> So I don't need the files themselves, but just the file 
> names 

Well, you can use the spider option (--spider, or in 
Options/Experts Only: "store html files only") and activate 
Options/Links/Test validity of all links. This will crawl 
the website and check all links. To get the list, look at 
the hts-cache/new.lst file. The current beta release (3.20-
b6) will also include a new.txt files which will be much 
more complete (mime type, scan result, and so on..)