Feed Discovery in HTML pages

Feed discovery: it would be neat if blam could discover feeds from an html page that provides them. Right now I have to manually read the source of the page to extract the feeds

(imported from old Blam Scratchpad)