I operate a few hundred bot accounts that syndicate US NWS weather content. Yeah, they are bots, but they are hopefully good bots (they only post, no following, likes, etc), just syndicating NWS content in real-time.
I have been running these for many moons since 2009.
The problem I have is that every so many months/years, I get ensnared within some new twitter anti-bot spam code and I have to manually get the bots unlocked. An example error is the generic:
{"errors":[{"code":326,"message":"To protect our users from spam and
other malicious activity, this account is temporarily locked. Please log in
to https://twitter.com to unlock your account."}]}
So the unlock procedure is always
- visit twitter.com as the bot account, OK
- Prove I am not a bot by doing the captcha and clicking on cars/fire hydrants/traffic signals (I am very good at this now)
- Failing to provide my cell phone number as the UI indicates it is being used by some other account, but unsure which one.
- Fill out the appeal form
- wait for the email to arrive (sometimes it comes immediately, sometimes many hours later)
- respond to the email with desperate pleas to talk with somebody regarding this, I write generic babble these days as what I write does not seem to matter
- wait one to a few days for the automated response to come saying that the account lock out was in error.
I would love to know how I can keep my bots from hitting whatever is currently triggering the spam 326 errors. Or get my bots whitelisted somehow. Or perhaps if twitter is really against these accounts and want them deleted, I can pull the plug on this and start doing other things during my day than captcha clicking 
Thanks