openrig vs paperclip
Define a team of Claude Code, Codex and Pi agents in YAML and boot it with one command: persistent tmux-backed seats with roles, shared context and owned work, steered via a lead agent. — versus — Open-source control plane for running fleets of heterogeneous AI agents as a "company" — bring your own agent, assign goals, org charts, budgets, governance, and an audited ticket system.
Both model a fleet of heterogeneous agents as a company with roles and assigned work; Paperclip is a control plane, OpenRig a CLI/TUI rig.
| openrig | paperclip | |
|---|---|---|
| Stars | 8.7k | 111k |
| Forks | 722 | 18k |
| Language | TypeScript | TypeScript |
| License | Apache-2.0 | MIT |
| Last activity | yesterday | yesterday |
| Topics | orchestration, coding, agents | agents, orchestration |
| Curated connections | 6 | 11 |
openrig — the curator's take
Use it when you already live in Claude Code/Codex and want a standing team — lead agent, specialists, stable addresses — rather than one-off sessions. Skip it if you want a hosted platform or Windows-native setup (tmux, macOS/Linux/WSL2 only), and read the permissions section first: the kernel operator runs with broad, prompt-free access by default.
paperclip — the curator's take
Reach for Paperclip when you run many agents across providers 24/7 and need a boss layer — budgets that hard-stop, org charts, goal alignment, audit trail, mobile monitoring. Sweet spot: "20 Claude Code tabs, lost track of who does what." NOT the tool to build an agent (it orchestrates ones you already have), and overkill for a single agent or one linear pipeline. It's a control plane, not a framework.