Can the security manager be disabled by a java agent?

[MadDev]

Hey,

I didn't look too much into it but was just wondering if the security manager has to possibility of being disabled by injecting a java agent upon the startup of the bot and 

setting the security manager and disallowing the bot to set their own policy. 

I wasn't fully able to get it working, but just wanted to let staff know in case this was actually a possibility of an exploit.

Although it would be hard to pull off if you were to affect anyone else since you'd have to add the java agent as an arg on startup of the bot.