While it is possible to use a single MultiTargetPath from the player to calculate a path to all following agents, I would first ask you if doing it the normal way is not completely fine from a performance standpoint? Do you have several hundred characters (or thousands) all converging on the same point?
I am not that far yet in development, but it will be for mobile, so when I can squeeze out some performance, I want to do it.
I am not sure how many enemies will be walking simultaneously because I need to test it with some mobile devices and what they can handle.
At the moment its my own curiosity.
While it is possible to use a single MultiTargetPath from the player to calculate a path to all following agents
It’s the other way around: Multiple agents to one player. That’s why I was thinking that using StartMultiTargetPath with multiple start points to one endpoint may fit.
It is possible to do what you want, but it requires a bunch of code. I would strongly recommend waiting until you see that pathfinding is actually a performance issue. Something I doubt will happen unless you have many hundreds of units calculating paths all the time.