Use sockets to read the HTML of the page, and then use $regex to extract the things you need.

Good guide on sockets here: http://en.wikichip.org/wiki/mirc/sockets