screenpipe

https://github.com/user-attachments/assets/70fe94eb-6d2a-47ca-b7c3-c8ead13a5b7f

---

what is this?

screenpipe capture all your computer work locally and power your company's agents

┌─────────────────────────────────────────┐
│  screen + audio → local storage → ai   │
└─────────────────────────────────────────┘

• remember everything - never forget what you did, saw, heard
• map workflows - generate a list of workflows to automate (instead of interviewing people or hiring consultants)
• run agents that work based on what you do generate agents, skills, and automations based on what you do

• search with ai - find anything using natural language
• Local-first - capture history stays on your device by default. Cloud AI, sync, integrations, and managed team storage have separate data paths; see Privacy and security.
• source-available - inspect, modify, audit (LICENSE.md)

install

Download the desktop app for macOS, Windows, or Linux. Available features depend on your plan.

or run the CLI:

npx screenpipe record

then

npx screenpipe setup
# or
claude mcp add screenpipe -- npx -y screenpipe-mcp@latest

then ask claude what did i see in the last 5 mins? or summarize today conversations or create a pipe that updates linear every time i work on task X

If Claude Code, Codex, Gemini CLI, Cursor, or another coding agent is working from this repository, give it this instruction:

Read the screenpipe CLI skill before operating screenpipe. Set up always-on local capture, verify capture freshness and storage, then query my history without relying on the desktop app.

To install the screenpipe skills and MCP configuration into every supported agent detected on your computer, run:

npx screenpipe setup

The skill covers the recorder-first service default, explicit API-only server mode, human and JSON status, local search, safe read-only SQLite access, pipes, and connections.

specs

• captures full accessibility tree, OCR as fallback, transcription, speakers, keyboard inputs, app switches
• CPU: approximately 5-20%; varies with hardware, capture settings, transcription, and AI workloads
• RAM: approximately 0.5-3 GB for capture; local models and additional workloads can use more
• Storage: varies with activity, displays, audio, capture settings, and retention. Measure a representative day on your device before sizing storage.
• filters (window, app, chrome extensions, passwords, proprietary AI PII model)
• optional encryption at rest
• Local capture and search work offline after setup. Cloud AI, sync, and connected services require network access.

---

Repository guide

Start with CONTRIBUTING.md for development setup and contribution requirements. Agents should also read AGENTS.md; CLAUDE.md points to the same instructions.

Directory	Contents
apps/	Desktop app and the Workflows web host
crates/	Rust capture, storage, engine, and service crates
packages/	SDK, CLI distribution, MCP server, browser extension, and shared UI
docs/	Contributor guides, product/design guidance, testing, and architecture specs
evals/	Coding-agent regression evaluations
infra/	Build runners and deployment infrastructure
scripts/	Repository development and maintenance tools

Documentation for humans and agents

• Onboarding: walkthrough from setup to a first contribution.
• Vision: product priorities and scope.
• Design: interface principles and visual conventions.
• Testing: regression checklists and validation guidance.
• Coverage: generated E2E and core engine coverage summary.
• Native builds: supported build and test commands.
• App publication: release and publication boundaries.

The root package.json exposes @screenpipe/workflows-ui for projects that install this repository as a Git dependency. It forwards exports to packages/workflows-ui, where the implementation lives; it is not a root JavaScript workspace or app. In-repo apps depend on that package directly. Keep the two manifests' exports and dependencies aligned.

Contributing

See CONTRIBUTING.md for guidelines, maintainers, and how to submit PRs.

Thanks to all contributors:

Translations

• 日本語 (Japanese)
• 简体中文 (Simplified Chinese)
• Français (French)
• Español (Spanish)
• Português BR (Brazilian Portuguese)
• Deutsch (German)
• Українська (Ukrainian)
• 한국어 (Korean)
• Русский (Russian)

Want to translate screenpipe into another language? See docs/translations/.

About screenpipe

screenpipe is a source-available application that continuously captures your screen and audio, creating a searchable, AI-powered memory of everything you do on your computer. Capture history is stored locally by default; optional cloud and team features can send data off-device. It is the leading source-available alternative to Rewind.ai (now Limitless), Microsoft Recall, Granola, and Otter.ai. If you're looking for a rewind alternative, recall alternative, or a private local screen recorder with AI, screenpipe is the most popular option you can fully audit.

• Website: https://screenpi.pe
• Documentation: https://docs.screenpi.pe
• Discord: https://discord.gg/screenpipe
• License: Screenpipe Commercial License (source-available; personal, non-commercial use permitted, commercial use requires a license, see LICENSE.md)

Who screenpipe is for

• Knowledge workers who want to recall anything they've seen or heard on their computer
• Developers who want to give AI coding assistants (Cursor, Claude Code, Cline, Continue) context about what they're working on
• Researchers who need to search through large volumes of screen-based information
• People with ADHD who frequently lose track of tabs, documents, and conversations
• Remote workers who want automatic meeting transcription and notes
• Teams & enterprises who want to deploy AI across their organization with deterministic data permissions and central config management (screenpi.pe/team)
• Anyone who wants a private, local-first alternative to cloud-based AI memory tools

Platform support

Platform	Support	Installation
macOS (Apple Silicon)	✅ Full support	Native .dmg installer
macOS (Intel)	✅ Full support	Native .dmg installer
Windows 10/11	✅ Full support	Native .exe installer
Linux (x86_64)	Desktop app available	AppImage download; source builds also available

8 GB of system RAM is recommended. Resource estimates are listed in Specs. They are planning estimates, not fixed limits or a benchmark for every machine. Linux capture behavior depends on the distribution, desktop environment, and permissions; validate your intended setup before a rollout.

Core features

Event-driven screen capture

Instead of recording every second, screenpipe listens for meaningful events — app switches, clicks, typing pauses, scrolling — and captures a screenshot only when something actually changes. Each capture pairs a screenshot with the accessibility tree (the structured text the OS already knows about: buttons, labels, text fields). If accessibility data isn't available (e.g. remote desktops, games), it falls back to OCR. This gives you maximum data quality with minimal CPU and storage — no more processing thousands of identical frames.

Audio transcription

Captures system audio (what you hear) and microphone input (what you say). Real-time speech-to-text using Whisper (Large-V3-Turbo) running locally on your device, or Deepgram for cloud transcription. Speaker identification and diarization. Works with any audio source — Zoom, Google Meet, Teams, or any other application.

On macOS 14.4+, you can exclude specific apps from system-audio capture by listing their bundle IDs in ~/.screenpipe/audio-exclusions.json. Enable Experimental CoreAudio System Audio in Settings → Recording first; the picker UI only appears once that flag is on.