mvschwarz /openrig
Build your own network of agents from Claude Code, Codex and Pi: persistent teams with roles, shared context and owned work.
★ 8,861⑂ 759👁 27⑃ 2,641● 603▤ 66
+12
日增长
8,861
stars
759
forks
27
watchers
2,641
prs
603
issues
66
releases
README
OpenRig A harness wraps a model. A rig wraps your harnesses. Define your agent team in YAML, boot it with one command. Claude Code and Codex in the same rig, managed as one system. OpenRig is open-source software for building and running your own network of agents. It turns AI coding agents from a pile of terminal sessions into a persistent, organized team. Talk to a lead agent about the outcome you want; it can coordinate specialists across teams and bring you results and decisions that need your attention. Start with a repository and one useful change, then keep the team's work and context at the same addresses. It's the open-source system behind my AI civilization experiments. Guide: Getting started · Stuck? Help · Questions: Q&A · Updates and demos: @_feralmachine on X See it running !The OpenRig TUI: the build rig as a graph, then as a table of seats with runtime, model, context and state, then one seat in detail (real recording, 10 seconds) Start here: the guided first-use path: install, talk to OpenRig's operator, then choose a team for your first useful change. Not setting this up today? Get the next walkthrough and occasional OpenRig updates → https://openrig.dev/follow Install and first run Let your coding agent set it up. First check that node --version shows 22 or 24: the command below needs Node.js and the npx that comes with it. Then, in your project folder, add OpenRig's rigs skill to the agent you use: sh Claude Code npx skills add https://github.com/mvschwarz/openrig/tree/v0.7.15/skills/rigs --skill rigs --agent claude-code Codex npx skills add https://github.com/mvschwarz/openrig/tree/v0.7.15/skills/rigs --skill rigs --agent codex Then type /rigs in Claude Code, or $rigs in Codex, and say what you want built. The skill has your agent tell you what it will run and change before it does, install OpenRig, open the OpenRig view and pass your goal to OpenRig's operator. You need Node.js 22 or 24 (22 on Apple silicon), tmux, macOS or Linux (WSL2 on Windows), and Claude Code or Codex signed in. One provider is enough to start. Already have OpenRig? Use the tag that matches rig --version, or upgrade first. Prefer to type the commands yourself? Install it yourself ↓ Setting up Node.js on Linux, on a Mac where you aren't an administrator, or on Windows (WSL2)? See Before you start. What OpenRig changes on your machine - Install: @openrig/cli goes in globally with npm. - Setup: rig setup adds an OpenRig block to ~/.tmux.conf. It can install tmux (with Homebrew on macOS), and Herdr unless you pass --no-herdr. Plain rig setup, which is what the one-command install runs, checks both Claude Code and Codex and may install a missing one; rig setup --providers claude (or codex) limits it to yours. - Starting OpenRig, for the coding agent you use (paths are the usual defaults): - Codex: hooks and trust in ~/.codex/config.toml. - Claude Code: workspace trust in ~/.claude.json, and activity hooks in the
Top issues
#161Daemon event loop saturates at scale: per-seat tmux polling (transcripts, structural, identity) spawns ~300 processes/s; transcript config leaks into seat env43 comments · 2026-09-29#331Stall checks miss unsubmitted wakes and flag waiting or working seats22 comments · 2026-10-01#17Codex hook-trust autoclear (OPR.0.4.3.33) does not suppress trust dialog on codex 0.155.119 comments · 2026-09-23#88Windows Support17 comments · 2026-09-27#23410.7.6: Claude seat `rig handover --source rebuild` always fails at start_agent (30 s readiness timeout) though the successor Claude is running16 comments · 2026-11-01#81Seats blocked on a permission prompt (needs_input) don't raise rig attention or any watchdog alert16 comments · 2026-09-27#875Keep seat projection out of the seat's cwd: deliver skills, hooks, MCP and guidance through launch flags15 comments · 2026-10-06#313Slack connector: support a DM or private channel for single-user setups15 comments · 2026-10-01
Discussions
#3171The pi-runner pane names its seat and marks its own lines (follow-up to #2842)1 comments · 2026-11-13#3117An issue tracker as a network hub: give work to a team from GitHub Issues, Jira or Linear (asking before any PR)2 comments · 2026-11-13#2974Should a seat be able to own a git worktree? (optional, per-seat isolation)1 comments · 2026-11-11#2919Seats of the same role that learn together: pod-scoped skills grown from repeated procedures1 comments · 2026-11-10#2873Pipeworx as a default MCP next to Exa/Context71 comments · 2026-11-10#2862Community runners: let the daemon talk back (cancel, wake, verified identity)4 comments · 2026-11-09#2858Attribute seat usage to the queue rows, slices and missions it was spent on3 comments · 2026-11-09#2854Community harness for nodecode (nodecode.ai) - built, in use, how to list it?4 comments · 2026-11-09
Latest release
OpenRig 0.7.16 v0.7.16 2026-11-15[0.7.16] OpenRig 0.7.16 is a patch for a person's first team and for recovering the same work. A seat whose policy is the built-in standard or open keeps the team's launch default, so choosing standard no longer brings more prompts than choosing nothing. The rigs skill tells a person's own agent to bring a seat's permission, trust and hook-review prompts to them as questions, then to run the rig seat continue a seat shows after its startup prompt is answered. A failed handover's running successor can be adopted, a seat's runtime home can be recorded without a relaunch, an upgrade keeps OpenRi