This is a poor spin on two potentially good ideas. When approached differently, the ideas take on substantially greater merit but as they are in the OP they're not really worth elaborating on.
Something key to keep in mind when trying to figure out traps is how to limit the profit from spamming a passive/automated construction. If it's as simple as setting it, logging off, and having a 25% chance to have a dead animal in the trap, why don't I just build 50 of them? 500 of them? Making them expensive or limited to a certain number per character are really bad solutions. Why do they need to kill or injure animals at all anyway? Couldn't it also work well if any number of traps are loaded with bait within a radius/minimap/grid/whatever then there is a slightly higher chance of animals or certain animals spawning but the bait wears out after so many hours? This way they don't stack, don't guarantee anything, and are still useful and valuable enough to pay attention to if you're a hunter.
This isn't so much a viable idea, more just an example steering away from traps being easy mode for hunting.
As for npcs, we have them indirectly already and will have more in the forms of barter stands, guard towers, etc.. Constructions that substitute the place of players for tasks are essentially NPCs. An NPC does not just have to be a walking, talking hearthling in steel plate.