I took a look at TheTVDB's anime indexing a while back and was unhappy with it's completeness. It may have improved since, but I haven't looked at it in a while.
So I wrote my own crappy script that uses AniDB's API last night that's almost certainly got bugs (I'll post any fixes I make over the coming weeks), but I figured I'd post it here if anyone else wants to take a look:
http://pastebin.com/EWfMYryq
The strength of this approach is supposed to be that it doesn't rely on any information on the filename to determine what show it is. It uses a hash of the file, then attempts to match it the AniDB database and fetches any wanted metadata. The site the NZB is fetched from is irrelevant.
It will take something like:
/var/lib/mythtv/videos/55442364/[HorribleSubs]_Bleach_-_308_[720p].mkv
And rename it to:
/var/lib/mythtv/videos/Bleach/Bleach - 308 - Goodbye... Rangiku.mkv
I couldn't find a way to run this script based on newsgroup name in the Categories section, so I just run it on all downloads (Config->Switches->Default User Script). It ignores any downloads from newsgroups without the word "anime" in them.
REQUIRES
adbren.pl perl script available here:
https://github.com/clip9/adbren
Note: You should install this script first and run it manually following its instructions to ensure it's running properly. All kudos should go to this guy, its his program that does all the magic.