Thanks for the response Andy.
Yeah, I did see the initial response and like the original author I had already read that page. I was referring to the subsequent lack of information.
Anyways, you said:-
If an application is detected by our automated / machine learning anti-spam and anti-abuse systems (aka “Botmaker”) then it can sometimes be muzzled and placed into a read-only mode, which would be visible via the apps dashboard.
I assume “apps dashboard” means apps.twitter.com? Yeah, I don’t see anything anything there about my app being muzzled. In fact the app continues to work for other authorized users. So this appears to affect just one user (and the behavior is exactly what the original author of this topic describes).
Twitter does not revoke tokens on behalf of users except in unusual circumstances that suggest that an account may have been compromised, which should trigger a visible message on use or login.
This isn’t what I am seeing actually. For the affected user (after re-authorizing) the first unfollow succeeds, the second and subsequent unfollows fail with the message:-
“401:Authentication credentials (https://dev.twitter.com/pages/auth) were missing or incorrect. Ensure that you have set valid consumer key/secret, access token/secret, and the system clock is in sync.message - Invalid or expired token.code - 89”
And if I inspect the affected user’s Apps settings my application that had been previously authorized and appeared in the list is now absent. That looks a lot like a revocation to me. The affected user receives no emails or any other indication that this has happened.
So to sum up. According to your descriptions my app appears NOT to have been muzzled and one particular user appears to be having his application authorization revoked which according to twitter you don’t do.
Not only frustrating but also embarrassing to have this happen to a Customer and be completely unexplainable.