Screenpipe logo
backed byF.inc logo

let AI access your screen history

Cross-platform, open source, 100% local, dev-friendly,captures all desktop activities 24/7

AI App Store

Discover powerful plugins built by the community

don't see what you need? request a custom app from our developer community

Two Ways to Use It

A

Use screenpipe as a whitelabel solution (MIT)

Use our sdk/api or clone our repo github.com/mediar-ai/screenpipe

B

Deploy to screenpipe Store

Use screenpipe sdk/api in your nextjs app, deploy it to screenpipe store with thousands of users that are already buying other apps

feature overview

screenpipe architecture diagram

What Developers Say

Screenpipe saved me months of development time building desktop context for my AI agent

Sam Stowers

Weights & Biases

I could easily combine screenpipe library with my productivity app

Mike Bird

Open Interpreter

The API is simple, the deployment is in seconds

Kyle Morris

Founder of Banana.dev

How Apps Are Built

Frontend

  • • NextJS App with TypeScript
  • • Tailwind for Styling
  • • ShadcnUI Components
  • • Deployed Directly to Desktop

Backend

  • • Call screenpipe localhost:3030
  • • Store data on users device directly
  • • Call LLMs through local endpoints
  • • Create actions: clicks, typing

Get Started in Minutes

1. Install screenpipe library

$curl -fsSL get.screenpi.pe/cli | sh

2. Build screenpipe frontend locally

3. Create your app

$bunx --bun @screenpipe/dev@latest pipe create

4. Deploy it locally

$screenpipe pipe install <path or url>

watch the demo to see screenpipe in action

power AI by your desktop history today

or
book a demo

for businesses

or
book a call

for investors