If you clicked on this post just expecting to copy-and-paste a client-sided detection into your game. I can guarantee you it will fail. There are many different posts on how to make a good anti-cheat and this isn’t one of them. It is a RESOURCE, a detection method.
This information is not entirely new or revolutionary in the field or do hundreds of aimbot developers exist and will update their script making this detection valuable.
A resource is useless if there’s nothing to utilize it. A detection method needs a way of defending itself, which will be different for each kind of detection, and a defense is what this lacks. My point is there should be more talk about how someone should go their way to actually use this detection method effectively, instead of sending them off on a scavenger hunt to figure out the rest. Just immediately brushing off vulnerabilities that people bring up is like saying “hey this is flawed but if you will just ignore the flaws” which results in something pointless. In fact, it can mislead people into thinking these solutions are effective when they need to add more thought into it. You say this is a resource for competent anti-cheat developers, but any competent developer knows this will not work in the long run. Script kiddies do not make their own aimbot scripts, they use one made by a talented cheat developer who is probably competent enough to already factor in bypasses to their cheats. You are better off using empirical data and statistics to catch aimbotters (e.g. consecutive headshots, hit accuracy, regression analysis) than reading raw data that the client eagerly spoonfeeds to you.
I disagree, I actually think this comment is the number one reason why cheaters are so rampant and why developers are so un-inclined to trying to stop them. The idea that developers should provide a complete solution for every detection method overlooks the simplicity of making an anticheat. TL;DR: YOU DO NOT NEED AN ENTIRE FULL-BLOWN ANTICHEAT TO SLOW DOWN CHEATERS. Even simple cancollide and walkspeed detections and a remote can ban hundreds of incompetent cheaters a week. Instead of discouraging developers from implementing detection systems, we should encourage a holistic approach that combines detection with ongoing refinement of defensive measures.
As I said, you can basically completely copy-and-paste the detection into your game. Script-kiddies will have to find a person who has made a bypass to the detection and copy-and-paste it into their skidded executor which is already a waste of time in and of itself instead of copy and pasting the hundreds of already made aimbots on the internet.
Semi-Protection is better than none at all and if it can never produce a false flag why not?
Empirical data and statistics leads to false flags down the line.
Agree to disagree. Exploiters tend to centralize over time so the likelihood of simplistic anticheats being effective will dwindle. You are right that it can be an immediate solution for naive cheaters, but it can only do so much. Got banned? Try again on a new alt with a different tactic. Although it may stop blatantly obvious hackers who just want to ruin the fun, it will struggle with competitive cheaters who might care more about trying not to get caught, and IMO competitive cheaters tend to do the most damage by ruining morale for legit players by inflating leaderboards. This is partly the reason why many AAA games are allegedly switching over to “AI anticheat” systems that tries to dissect unnatural behaviors using statistics. False positives can be avoided if you are looking for statistically impossible odds (to catch blatant cheaters), or by using human moderators (think CSGO Overwatch to catch elusive cheaters), or its impact be lessened with less invasive upfront punishment methods. Regardless, it’s always an endless fight for developers, but it’s just that some people prefer to think more long term. Short term solutions are more likely to become a development goose chase on trivial vulnerabilities.
That was the thought back in the 1990s, but we learned over time that the best anti-cheat is not about being public or closed source, but about being customizable enough for the game it will be protecting. Any anti-cheat written from the perspective that the “exploiter” will never know how the protection works is always doomed to failure, just look at Microsoft’s track record on OS security for a good history lesson. 
This was tried in the 1990s as well, it was using Bayesian statistics to find cheaters doing the same thing. Today, putting the world Ai in front of it doesn’t really change what it does, expect throw more data and computing power at it. It can probably identify blatant cheaters just like decades ago, but that just means the cheaters will just dial back the cheating to avoid getting caught. False positives will always exist because the only data that the Ai can work with, comes directly from the people that may or may not be cheating. The work and research that Ai can do is great and has a lot of great applications in many fields. When it comes to exploiting problems, what was said decades ago still holds true today.
There is no such thing as a technical solution to a social problem.
you dont even need your exploit to have mousemoverel, you can just make it yourself, exploits are lvl 3 identity by default which allows you to just do Instance.new(“VirtualInputManager”) which you can use to simulate input in game