Automated reply bots that have long been a nuisance on Twitter are now appearing on Bluesky as well. The author of the tool notes that, with a substantial follower count, posts on Twitter often attract dozens of mindless automated replies — and similar behaviour has started to show up on Bluesky.
Unlike Twitter, Bluesky still provides a freely available and useful API. That availability does not stop bots from operating, but it does make investigating them easier because researchers and developers can access activity data directly.
Using Opus 5.5 "vibe code," the developer created a tool called the Bluesky reply bot checker that inspects any Bluesky profile for signals indicative of a reply bot. The checker evaluates several heuristics, including:
- Rapid replies: it flags accounts that reply to other posts within seconds.
- Lack of original content: it looks for accounts that rarely or never publish their own posts, images, or links but consistently reply to others.
- Targeted replying to higher-follower users: accounts that mainly reply to posts by users with more followers are treated as more suspicious.
- Use of question marks: the tool notes replies containing question marks, reflecting the creator’s particular annoyance with bots that attempt to bait users into responding to questions no human asked.
The tool generates probabilistic indicators rather than definitive labels; it infers likely bot behaviour from observed patterns. The aim is to offer an accessible means for Bluesky users and researchers to identify and investigate potential reply bots on the platform.
While such tooling can help reduce automated, meaningless interactions and improve the quality of discourse, its accuracy and effectiveness will depend on further testing and refinement.
