Bluesky reply bot checker
- ID
- 29195
- Status
- summarized
- Published
- 28 Sep 2026, 2:41 AM
- Fetched
- 30 Sep 2026, 1:02 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Sep/27/bluesky-bot-check/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 5.5
- Created
- 30 Sep 2026, 1:03 AM
- Tags
- Audience
- developersvibe_codersai_agent_usersai_ml_learners
What happened
Simon Willison published a small tool that checks a Bluesky profile for signs of being an automated reply bot, built by vibe coding with Opus 5.5. The heuristics it uses are concrete: replies posted within seconds of another account's post, accounts that never publish their own posts, images or links but only reply to higher-follower users, and replies containing question marks. He notes Bluesky still offers a free, useful API for this kind of investigation, unlike Twitter/X, where the same bot behaviour exists but is harder to inspect.
Why it matters
If you run automated replies on Bluesky for marketing or support, this heuristic list (sub-second replies, zero original posts/images/links, question-bait) is effectively the detection spec you will be judged against. If you build on Bluesky's free API instead, you can replicate this kind of profile audit in an afternoon, which is not possible on X without paid API access. Nothing here affects Malaysia specifically; treat it as a tooling and AI-misuse pattern, not local news.
Discussion angle
Compare rule-based bot detection against a trained classifier: which of Willison's four signals (reply latency, no original content, replies only to high-follower accounts, question marks) would survive an adversary who deliberately posts one original image a day and waits 60 seconds before replying?