What Paperclip is
Paperclip is an open-source application for running AI agents as part of a team's daily work. Instead of juggling separate terminal sessions, chat threads, and dashboards, it gives agents a shared workspace with tasks, logs, and handoffs visible in one place.
Why it is trending
Agent tooling has moved past single-shot prompts. Teams now run long-lived agents that plan, execute, and report back. Paperclip sits in that gap: it treats agents like coworkers with assignments, status, and history, rather than anonymous API calls.
Core ideas
- Tasks as first-class objects. Work is broken into tracked items an agent can pick up, update, and close.
- Shared context. Agents read the same project state instead of starting from a blank prompt each run.
- Observability. Runs, tool calls, and outputs are recorded so a human can audit what happened.
- Handoffs. An agent can pass work to another agent or back to a person with the context attached.
When to use it
Paperclip fits teams that already run coding agents (Claude Code, Cursor, Codex) and want a single place to assign work, review results, and keep an audit trail. For solo experiments, a plain terminal is lighter.
Getting started
Clone the repository, follow the setup in the README, and connect your preferred agent runtime. Start with one agent and one project before scaling to a fleet.
Repository: github.com/paperclipai/paperclip
