# paperclip

> A Node.js server and React dashboard for running a team of AI agents like a company

- Repository: https://github.com/paperclipai/paperclip
- Category: AI agents
- Stars: 106,218 · Forks: 17,665 · Language: TypeScript · License: MIT
- Experience: Some experience
- Page: https://ir.adyog.com/repos/paperclipai/paperclip/

If you run many AI agents at once, say a pile of Claude Code terminals, it is hard to know who is doing what, what it costs, and what happens after a reboot. Paperclip puts them behind one task manager. You describe a goal, set up the agents as an org chart, and watch the work from a dashboard.

It is a Node.js server with a React interface. Each agent is added through an adapter, and the README says anything that can receive a heartbeat can be hired. Agents wake for assigned tasks, follow-up messages or schedules, and goal context passes down from the company to the project to the task.

The README contrasts it with running agents by hand or with folders of scripts and configs. Paperclip adds ticketing, delegation, approval stages, budgets with automatic pauses, and run history. It names OpenClaw, Claude Code, Codex, Cursor, Gemini CLI and OpenCode as supported agents.

## What it does

- **Bring your own agents**: Adapters cover Claude Code, Codex, Cursor, Gemini CLI, OpenCode and others. Custom processes and HTTP endpoints can be added too.
- **Org chart and governance**: Agents get roles, reporting lines and permissions. You can set review and approval stages, approve hires, and pause or reassign work.
- **Budgets and cost tracking**: Set budgets per company, agent or project. It tracks reported spend, sends threshold alerts and pauses work at the limits you configure.
- **Tasks, routines and heartbeats**: Task threads keep conversation, files and run history attached to the work. Routines create recurring tasks on a schedule, or from an API or webhook.
- **Skills and connections**: Share skills across the org, test them with saved inputs and restore older versions. Connect services like GitHub and Notion or your own MCP server, with per-action permissions.

## What you could build

- Coordinate several coding agents on one project from a single dashboard instead of many terminals.
- Set spending limits so a runaway agent loop gets paused before it burns your quota.
- Schedule recurring jobs such as reports or support triage and inspect each run's output.
- Run separate organizations with their own agents, permissions and histories on one deployment.
- Install a ready-made team with roles, skills and routines, then change its runtimes to suit you.

## Who it's for

Suits developers and small teams already running several AI coding agents who want oversight; you should be comfortable with Node.js and know how your agents are set up.

## Get started

The README excerpt points to a Quickstart section but the install commands were cut off, so check the Quickstart on GitHub or at docs.paperclip.ing for the exact steps.



Tags: agent-orchestration, multi-agent, dashboard, cost-control, typescript