OpenRig · 2026-09-27 · notable
OpenRig — Claude Code and Codex agents run as one team from YAML
OpenRig is an open-source harness that runs Claude Code and Codex sessions as one team: a YAML file defines the seats, rig up starts them in tmux, and agents message each other. It gained 781 GitHub stars in a day.
A control plane that turns a pile of coding-agent terminals into a named, persistent team.
Key specs
| GitHub stars | 1,578 |
|---|---|
| Stars in a day | 781 |
What is it?
OpenRig coordinates several AI coding agents, currently Claude Code, Codex, plain terminal nodes and a Pi adapter, as one system. A YAML RigSpec describes the team as pods of seats with edges between them, and starter rigs such as first-project, product-team, adversarial-review and research-team come included. Version 0.5.17, released on 27 September, lets you install it with Bun as well as npm.
How does it work?
The rig up command boots each seat in its own tmux session, writes its startup files and waits for readiness checks. A Hono HTTP daemon with SQLite keeps the state, and a CLI, a TUI with a topology graph, and an MCP server sit on top. Agents talk through rig send, rig broadcast and rig chatroom, and rig grow and rig shrink change the team while it runs; snapshots save and restore a whole topology.
Why does it matter?
Running Claude Code and Codex side by side usually means juggling terminals by hand. OpenRig makes the team setup a file you can version and share, and it can adopt sessions you already started in tmux. It runs on macOS and Linux with Node.js 20, 22 or 24; it also edits Claude Code and Codex trust settings, which the README lists up front.
Who is it for?
developers running several coding agents at once
Try it
npm install -g @openrig/cli && rig setup --dry-run