June 18, 2026
Agents of chaos, meet agent chat
Ask HN: Is anyone using the A2A protocol?
Google’s agent chat idea is getting side-eye, shrugs, and a few corporate believers
TLDR: A Hacker News user asked whether anyone is really using Google’s system for letting AI assistants talk to each other. The community response was mostly skeptical, with critics calling it impractical and overbuilt, while a smaller camp said big companies still see it as the next big thing.
Someone on Hacker News popped the question: is anyone actually using Google’s A2A, short for “agent-to-agent,” a system meant to let AI assistants talk to each other directly? The post itself was curious and almost hopeful: maybe the real point of contact in the future isn’t an app or website, but your AI helper, and maybe those helpers will need to call each other. Sounds futuristic. The comments, however, showed up with the emotional energy of people inspecting a concept car and asking where the wheels are.
The biggest mood was skeptical-to-brutal. One commenter flat-out torched it as something “designed on paper” without enough thought for speed or cost, which is basically the internet equivalent of saying, “nice slideshow, shame about reality.” Another said their team built something similar with a registry and messaging setup, only to realize they didn’t actually need another agent on the other end—ending with a painfully funny “So, no.” Ouch.
Still, this wasn’t a total pile-on. One voice insisted big companies are into it and called it the future for enterprises, linking to a real-world example. But that optimism got drowned in a sea of “cool idea, unclear payoff.” Even the linked earlier A2A discussion, with 280 comments, loomed over the thread like unfinished business. The vibe? A2A has intrigue, but the crowd seems stuck between “this changes everything” and “this is just more AI theater.”
Key Points
- •The post asks whether anyone is currently using Google’s A2A agent-to-agent protocol.
- •The author reviewed A2A about six months earlier but says they did not then understand how to use it.
- •The post places that uncertainty in the context of a period when many people were still figuring out agent systems.
- •The author notes that the MCP protocol became a major point of attention during that time.
- •The author now believes direct agent-to-agent interaction may become important as agents accumulate tools, services, data, and contacts.