summary() does not count missing characters

as.character(NA) |> rep(times = 10) |> summary()
vs.
as.numeric(NA) |> rep(times = 10) |> summary()
This feels inconsistent, doesn't it?

I constantly need to count missing characters in some data.frame columns.
It would be helpful if summary() can do that. Thanks.

Sincerely,

Yifan Liu