Question regarding rules of automated tweets with hashtags



Hi all,

I have a question to ask about the Automation Rules and Best Practices.

In said rules, there is the following line regarding automated tweets:

Automatically Tweeting to trending topics
Trending topics are the most Tweeted-about topics on Twitter at any moment. Posting automated Tweets to trending topics can degrade the experience for other users and is prohibited.

As I understand it, there is no rule against automated tweets that include hashtags. What I want to clarify is whether the rule means one of these:

A) Automated tweets must never tweet to trending topics under any circumstances, and apps are therefore required to use the GET endpoint for trending information to ensure no hashtags are included in a tweet at the time of posting that appears in the trending topics (I assume an app would accomplish this by querying the GET trends/place endpoint with the global parameter 1, but it isn’t clear)

B) Apps that deliberately attempt to find trending topics and tweet to them regularly will be restricted or suspended

I run an app that posts artwork with relevant hashtags on an automated basis, and essentially I’m looking to see whether tweeting something with a hashtag that just so happens to be trending at the time violates this rule or whether the rule is more of a “don’t try to tweet to trending topics to boost your followers” sort of thing.

Automation best practices: Multiple links in automated tweets

As I understand it, the rule is certainly focused on the “do not game the trending topics” situation.

However, our antispam systems are automated and look for patterns that may conform to known or suspected bad actors, and these can include (for example) frequent uses of hashtags, @mentions, and URLs. It’s hard to put specific conditions in words for these, as often these systems are acting on machine learning. Just wanted to give you a heads-up on that.


Yeah, that’s what I’m trying to figure out. I use my app to post tweets of the form:

artwork name, by artist name or twitter handle: URL to artwork #(1-3 hashtags)

The reason I do is this because I want to properly credit the artist who made the work, by providing the direct URL to the artwork on their site and providing either their name or twitter handle if they have one, but unfortunately that’s more likely to make the antispam systems flag it up…

Ideally, I’d like to be able to post tweets every 5-10 minutes for 12 hours a day with an image and a maximum of say 2-3 hashtags per tweet, but I realise the systems that look for bad actors are likely to potentially flag that on occasion even if it isn’t spam or isn’t duplicated - so I’m trying to figure out exactly what I can do to minimise the chance of that happening.

From a personal standpoint, I’m not too concerned if the antispam systems occasionally restrict my app by mistake, so long as that gets recognised and then subsequently the restriction is lifted. I just want to be absolutely sure that that’s the only reason it gets restricted.

On that note - if an app is write restricted, and has its restriction lifted due to it being mistakenly restricted after being mistakenly flagged up by the antispam systems (as was the case the second time my app was restricted quite recently), does this count against an app in terms of having restrictions/violations recorded?


My understanding is that if an app repeatedly gets caught by the system, there’s a likelihood that at a certain point it would not be re-enabled. For your information, humans are also involved the review process if this happens repeatedly, but if something is triggering the rules, then ultimately it may continue to do so, so it would not be re-enabled after a number of times.

Totally get what you’re aiming to do with your app / bot. With the rules and systems as they are right now, all I can do is share these facts in an effort to help - I don’t have many suggestions that would necessarily help to skirt around the issue.


Hmm, fair enough. I guess the best thing I can do is to hope it doesn’t get flagged, probably avoid using hashtags, and if it does get flagged again to ask support what I can do or what they would want me to implement to try and avoid future issues. Given it’s an adaptive algorithm I doubt there is an awful lot to be done there, but I’ll just have to do whatever I can (and whatever support asks of me) and hope that’s enough.

Anyway, thanks for your answers Andy, I’m grateful for the clarification. :slight_smile: