From the StackOverflow question "Check if a string is html or not" there are many examples and a few approaches, some more expensive than others. Full DOM parsing and detecting constructed DOM elements is sure-fire, but slow, requiring a browser with parser and DOM representation. Regular expressions are often frowned upon, but they may get a bad rap in solving this problem. This isn't trying to parse HTML, only identify if it is likely present. A heuristic, not a definitive answer.

Made more complicated by virtue of the fact that browsers will accept almost anything and go out of their way to attempt to parse it HTML.