Hey everyone! 👋
I've been working on a project called SentryX Guard for about 3 years now, and I recently decided to make it publicly available.
I figured I'd share a bit about why I started building it, what it actually does, and some of the challenges I've encountered along the way.
The problem that got me started
If you've ever managed a FiveM server, you've probably dealt with cheaters or players who just can't seem to follow the rules.
You ban them, move on, and think that's the end of it.
Except sometimes, it isn't.
They come back with another account, different identifiers, a spoofer, and suddenly you're dealing with the exact same person again.
Anti-cheats are great at detecting cheats, but what happens after someone gets banned?
That's the question that originally got me interested in building something different.
So I started building SentryX Guard
The idea was pretty simple: instead of focusing on detecting cheats, I wanted to focus on identifying players who try to evade bans by changing their identity.
Of course, the implementation turned out to be a lot more complicated than the idea itself lol.
Over time, the project evolved into an anti-spoof system with an identity correlation engine, connection verification, and a web dashboard where server owners can monitor what's happening.
The goal is to help server administrators identify suspicious returning players before they can rejoin.
The interesting part: identity correlation
One of the biggest challenges is figuring out whether two different sets of identifiers actually belong to the same player.
You can't simply rely on a single identifier.
For example, two legitimate players might share the same public IP address. That doesn't automatically mean they're the same person.
And on the other hand, someone attempting to evade a ban might change several identifiers between connections.
So a lot of the work has been about understanding relationships between identifiers, tracking historical connections, and trying to make decisions without punishing legitimate players.
False positives are just as important as detection accuracy.
Blocking someone innocent because two connections look similar isn't exactly a great security solution.
Three years later...
What started as an experiment has grown into a much larger project than I originally expected.
There's now a backend engine handling connection verification, real-time monitoring, identity investigations, and a dashboard for managing server protection.
I've also spent quite a bit of time on the less exciting parts: infrastructure, databases, reliability, and making sure the system remains responsive.
And honestly, there's still plenty to improve.
I'm not going to claim it's impossible to bypass. I don't believe any security system can realistically make that promise.
It's an ongoing process of testing, finding weaknesses, fixing them, and learning.
Why I'm sharing it now
After working on it for so long, I realized that keeping everything private wasn't going to help me improve it much further.
So I recently made SentryX Guard publicly available, including a free option for FiveM servers with up to 200 players.
At this point, I'm particularly interested in hearing from server owners who regularly deal with banned players returning, as well as other developers who have worked on similar problems.
I'd love to get feedback on the detection accuracy, false positives, reliability, or even the overall architecture.
If you're curious, here's the project:
And if you've ever worked on identity correlation, fraud detection, or anything remotely similar, I'd genuinely love to hear about your experience.
Still learning, still building :P
Thanks for reading!
Top comments (0)