Advanced behavioral analysis powered by Machine Learning. Optimized for the Crystal PvP meta.
Servers that collaborate with us
Most anticheats look for cheats they already know. Anything not on the list walks straight through. We do the opposite: learn what normal play looks like, then flag what doesn't fit. Private clients included.
Analyzes combat patterns 20 times a second, detecting anomalies invisible to traditional checks.
Execution happens on the cloud, so the processing weight isn't on your server. Stays at 20.0 TPS flawlessly.
Syncs bans across hundreds of servers. If they cheat once, you'll know instantly.
Hooks into Lunar, Badlion, and Feather to instantly ban unallowed mods used in tier testing.
Open the full benchmark report, methodology, and live references.
Open page →One jar. Spigot, Paper, Folia and Minestom, on 1.20.4 through 1.21.x. Drop it in, restart, done. No proxy module. No client mod. No key to paste anywhere, because your licence is built into the download.
Natively hooks into the most popular PvP clients to ban unallowed modifications and payloads, ensuring 100% fair tier testing and competitive play.
Instant detection of injected modules and unallowed mods natively via webhook APIs.
Official site →
Plugin message hooking to strictly verify client integrity and allowed mod lists.
Official site →
Native support for banning restricted Forge/Fabric mods through handshake packets.
Official site →Everything from in game. Pull up a player's flags, watch the replay of the moment they tripped, change how strict a check is, ban. All of it lands in the audit log.
Global Bans
Caught on one server, marked on all of them. Someone who shows up already flagged elsewhere is treated as a risk from their first tick, so hopping servers no longer buys a clean slate. It never bans on its own. That call stays yours.
I've been backing this since before it was a finished product. On one of the biggest practice networks in Europe people try everything on you, every single day. It holds up, and I don't spend my evenings arguing about bans anymore.
Cheaters were killing my SMP. People were leaving and the server was starting to get known for it. This turned it around. The players who left actually came back.
I ran it on my practice server back when nobody knew about it. It caught stuff nothing else was catching. I ended up helping the team make it better because I wanted it to win.
Quick answers about Ecstacy deployment and detection behavior.
No. All inference runs on Ecstacy Cloud infrastructure, not on your hardware. The plugin only collects and transmits detection data, with minimal overhead and no measurable TPS impact.
Crystal Aura detection reaches 97.0% precision and 91.1% recall, an F1 score of 93.9%, measured per-window on held-out high-tier Crystal PvP match data. Anchor Macro model achieves 99.0% precision and 95.2% recall (97.1% F1 score). These are evaluation figures, not a promise about your server: a model measured on high-tier competitive play behaves differently on a casual SMP, in both directions. The full methodology, and what the numbers do not cover, are published on the benchmarks page.
Ecstacy supports Minecraft 1.20.4 to 1.21.x on Spigot, Paper, Minestom and Folia servers. Installation is a single plugin jar. There is no proxy module, no client-side mod, and no licence key to paste anywhere, because your licence is built into the download.
Ecstacy scores behaviour rather than matching known cheat signatures, so it catches blatant and deliberately toned-down cheating alike. Evidence accumulates across a session, which is what exposes players who only enable a module for a few fights at a time.
An AI anticheat uses machine learning trained on real gameplay to recognise cheating patterns. Where a rule-based anticheat can only check for signatures somebody already wrote a rule for, Ecstacy analyses player behaviour twenty times per second and judges how far it sits from legitimate play. That is why it flags cheats it has never specifically been told about.
Crystal Aura detection analyses combat timing, rotation and interaction sequences twenty times per second, using models trained on real high-tier Crystal PvP matches. It reaches 97.0% precision, so about three flags in a hundred are false positives.
You're about to let someone else decide who gets banned on your server. Here is how to reach us, what happens to your data, and what you can check for yourself before paying anything.
Discord is the fastest route and where tickets are handled. For anything billing or legal, write to Contact Support. Security issues have their own channel: see security.txt.
Hosting is in the EU. For your players' data you are the controller and we are the processor. The terms are in the Data Processing Agreement, and the Privacy Policy sets out exactly what is collected.
The detection figures on this page come with their methodology. Uptime is public and unedited on the status page, incidents included. The plugin API and its source are on GitHub, and the full documentation is readable before you buy.
Detection models, benchmarks and documentation last updated . Ecstacy has been in production since 2024 and is not affiliated with Mojang AB.
Secure your server with the most advanced ML Anticheat on the market.
Get Started
