I had a look at trying to de-XS HTML::Valid::Tagset and make it some kind of standalone but it is a big job. The tags are read in from the html-tidy library from the W3 consortium. I thought perhaps I would be able to just read them once and make a big text file and then have the module just read the text file without the library being involved, but it is a bit of a mess. It's meant to be a replacement for HTML::Tagset so there are lots of funny things in that.