AceGuardian Open-Sources Superuser Detection Code Following Targeted Poker Software Compromise
AceGuardian has open-sourced its superuser detection models on GitHub after a scandal where tampered poker software exposed hole cards to cheaters on sites including GGPoker. The tool flags anomalies via equity comparisons, oracle folds, bluff index and timing, having operated with seven platforms since 2019. A reviewed case flagged 72 hands in 757 dealt at $25/50.
SCCG Take — Platforms gain direct access to battle-tested detection logic that requires multi-signal confirmation, enabling faster post-incident audits without custom builds. This reduces detection gaps exposed in the June 2025-June 2026 compromise.
AceGuardian Research by A5 Labs has published its complete code repository on GitHub. The release provides online poker operators and researchers with tools to run automated detection on hand histories after a high-profile superuser incident.
As reported by PokerNews, the scandal involved a remote-access Trojan hidden in tampered versions of Jurojin Poker and IntuitiveTables. The compromise occurred through a highly targeted operation between June 2025 and June 2026. It affected more than two dozen high-stakes players, including suspicious activity on GGPoker that led to account bans.
One reviewed heads-up case following the September 29 incident flagged 72 suspicious hands. The suspect played 757 hands at $25/50 over 14 sessions across ten weeks, with 95% heads-up against one opponent. The player won approximately $45,000 in those sessions.
Detection Metrics and Operational History
AceGuardian has run anti-cheat systems for operators since 2019 and now supports seven platforms. It processes tens of millions of decisions daily to identify collusion, bots, real-time assistance and superusers. The models evaluate post-facto when all hole cards are known, using equity comparisons, oracle folds, bluff index, win-rate outliers compared to bb/100 against similar samples, and decision timing measured against a player’s baseline.
No single metric triggers a flag. Signals must align across many hands. The company accepts direct hand history submissions from operators or players for analysis. The full Python and data pipeline are available on GitHub for internal implementation. Dr. Thanh Tran is founder, CEO and CTO. John Andress serves as head of game integrity.
Operators can contact [email protected] for submission details. The release equips platforms with the exact methodology applied in the reviewed case.
Reporting: PokerNews
Generated by SCCG’s automated editorial system from published source reporting. SCCG Management holds editorial responsibility.
Gaming, betting and prediction markets — the desk’s read, every weekday.
Subscribe →