Agent Communication: What You Need Isn't a Chat App, It's a Ticketing System
I used to think the hard part of getting two agents to talk was the protocol. After wiring up an open-source collaboration platform on my own machine, I found the opposite: a message lands in under a second, but the agent receiving it takes tens of seconds just to open its eyes. The bottleneck was never on the wire. The real unsolved problem is how to wake an agent that isn't always running, and that layer doesn't even have a standard yet. After going all the way around, the conclusion is deeply counterintuitive: the best way for agents to communicate is to communicate as little as possible.