More on this..
You're thinking about old languages/platforms
The potential "risk" you're describing is something that can happen with non-managed code, i.e. where received data is held in a memory pointer and a malicious user attempts to overflow the buffer and gain access to memory that the code isn't managing.
Things don't work like this in .Net Framework languages - unless you specifically write and declare you want to use unsafe code (which is something you'd only do for legacy reasons).
The socket server for this remote obviously doesn't do this