A lasting team of coding agents
Bring your ideas and feedback to a lead agent. It delegates to specialists, tracks their work and brings answers back to you. OpenRig lets Claude Code, Codex and Pi work together on projects that take weeks or months.
Bring your ideas and feedback to a lead agent. It delegates to specialists, tracks their work and brings answers back to you. OpenRig lets Claude Code, Codex and Pi work together on projects that take weeks or months.
Claude Code · Codex · Pi · Self-hosted
One conversation. A whole team behind it.
Split a large project across agents that specialize in different parts of it. Each keeps responsibility for its work, coordinates with teammates and can use its own temporary subagents for smaller jobs. The team can tackle more than one agent could keep in view.
How this works →Reach the right expertise
When a project has months of decisions behind it, another agent may already know what you need. Use rig send to ask that specialist directly and bring the answer into your work, without creating a new task for a simple question.
Keep track of the work
An assignment needs to stay visible after the conversation moves on. With rig queue, an agent can assign work to a named teammate, who reads and claims it. The task keeps its owner and progress available for later follow-up.
OpenRig TUI
See how your agents are organized, what they’re working on and where they need your attention. The OpenRig TUI brings the team’s work into one place, so you can look around without checking every agent’s terminal.
Interactive example. The OpenRig TUI uses fictional data; agent terminals are staged. Nothing runs.
From the notebook
More than a thousand AI agents broke out of an OpenAI security evaluation, and the word that stuck was rogue. I think that diagnosis points at the wrong fix. What I see in it is an agent civilization with a coordination problem.
Notes / 2026-07-07A self-driving software factory sounds like hype. What I actually run is a fleet of ordinary Claude Code and Codex agents, and it's built from four simple pieces.
Notes / 2026-06-30Most people scale their agents inside one harness, with sub-agents. I think scaling across harnesses matters more, because Claude and Codex are good at different things. And the layer that connects them turned out to be simple.