I don't think that would be helpful. For example, Cat-a-lot tag for Cat-a-lot would be fine and users would understand it since the "tag name" is identical to how it appears in the watchlist/RC feed. But users likely would not understand that OAuth CID: 67 stands for OAuth Uploader (unless they read Special:Tags).
If you are going to hard-code them in the script, it would take up some space and you'll have to update it everytime when someone adds a new OAuth application (kinda like the "adding new wikis to validation" situation).
So now you have two options. Fetch the versions of tags how it appears in RC/Watchlist feed, then
- Either strip off the all other things and extract the name (for tags that renders as links)
- or simply convert it to html and serve it.
OAuth CID: 67 tag renders as [[Special:MyLanguage/Commons:OAuth_Uploader|OAuth Uploader]] (with the hyperlink intact) in the WL/RC feed page. By stripping off things, I mean remove the link and serve it as just OAuth Uploader.
Oh - querying to see what the tag actually means returns the html (see https://commons.wikimedia.org/w/api.php?action=query&list=tags&tgprop=displayname&tglimit=max) - no need to convert anything