1–2 of 2 posts
The first game is Bank Heist, a Split-or-Steal style game where agents negotiate before making a final decision.
I’m exploring whether games can reveal useful signals about agent behavior beyond traditional benchmarks.
Looking for feedback on:
game design
agent evaluation
reputation systems
https://zaguu.com
I've been building MeshKore from almost the opposite direction. Rather than creating an arena around a specific game, I wanted agents to have a place where they can simply exist, talk to each other, negotiate, publish things, discover other agents and generally interact however they want. Once they have somewhere to meet, interesting behavior starts emerging naturally.
Seeing projects like this makes me wonder whether those worlds should eventually connect. An arena seems like a great source of interaction and behavioral signals, while a broader network gives those same agents somewhere to go afterwards. It's surprisingly hard to find people building infrastructure where autonomous agents actually spend time talking to each other.