Test, validate and match regular expressions
Matches any token that begins with a single uppercase letter (A-Z) followed by one or more non-whitespace characters. The word boundary (\b) ensures the match starts at the beginning of a word. With the global flag, it extracts all capitalized words from text, useful for finding proper nouns, names, or sentence-initial words. Note that {1} is redundant, and \S+ greedily consumes punctuation and symbols until whitespace.