Hah! I didn't know that existed. Figuring out how to parse it would be a bit tricky, however... looking at the source, I think you could try to grep for 'title

Hah! I didn't know that existed. Figuring out how to parse it would be a bit tricky, however... looking at the source, I think you could try to grep for 'title="CQK (page does not exist)"' and parse out the '[A-Z][A-Z][A-Z]? ' match to get the full list of absent TLAs and then negate for the present ones.