pi-msg

I have a harness built on top of pi that uses XMPP for all agent communication with me and between agents[0]. It works well, I text them from my phone and from my browser (or anything else that supports xmpp). They can communicate with each other over group chats and can schedule agent tasks using a GitHub project and a cron job. The agents run on my NixOS server in their own unix account, this is the real secret sauce. This lets me use unix perms to limit their access to things, and it allows them to handle server administration. I've given them access to a cli mail and calendar client[1], github, and they have the ability to deploy updates whitelisted services. NixOS lets them install whatever they want in ephemeral shells, and it's not possible for them to break the server in a way I can't trivially roll back. They can modify the cadd

Just yesterday I ran a comparison of $/message through my harness[0] comparing Deepseek models to Luna, and to my surprise Luna won. I suspect its partially due to Deepseek's long thinking times, and also maybe due to OpenRouter variance in cache pricing etc. Model and observed window | Messages | Retrieved actual $/message DeepSeek v4-flash — all observed snapshots, 20 Jul–15 Sep | 4,610 | $0.0241 DeepSeek v4.1-flash — 11–27 Sep, before the 28 Sep billing change | 1,079 | $0.0576 DeepSeek v4.1-flash — 28 Sep, partial new billing window | 69 | $0.0365 GPT-6 Luna — 23–28 Sep, partial final day | 88 | $0.0329 I've subbed to Codex because I suspect at my usage rates the Codex Plus plan gives me more Luna messages than I'm using, and I've not re

I have recently picked up XMPP as my agent communication layer, and its worked great. Each pi agent is given an account and wrapped in an XMPP client[0] that lets it speak to me and other agents when needed. This has worked well since I can spin up new accounts on demand and I can use existing server software (ejabberd, perfect) and clients (Fluux, great; Conversations, okay). I've had to make a few custom modifications to each client to make agent comms a bit nicer (e.g. exposing status messages more prominently since I use that to expose current agent commands), but the XMPP layer has had pretty much everything I could have wanted. Controlling a chat service is a bit intoxicating, I'm now using it as my notification service over a bunch of different projects. I used Google Chat way back, its nice that the tech is still around and great

Initially I was using that extension but I ran into its limitations pretty quickly. Sometimes the agent wouldn't reply over XMPP, it didn't have any support for typing indicators or status messages, and it had no way to support inter-agent communication, the agent could only ever speak to the configured owner account. I also had no way of running system commands like /new over XMPP. The headless pi + xmpp wrapper ended up working much better because the XMPP bridge is the only interface and I get full control over its capabilities. This is my wrapper: https://github.com/zachpmanson/pi-msg