You can start with URI::Find or build a little parser using HTML:TokeParser::Simple; our Monastery is full of examples, just use Super Search :) Added: for example Extracting href's.

Ciao, Valerio