Could this pattern "^\s[A-Z0-9]{64}\s$" simplify your processing?
or "^[A-Z0-9]{64}$" with .strip()

In most (maybe every) cases, when the config file is hardcoded, we could even get to the situation, that fedpkg-stage script is used. It has its own config name. You can use self.name (which contains fedpkg or fedpkg-stage) to distinguish the script.

self.name instead of const string.

Will this write the whole (global+user) config to the user config?
If yes, this would effectively do a lot of changes. I prefer to do just partial changes in the token section. But it needn't be much more complex. What about use ConfigParser for loading separate user config, update it and write?

We could also think about the detection of the expired token. Earlier I sent you a method on how we can verify the API functionality. If it fails, we can show a hint about the new methods to the user.

I updated the code to reflect the optimizations that you suggested, apart from the final optimization concerning checking the api token. I think that I'll open another ticket later for this since it adds another layer of complexity to the function and will require writing a batch of Mock tests.

after_more seems to be useless here. Both commands could be moved to others on line 99.

Made changes to the autocomplete script, included the names of the new functions on line 99 as requested.

Checked and it looks good.

Commit 92400b5c fixes this pull-request