Hmmk. In a nutshell, it would just download everything off the site via some HTTP transfers. Preferably done on a loop, each iteration being a file. Brownie points for making it multithreaded. The best way I can think of doing it multithreaded involves chucking all the downloads into a queue, and then assigning each thread an index in the queue. You'd need to parse the JKFiles page for the div tag (or...shudder...table tag) that contains the download. It follows a fairly generic setup, and what you're looking for is the page link to Filefront. From there, it sends...something to the client to invoke a download prompt on the browser. You'd want to intercept this and make it chuck all the .zips into the queue. The main thing we need to do before the big pull takes place is put in some kind of blacklist of files, to prevent files we don't want from being downloaded. But before the big download takes place, you'll want to try it out on one or two files first anyway. That's the basic gist of it anyway.