Can Replying Too Much Get You Flagged on X?

You’ve heard the advice: reply more to grow. Then you see someone say they got shadowbanned for it. Both can be true. Below: what X’s code says about reply spam, what actually looks like spam, how to check your own labels and a daily routine that stays safe.

Can replying too much get you flagged on X?

It can, if your replies look like a bot wrote them. X’s Authenticity policy bans “bulk, aggressive, high-volume unsolicited replies”. The words that matter are bulk and unsolicited. Lots of replies that answer the post are normal use.

There is no public number. X’s code hides its real thresholds, and the models that score you are not published. So nobody can tell you “stay under 200 a day”. What you can control is how your replies look.

What X’s code says about reply spam

X added named rules for reply-spam and post-spam bots. They landed in commit 07d159b of X’s open-source repo on October 23, 2026. I compared it with the day before. Here is what changed in the account rules.

From enforcement_user.yaml at commit 07d159b, checked October 25, 2026. A liveness check is X’s name for a check that a real person runs the account.

Single replies get labeled too

Some rules work on one post at a time. The post rules already had a “fast reply spam” rule. It puts SPAM_HIGH_RECALL on the reply for 30 days, so people who don’t follow you won’t see it. Gibberish posts get the same label. AI slop posts get a label of their own.

What the code can’t tell you

Treat this as a rough guide. The file says it mirrors a live config, so production can change without the repo changing. The follower and credibility cutoffs are mock values. Big accounts are skipped before any of this runs, which is why small accounts feel it most. No @XOpenSource post explained the change, and the same commit also dropped some older rules from the file.

What reply spam looks like to X

Spam means repeating yourself and ignoring the conversation. The spam list comes from X’s own policy. The fine list is what I do every day without labels.

The spam list paraphrases the Content Spam and Engagement Spam lists in X’s Authenticity policy.

Two of those catch normal people by surprise. Pasting a saved “great point, I wrote about this here” under ten posts is copypasta, even if you typed it once. And a link in every reply looks like traffic farming, even when the link is useful.

My numbers: 400 replies a day, 1 label

High volume did not get me flagged. My first month I did 50 to 100 replies a day. Later I did 8 posts and 400 replies a day, for 50 days straight.

One labeled post out of almost 12,000 is noise. My account had no labels. One honest caveat: about 90% of those replies were on my own posts, answering people who replied to me. Nobody calls that unsolicited. If all 400 had gone under strangers’ posts, I’d expect more risk. How I picked the posts to reply to is in the reply guy strategy.

How to check if you’ve been flagged

X now shows you. The Under the Hood report lists the labels on your posts and your account for the prior month.

1. 1

   Open x.com/i/jf/under_the_hood.

   You need an account at least one year old with 10 or more posts last month. There is no menu link.
2. 2

   Look at account labels first.

   SpamHighRecall here is the one that hurts. It drops your posts from For You, followers included.
3. 3

   Then count the post labels.

   One or two SPAM_HIGH_RECALL posts a month, like my one, is noise. Dozens likely means something in your pattern looks automated.
4. 4

   Look up any label you don’t know.

   Our label lookup says what each one hides and what to change.

The full walkthrough is in X Under the Hood. For the two spam labels, see SPAM_HIGH_RECALL on a post and SpamHighRecall on an account. Every other label is in the X label lookup, and how labels feed X’s filters is in X algorithm labels and visibility filtering.

A safe daily reply routine

Write every reply for the post in front of you, and spread them out. That covers most of what X’s policy names.

1. 1

   Start with 10 to 20 replies a day.

   Add more while each one still takes thought. I went to 50 to 100 in my first month.
2. 2

   Split them into two or three sessions.

   Twenty minutes in the morning and twenty at night beats one long burst.
3. 3

   Read the post before you reply.

   Answer what it says. Add a number, a story or a question.
4. 4

   Write every reply fresh.

   No saved text, no template pasted under several posts.
5. 5

   Keep links out of replies to strangers.

   If someone asks for the link, send it then.
6. 6

   Tag only people in the thread.

   Don’t pull ten accounts into a reply to get their attention.
7. 7

   Answer replies on your own posts.

   Most of my 400 a day went here. It’s the safest place to reply a lot.
8. 8

   Check Under the Hood every month.

   Compare the post labels with last month. Fix the pattern if they climb.

If you’re already flagged or suspended

Change the pattern first, then wait. In the rules X published, the reply-spam account label lasts 30 days and the 7-day versions last a week. Keep posting normally while it runs out. The label runs on a timer in the code, so going quiet does not clear it faster.

If you’re suspended, the code marks these spam suspensions as not permanent. Appeal once, clearly, and don’t make a new account. X can suspend other accounts it links to you. The steps are in what to do when your X account is suspended.

Frequently asked questions