I'm not the regexp guru, but you might try:
"/(<[>]+>)+$/"
(I don't know why, but my preview is showing a space before the closing paren; that shouldn't be there; it isn't there in my original.)
I haven't tried this myself, but I think it says to find a pattern which matches the entire string (the '' and the '$'), consisting of a repetition ("(...)+") of one or more patterns, each pattern containing a "<" followed by any sequence of one or more characters (other than >), up to a ">". It would match:
<b>
<br><br><pre></pre>
<img height="130"></pre>
as well as:
< >
It should not match:
<>
or
<b>A second paragraph</b>
If you need only alpha, numbers, space, and '/' inside the tag (as in <b>), then mod the regexp to:
"/(<[A-Ba-b0-9 \/]+>)+$/"
(There should be a space after the '9', but there should not be a space before the closing paren.)
If you need to get really precise, and ensure that the tag name starts with a char, includes quotation marks, etc., you can mod that appropriately.