OpenRig
Turn Claude Code and Codex into a persistent agent team: define the topology in YAML, boot it with one command, manage and restore it as one system.
- Source repo
- mvschwarz/openrig
- Stars
- ★ 8.9k
- Last updated
- today
- License
- Apache-2.0
- Primary language
- TypeScript
- FA score
- 79/100 · Good
At a glance
- How it runs
- Works with
- Universal · cross-platformCodex · Claude Code
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A solo developer starts an owner/checker pair (first-project starters) in their repository, with a Claude owner coordinating a Codex checker to ship one reviewed change
- Not a fit if
- Native Windows users or teams needing tested WSL2 support
- Developers who want a single coding agent with no multi-agent coordination
- Environments unwilling to have provider trust settings and hooks written locally
- Website
- openrig.dev ↗ set on GitHub
- Maintenance
- Commits in the last 30 dayslast commit 2026-11-21latest release v0.7.20 (2026-11-20)+9.6 ★/day over 7 days
- Source review
- 79/100 · Good
What does this agent do, and when should you use it?
OpenRig is an open-source multi-agent harness: instead of wrapping a model, it manages the system of coding agents you run together. It runs locally as a daemon + CLI + terminal UI + MCP server, built on tmux and SQLite. You declare topologies as YAML RigSpecs (pods, members, edges, continuity policies), then boot the whole team with rig up, with every agent living in a tmux session you can attach to. It runs Claude Code and Codex natively in the same rig, plus Pi and Oh My Pi via RPC runners. Shipped starter rigs include first-project, conveyor, product-team, and a secrets-manager rig where a specialist agent operates a HashiCorp Vault instance. It requires Node.js 22 or 24, tmux, and currently supports macOS and Linux only.
OpenRig reads YAML RigSpecs (pods, members, edges, continuity policies, CULTURE.md), and rig up creates tmux sessions, supplies seat identity and daemon connection environment, projects startup files, and runs readiness checks. At runtime it provides: cross-agent communication via rig send, rig broadcast, and rig chatroom; inspection via rig ps and a TUI with topology table, graph, and seat detail views; session discovery and adoption via rig discover and rig adopt; snapshot/restore via rig down --snapshot and rig up <name>; live topology evolution via rig grow, rig shrink, rig launch, and rig remove; task queues via rig queue list; a typing guard for human-typed seats; Slack via rig slack manifest; and permission governance via rig policy permissions and rig seat set-permissions. MCP tools (rig_up, rig_ps, rig_send, rig_chatroom_send, etc.) let agents manage their own topology. Architecture: Hono HTTP daemon, domain services, SQLite + tmux + runtime adapters.
- A solo developer starts an owner/checker pair (first-project starters) in their repository, with a Claude owner coordinating a Codex checker to ship one reviewed change
- An engineer juggling multiple AI terminal sessions uses rig down --snapshot and rig up <name> to save and restore the whole team after a reboot
- A team that wants mixed vendors runs Claude Code agents and Codex agents in one rig with shared context and queues
- A user with scattered Claude/Codex tmux sessions uses rig discover + rig adopt to bring them under management
- A larger product squad boots product-team: two orchestrators, implementation, QA, design, and two independent reviewers
- An operator runs the secrets-manager rig and asks vault-specialist to check a managed Vault's health via rig send --verify
Key setup facts
Pulled from the install and usage notes below. The project README has the current values.
- Install / run
npm install -g @openrig/clibun add -g @openrig/cli
- Config files
~/.openrig~/.claude/skills~/.agents/skills- Runtime
- Node.js 22
- Install notes cover
- macOS · Linux
How do you install or deploy this agent?
Prerequisites: Node.js 22 or 24 (22 on Apple Silicon Macs), tmux, macOS or Linux; a working Claude Code and/or Codex account. Install the CLI and preview setup:
bash
npm install -g @openrig/cli
rig setup --dry-runOr install with Bun:
bash
bun add -g @openrig/cliBefore installing, read the README section "What OpenRig changes on your machine" and back up relevant files: setup writes tmux config, daemon startup creates instance state under ~/.openrig, seeds the openrig-skills skill in ~/.claude/skills and ~/.agents/skills, and writes trust and hooks settings into Claude/Codex provider configuration.
How do you use this agent?
Enter your repository, pick a starter (first-project, first-project-claude, or first-project-mixed), preview, plan, and boot:
bash
cd /path/to/your/repository
starter=first-project # or first-project-claude / first-project-mixed
rig specs preview "$starter" --kind rig
rig up "$starter" --cwd . --planrig up "$starter" --cwd .
rig tui --sharedCheck seat readiness and give the owner one bounded task:
bash
rig ps --nodes --rig "$starter"
rig send "dev-owner@$starter" 'Implement <one useful change>. Track the task in the queue and return its ID. Keep it local, verify the behavior, ask dev-check in this rig to check the exact candidate, and record the result and how I can try it.'
rig queue list --destination "dev-owner@$starter" --limit 1000Press Ctrl-b then d to detach from the shared TUI without stopping the dashboard; rig tui --shared returns to it. Browse the full spec library with rig specs ls.
What are this agent's strengths and limitations?
- Cross-vendor teams: Claude Code and Codex collaborate in one rig; the mixed starter (first-project-mixed) works out of the box
- Persistence and recovery: snapshot/named restore (rig down --snapshot, rig up <name>), stable seat identities that survive occupant changes
- Declarative, evolvable topology: YAML RigSpecs with pods/edges/continuity policies, adjusted live via rig grow/shrink/launch/remove
- Agent-managed operation: MCP tools let agents boot topology, message each other, and manage queues; typing-guard protects human-typed seats
- Full operational surface: TUI graph/table views, rig doctor health diagnostics, RigBundle portable archives with SHA-256 integrity
- Platform limits: no native Windows support, WSL2 untested; macOS and Linux only
- Modifies local provider configuration: writes ~/.codex/config.toml, .claude/settings.local.json, trust records and hooks, with no complete preservation/rollback guarantee — back up first
- Requires Node.js 22/24 and tmux; Node 20 is refused at install, and Apple silicon has a documented Node compatibility limitation
- Upgrade migration risk: crossing the 0.5.9 layout boundary requires the multi-phase openrig-upgrade skill migration scripts
- Requires an existing paid Claude/Codex account with model usage billed separately; the React web UI is in maintenance mode
How does this agent compare with similar options?
The README compares OpenRig with Claude Managed Agents: OpenRig is open source and self-hosted, can mix Claude Code and Codex in the same team, runs on your own infrastructure, and the selected providers' model usage costs still apply. A full comparison is at openrig.dev/compare/claude-managed-agents.
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| OpenRig This agent | 79 · Good | CLIFree + model costs | ★ 8.9k | today | TypeScript | Codex · Claude Code |
| NTM (Named Tmux Manager) | 61 · Some gaps | CLIFree + model costs | ★ 465 | today | Go | Codex · Claude Code |
| Agent Manager | 75 · Good | CLIFree | ★ 701 | today | Go | Codex · Claude Code |
| Kangentic | 65 · Some gaps | Desktop appFree + model costs | ★ 150 | today | TypeScript | Codex · Claude Code |
How does FollowAgents rate this agent?
Why each dimension lost points
The README and getting-started guide enumerate setup, startup, and team-launch writes to configuration, trust records, hooks, and project guidance. They describe activity-hook payloads as excluding prompt text and tool arguments and distinguish Claude context metadata, supporting strong data-flow transparency and a basic sensitive-data boundary. The skill requires advance disclosure and consent for machine changes and launches, while allowing reuse of consent for the same launch. Deductions: kernel defaults are broad, and workshop may grant broad access; the supplied material does not establish dependency auditing or mitigation; and the docs disclaim guaranteed backup or rollback. Publisher identity is unverified, so attribution receives only limited credit for repository author and contribution-path information.
The README, quickstart, and setup reference provide concrete prerequisites, readiness checks, partial-start handling, permission-prompt recovery, timeout checks, and recovery commands. Deductions: the packet mixes release, current-source, and route-specific guidance, and some docs acknowledge that maps can become stale or described capabilities may not be implemented; required dependencies are named, but availability depends on local installation, sign-in, and permissions.
The docs cover first installation, one or multiple providers, several team sizes, upgrades, WSL2, and headless SSH. They distinguish kernel, team, and integration-author responsibilities and label draft or unsupported interfaces. Deductions: the rigs skill has a broad trigger that includes mentioning OpenRig or the rig command, and supported environments are concentrated on macOS/Linux and WSL2 with specific Node-version constraints.
The README offers clear entry points, with references, skills, and command examples linked across the packet. Installation, Q&A, limitations, license, and release information are covered. The development skill warns that maps may be stale and code takes precedence; the integration contract separates implemented, draft, and unsupported items. Deductions: naming and compatibility are complex across harness, provider, and seat concepts; version details appear in several places, and the sample does not establish one consistent update-responsibility mechanism. Contribution channels and response targets are documented, but publisher identity is unverified and actual performance is not evidenced.
The product organizes agents across harnesses into persistent teams with roles, shared context, a task queue, an operator, and a TUI. Examples lead to actionable installation, launch, task assignment, and recovery steps; the coordination layer offers clear value over separate terminal sessions. Deductions: setup and operation involve a global CLI, daemon, tmux, provider configuration, and writes across several locations, creating meaningful setup and maintenance costs. The payoff depends on a genuine need for persistent multi-agent coordination.
Multiple user-facing documents corroborate concrete claims about machine changes, data flows, and permissions. The integration contract records versioned implementation boundaries, while the development skill asks maps to identify the source commit checked. The docs generally distinguish stated behavior, user reports, drafts, and inference. Deductions: this is a limited file sample and cannot establish consistency across all implementation and documentation; no execution or independent verification was performed, so confidence is low.
- Kernel defaults are substantially broader than ordinary team defaults; review the actual write locations and selected team permissions and confirm the intended scope before setup or launch.
- The docs do not guarantee preservation or rollback of configuration files; back up relevant tmux, provider, and project configuration first.
- Activity hooks, Claude context collection, and selected MCP services have distinct data flows; inspect local state directories and external-service configuration as appropriate.
Recent releases
Latest entries from the project's GitHub Releases. Notes are quoted from upstream as written, not rewritten or tested by FollowAgents.
-
v0.7.20 OpenRig 0.7.20
Release notes excerpt
OpenRig 0.7.20 tells a person's own agent the consent rules first and where a handed-over task's result is, holds automatic delivery while a seat's tmux server is gone, and, when a parked row's wake is refused at a Claude seat's usage limit, retries it at the stated reset. It also lets a workflow instance stay on one pod, adds an opt-in Claude status-line throttle, and fixes a run of smaller CLI, TUI, scope and workflow problems. Before you upgrade …View on GitHub ↗ -
v0.7.19 OpenRig 0.7.19
Release notes excerpt
OpenRig 0.7.19 improves how a team brings permission prompts to a person, reports interrupted recovery, and starts again after a failed launch. It also adds optional Claude hook delivery, per-seat Linux memory limits, and a versioned interface for apps answering human questions. Before you upgrade Run the upgrade in the environment that owns your daemon, keeping the same OPENRIG_HOME (legacy RIGGED_HOME) if you set it. Stop the daemon, back up its database, install the patch, then restart it. Your agents keep running; do not use rig down for this upgrade. …View on GitHub ↗ -
v0.7.18 OpenRig 0.7.18
Release notes excerpt
[0.7.18] OpenRig 0.7.18 can send a team's seat prompts to the conversation you handed the work over from, restore says more truthfully what it left running, and three settings can be chosen per seat: native input, Claude's auto mode and, on Linux, a task limit. Many fixes come from people using OpenRig. Before you upgrade - **Stop the daemon, back up the database, install 0.7.18, then start the daemon.** Your agents keep running through a daemon restart; don't use rig down, which ends them. Each step runs only if the one before it …View on GitHub ↗
FAQ
Does it require a paid subscription?
What does it change on my machine?
Is YOLO mode on by default?
How do I troubleshoot problems?
What should I watch out for when upgrading?
Is OpenRig open source?
Is OpenRig safe to use?
Is OpenRig still maintained?
Who makes OpenRig?
What is OpenRig's official website?
Thanks — we use this to improve the page.