Skip to main content
Run Paperclip locally for development, testing, or single-operator workflows. Local deployment uses embedded PostgreSQL and requires no external dependencies.

Quick Start

The fastest way to get started:
Or clone and run manually:
The server starts at http://localhost:3100 with an embedded PostgreSQL database.

Prerequisites

  • Node.js 20+ - Required for running the server
  • pnpm 9.15+ - Package manager (use corepack enable to install)

Default Local Setup

By default, local deployment runs in local_trusted mode:
  • No login required - Optimized for fastest local startup
  • Localhost-only binding - Binds to 127.0.0.1 for security
  • Embedded database - PostgreSQL runs in-process
  • Local file storage - Attachments stored on disk
  • Company deletion enabled - Easy cleanup during development

Data Persistence

All data persists in ~/.paperclip/instances/default/:

Custom Data Directory

Override the default location:

Multiple Instances

Run multiple isolated instances:
Each instance gets its own directory under ~/.paperclip/instances/.

Development Commands

Health Checks

Verify your local installation:

Reset Local Database

To wipe data and start fresh:
The database will be recreated automatically on next startup.

Docker Local Deployment

Run Paperclip in Docker without installing Node locally:
Or use Docker Compose:
Data persists in the mounted volume (./data/docker-paperclip).

With AI Agent Support

To use local Claude or Codex adapters, pass API keys:

Private Network Access (Tailscale)

For private network access while keeping authentication optional:
This runs in authenticated/private mode and binds to 0.0.0.0.

Allow Custom Hostnames

Add private hostnames (e.g., Tailscale machine names):

Troubleshooting

Port Already in Use

Change the default port:

Database Migration Issues

Run migrations manually:

Check Logs

Logs are written to .paperclip/logs/server.log:

Next Steps

Configuration

Configure deployment mode, storage, and secrets

Database Setup

Switch to external PostgreSQL for production