Stack Exchange does not divide up by topic, it divides up by _readership_.
This is something that people stumble over, but if you look at all of the site descriptions in the drop-down list of "Stack Exchange communities" at the top of this very page you'll see that all of the communities are described in terms of _the people_ that read and write the questions and answers, not in terms of the subject matters.

It takes a while to spot this, especially if one has a background in the likes of CompuServe, Fidonet, Usenet, I-Link, and so forth, which did divide up by topic, and whose FAQ documents did as well.
But it is how things are structured by the company that runs the sites, and is the methodology that one can see all the way back to 2009 (where [the question was "Which community do you consider yourself a part of?"](https://stackoverflow.blog/2009/07/26/why-cant-you-have-just-one-site/)) if one goes and looks.

This WWW site is "for users" of Unices and Linux-based operating systems.
Stack Overflow is "for programmers".

So the correct approach is to consider the readership.

If an answer is aimed at a _user_ of the operating system, then it is appropriate.
It seems to be surely the case that one-liners in many of the interpreted languages that one can commonly (and even as standard in some cases) find on Unices and Linux-based operating systems, are usable by users.
Conversely, a question or an answer that is _for programmers_ is better on Stack Overflow.

I for one would object to any notion that everything outwith `grep`, `sed`, and `awk` was outwith the scope of using these operating systems.
It is both narrowminded and blinkered, and definitely not in accord with _my_ experience as a user.
Indeed, a lot of the history of Unices has been the expansion to include _all sorts_ of useful tools that users can employ, from `perl` through `rs` and `jq` to Miller.
This is, moreover, a process that is ongoing today.

So your question is ill-founded to an extent.
It is not the case that there is a set of on-topic and off-topic programming languages.
It is, rather, whether a question and an answer is appropriate _for a readership of users_.

There is an observable widespread tendency to give an `awk` answer to almost everything, but that should not be inferred as a rule to be followed, and if there's (say) a Python answer that involves _less programming_ then surely that is quite on point as an answer _for a readership of users_.

I went against the grain, applying _other tools_ that people have written over the years to directly perform the job at hand which _do not_ involve _entering a program_ for `awk` or a shell to run, with answers like https://unix.stackexchange.com/a/574309/5132 and https://unix.stackexchange.com/a/578242/5132 .
Others have done similar.
https://unix.stackexchange.com/a/584274/5132 and https://unix.stackexchange.com/a/569600/5132 are (for examples) answers that show alternative tools to answers employing shell script and (yet again) `awk` programs, namely John A. Kunze's `jot` and `rs` (reshape), which have been around since 4.2BSD for goodness' sake!

Similarly, it's not about which section number a library function has assigned to it.
It's about whether the Q&A is for programmers or for users.
Knowing that there's wacky semantics to the (Linux) `settimeofday()` system call that make the system time jump around at bootstrap is something that concerns users, especially when it affects things like filesystem checks (c.f. https://unix.stackexchange.com/a/600490/5132).
But whether one can pass a `time_t` to `settimeofday()` is a question for programmers (c.f. https://stackoverflow.com/q/3374659/340790).
It's the same system call, but two different readerships.