To me, it depends on whether there actually exists only a small number of standard, widely known and used resources of the type that the questioner is asking about. This is part of why some of the piano questions seem OK despite being contrary to a literal reading of the on-topic guidelines. (The other part of the good ones is that they are very focused on a relatively focused aspect of technique).
The main problems that are brought up against resource requests are:
The utility/importance/desirability of the resources are going to change.
This is especially true for the more technology oriented Stack Exchanges, where today's standard reference becomes obsolete in a year or two.
The nature of the question promotes the construction of a list of answers, none of which is a the clear (objective) winner -- alternately you
could phrase it as any of a large number of possible answers would be equally applicable.
If a question manages to spell out all of the soft/subjective features of the OP's actual problem, you won't get a question that anybody else would find useful.
I believe that for some aspects of music pedagogy/practice there might be a relatively small number of standard etudes or other resources that these kinds of questions could work. This would mean that some resource requests would be acceptable on a case by case basis. If we went this route it might be desireable to add a meta discussion along the lines of this Physics SE one.
One side effect of this might be the appearance of a bias towards classical music: it is my impression that in that field there is a more stable set of materials that are commonly used. So when a guitarist comes here, and his/her resource request question gets shot down, they'll be able to point to all the apparently similar questions that were OK by this criterion. But, if my premises are correct, this is a function of the fact that there are not a limited set of standard, canonical, resources for guitar technique (yet).