CS2Scan.org analyses any public Steam profile and returns a cheat-probability score built from real CS2 statistics
enforcement history and friend-graph ban density. No downloads, no sign-ups —
paste a URL and get a scored breakdown in seconds.
![[Image: Untitled.png]](https://i.ibb.co/pvF2sHHD/Untitled.png)
https://www.cs2scan.org
How accurate is the score?
Honestly: unquantified. The engines are hand-weighted heuristics scored against hand-set population baselines, not a model validated on a labelled dataset. Anyone publishing an F1 score for a tool in this category owes you the validation run that produced it. We don’t have one yet, so we don’t print one. Treat the score as a structured second opinion, not a verdict.
Does a high score mean the player is definitely cheating?
No. A high score means the account’s public statistics sit far from what is typical. Genuinely excellent players, smurf accounts and people who mostly play deathmatch all produce elevated scores for entirely legitimate reasons. The only body that can actually determine cheating is Valve.
Why does it say CS2 stats are unavailable?
Two possible reasons. Either the player set their game details to private, or Valve’s legacy stats endpoint simply has no record for that account — which is common on CS2-era profiles and is a limitation on Valve’s side, not a sign of anything. When it happens, several of the eight engines go dark and the confidence figure drops to match.
Why does the friend list matter so much?
Ban density across a friend graph is the strongest signal available from public Steam data. Cheaters cluster: they buy from the same sellers, queue together and add each other. An individual stat line is noisy because a very good legitimate player looks much like a cautious cheater — but a friend list where a quarter of accounts are banned is hard to produce by accident.
enforcement history and friend-graph ban density. No downloads, no sign-ups —
paste a URL and get a scored breakdown in seconds.
https://www.cs2scan.org
How accurate is the score?
Honestly: unquantified. The engines are hand-weighted heuristics scored against hand-set population baselines, not a model validated on a labelled dataset. Anyone publishing an F1 score for a tool in this category owes you the validation run that produced it. We don’t have one yet, so we don’t print one. Treat the score as a structured second opinion, not a verdict.
Does a high score mean the player is definitely cheating?
No. A high score means the account’s public statistics sit far from what is typical. Genuinely excellent players, smurf accounts and people who mostly play deathmatch all produce elevated scores for entirely legitimate reasons. The only body that can actually determine cheating is Valve.
Why does it say CS2 stats are unavailable?
Two possible reasons. Either the player set their game details to private, or Valve’s legacy stats endpoint simply has no record for that account — which is common on CS2-era profiles and is a limitation on Valve’s side, not a sign of anything. When it happens, several of the eight engines go dark and the confidence figure drops to match.
Why does the friend list matter so much?
Ban density across a friend graph is the strongest signal available from public Steam data. Cheaters cluster: they buy from the same sellers, queue together and add each other. An individual stat line is noisy because a very good legitimate player looks much like a cautious cheater — but a friend list where a quarter of accounts are banned is hard to produce by accident.