Bluesky Logo - Growth Tracker

Bluesky Bot Detector

Bluesky Bot Detector

Check if an account is likely to be a bot. Our algorithm analyzes multiple factors including posting patterns, profile characteristics, and network behavior. It is still a beta, take the results with a grain of salt...

bskycheck is free, with no ads. If you find this tool useful, please consider a small tip to keep it that way. Support on Ko-fi

Analyzing account...

This may take a few seconds while we assess bot-like behavior.

Initializing...
Connecting to server...
bskycheck is free and ad-free — supported by its users.
?
Bot Score
Analyzing account...
bskycheck is free, with no ads. If you find this tool useful, please consider a small tip to keep it that way. Support on Ko-fi

Detection Factors

How the Bot Detection Works

Our bot detection algorithm analyzes multiple factors to determine if an account is likely to be a bot:

  • Database Check: We first check if the account appears in our known bots database.
  • Profile Analysis: We evaluate profile completeness, bio content, and avatar usage.
  • Posting Patterns: We analyze posting frequency, timing patterns, and content variety.
  • Engagement Metrics: We assess the account's follower/following ratio and engagement rates.
  • Content Analysis: We check for link spamming, duplicated content, and suspicious domains.

The bot score is calculated using a weighted combination of these factors. Higher scores indicate a greater likelihood of automated behavior.

Note: This tool provides an estimate based on available data. No detection system is perfect, and some accounts may be misclassified.

Frequently Asked Questions

How do I check if a Bluesky account is a bot?

Enter the account's handle (e.g. name.bsky.social) in the checker above. bskycheck analyzes the account's public activity through the AT Protocol and returns a bot-likelihood score in seconds. It's free and requires no login.

What signals suggest a Bluesky account is a bot?

Typical signals include highly regular posting intervals, an account created during a known bot registration wave, an incomplete profile (no bio or avatar), an unusual follower/following ratio, repeated or duplicated content, and link spamming to suspicious domains.

How accurate is the bot detection?

The score is an estimate based on observable behavior patterns, not proof. Most clearly automated accounts score high, but some legitimate accounts (news feeds, bridges) can look bot-like, and sophisticated bots can look human. Use the score as one signal among others.

Does a high bot score mean Bluesky will remove the account?

No. bskycheck is independent and not affiliated with Bluesky; its score has no effect on an account. To see whether Bluesky or a community labeler has actually flagged an account, for example as spam, use the Label checker.

Can I check any Bluesky account for free?

Yes. The checker works on any public Bluesky account and only uses publicly available AT Protocol data. There is no login, no rate limit for normal use, and no cost.

Is this bot detector only for Bluesky?

Yes, this bot detector is designed specifically for the Bluesky / AT Protocol network. It reads public account data directly from the AT Protocol to assess automated behavior patterns.

How are the account types decided?

From public figures only: each account's handle, the date it was created, its followers, the accounts it follows, its posts, and its recent posts when they were read. Each account gets the first type below whose rule it matches, checked in this order. "Ratio" is the number of accounts it follows divided by its followers.

1. Bridged Accounts: accounts from another network (Mastodon and the fediverse, or a website) mirrored into Bluesky by Bridgy Fed: their handle ends in .brid.gy.
2. New Accounts: created in the last 30 days.
3. Inactive Accounts: never posted, or no post or repost for more than 90 days.
4. Mass Following: follows far more accounts than follow it: a ratio above 7 while following over 400, above 4 while following over 1,000, or above 2 while following over 3,000.
5. Star Accounts: over 150,000 followers, following fewer than half that number.
6. Aspiring Stars: over 10,000 followers, following fewer than 30% of that number.
7. Content Broadcasters: over 1,000 followers and 1.5 times more followers than accounts followed, over 100 posts, and more than 60% of its recent posts its own (not replies or reposts).
8. Engagement Specialists: over 100 posts, more than 40% of its recent posts replies, and a ratio between 0.7 and 1.3.
9. Content Consumers: fewer than 50 posts, following over 200 accounts, and a ratio of 0.8 or more (following about as many accounts as follow it, or more).
10. Reciprocal Networkers: a ratio between 0.8 and 1.2 (about as many followers as accounts followed), following over 200.
11. Selective Networkers: over 500 followers, and at least 1.5 times more followers than accounts followed.
12. Standard Users: every other account.

Three rules need an account's recent posts: Inactive (for an account that has posted before), Content Broadcasters and Engagement Specialists. The profile viewer reads them for the account it shows. The follower tool reads them for the accounts on screen and for a sample of 150, so these types can appear as you browse, and their counts cover only the accounts read. The follower tool and the profile viewer use the same rules. The follower tool's audience quality card uses the same meaning of New and Inactive, but it counts signals, and one account can show several: an account created last week that never posted counts as both New and Inactive there, while its type is New, the first rule it matches. A type describes how an account uses Bluesky, not its quality, and it does not say whether an account is a bot: use the Bot checker for that.