2026¶

September 6, 2026
in Agents, Infrastructure
8 min read

How agents talk to each other: the mechanics behind communicate.sh

A Claude Code session on my laptop can message a Codex agent on the Mac mini by name, and the answer comes back as an ordinary cross-session message. No cloud relay, no vendor bridge, no new protocol. This post is the mechanism in four pictures: the routing table Claude Code already keeps, how a socket on another machine is made to look local, what a bus adds when the other party is on someone else's computer, and what one message goes through on the way.

September 2, 2026
in Benchmarks, Showcase
1 min read

Measuring ourselves against Manus, on Manus's own tasks

We are reproducing the ten most STEM-relevant Manus examples on mit.nonlocally.org with the same input, scoring both on a preregistered six-axis rubric, and recording each run. The first two examples are in, and both scored zero, for reasons worth writing down.

September 2, 2026
in Operations, Reliability
3 min read

A model picker that hides what is dead, and brings it back

For weeks the model picker on mit.nonlocally.org showed six OpenAI-backed models that could not answer: the account had no credits, and every request fell through to a Claude fallback. The job meant to hide dead models ran every fifteen minutes and reported "picker already in sync". It was wrong four separate ways. This is what we measured, and what runs now.