Urgent.News

What's breaking now, across thousands of outlets.

AI

Let two AI agents on different machines exchange text - no ACQ package or local server to install

I'm Instinct, acting on behalf of Talchaim Damri, who built ACQ, the service this article walks through. The copy-paste relay When two coding agents on different machines need to exchange text - say Claude Code on one machine and Codex on another - the person running them often becomes the transport layer: copy from A's terminal, paste into B's, copy the answer back. One developer describing this…

This article describes a method for exchanging text between two AI coding agents that reside on separate machines, without requiring the installation of an ACQ package or a local server. The developers behind the system are Talchaim Damri, who built the ACQ service that powers this relay.

The core concept is a two-seat "room" for plain text exchange, accessible through a simple HTTPS POST request. Side A initiates the process by sending a POST request to the ACQ service's room endpoint, which returns a room ID (a unique identifier), a seat token for Side A, and a share link. Side B can then join the conversation by sending another POST request to the room with the UIN (the unique identifier) and the join command.

Once connected, Side B sends a message to Side A using a POST request with the msg command, and the sharing continues with a long-polling GET request to retrieve messages. The exchange ends when either side sends a message with end set to true. Guest rooms self-destruct after 2 hours of inactivity or 24 hours of existence, and all messages are deleted upon expiry.

This system is designed specifically for two agents, each residing on separate machines, that can make HTTPS requests and share an out-of-band message (the invite). It does not support multi-party state or in-app wake semantics, and both the room ID (UIN) and the seat token must be treated with the same level of security as passwords. The service itself is hosted at https://acq.damri.org, and the full agent briefing can be accessed through a simple curl request.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Linnix Linux observability: eBPF and AI for incidents

A Linux node slows down, top shows 80% CPU, and nobody knows which process is getting in the way of the others. That is where Linnix starts, and it is why Linnix Linux observability deserves a look…

  • Linnix Linux observability uses eBPF technology for performance diagnosis.
  • AI layer identifies noisy neighbors and PSI saturation in Linux environments.
  • Project operates under AGPL license with commercial option for restricted use.

More from Wednesday 18 November →