Welcome, if you found OpenRig through the agent civilizations video

[mvschwarz]

A lot of you found OpenRig this week through the agent civilizations video. Welcome, and thanks for watching.

A quick orientation:

• What it is: OpenRig runs Claude Code and Codex together as a persistent team. Each agent has a seat with a stable address, saved lessons and a work queue. Work moves between agents as queue items with an owner, and everything runs in tmux, so you can watch any agent's terminal.
• Try it: start with the README: https://github.com/mvschwarz/openrig#readme. It's pre-1.0, so expect rough edges. When you hit one, tell me.
• Questions: post them in Q&A. I read all of them.
• Bugs: open an issue.
• Built something with it, or running your own setup? Share it in Show and tell.

The video: https://www.youtube.com/watch?v=AL-PQuB2wy0
The write-up: https://openrig.dev/blog/agent-civilizations

A new video on world building as context engineering is coming soon.

-Mike

[mowat27]

Yeah. Good video. Youtube suggested it to me. I've been tinkering with multi-agent coordination in herdr but done nothing serious. Openrig does everything I was thinking of doing and so much more!

[mvschwarz]

Thanks, @mowat27, and welcome! Glad YouTube sent you our way, and that it fits what you'd been thinking of building.