted, so this one might be poor too), why not trust based on class characteristics (scientific articles in Nature tend to be really good, so this one might be good too)?
You're right, I don't think either of these options is better than the other.