I think the rule of thumb would be something like ...

# stdlib

[orangy]

I think the rule of thumb would be something like this. Consider a 2x2 matrix, on one axis “all devs” and “some devs”, on another “use often” and “use rarely”. Basically, if the feature falls into “all often” it should be in stdlib. If it’s “all rarely” or “some often” then we need additional justification, like it saves a lot of code, make code better to read, etc. “some rarely” is almost always no go.