Screenpipe
Source-available AI screen recorder that captures your entire workday locally on Mac and Windows - search every app, meeting, and website in plain English.
Screenpipe is a source-available, privacy-first screen and audio capture tool that records everything on your Mac or Windows PC locally - every app, browser tab, terminal, and meeting - and makes it all searchable in plain English through local AI models. Founded in 2024 and backed by Y Combinator, Screenpipe has accumulated 21,000+ GitHub stars from developers who want a self-hosted alternative to cloud-based AI memory tools like Rewind and Limitless. The codebase is published on GitHub and publicly auditable, but it ships under the Screenpipe Commercial License rather than an OSI-approved open source license: personal, non-commercial, and evaluation use can be free, while commercial use requires a paid license. The platform uses OCR to extract text from every captured frame, enables natural language search of the full history, and exposes a Pipe plugin system so developers can build custom automations on top of the screen timeline using TypeScript. A native MCP server lets Claude Desktop, Cursor, and other MCP-compatible AI tools read current screen context for more accurate assistance without manual copy-pasting.
Key Features
- Continuous screen and audio recording runs locally on Mac and Windows with no cloud uploads - all captured data stays on the local machine by default
- OCR pipeline extracts searchable text from every screen frame, making all on-screen content - including terminal output and PDFs - findable by keyword
- Local AI processing with Ollama integration keeps all data on-device - screen content is not sent to external servers without explicit per-session consent
- Pipe plugin system lets developers build custom automations on top of the screen capture timeline using a TypeScript SDK with access to OCR and audio data
- Native MCP server exposes screen memory to Claude Desktop, Cursor, and other MCP-compatible AI tools so they can read current screen context for better assistance
- Natural language search queries the full history of everything seen, typed, or heard across all apps with results in under 200ms using local embeddings
- Timeline view shows a scrollable visual history of all screen activity organized by app, website, and time period for retrospective review of any workday
Use Cases
- Developers using the Screenpipe MCP server to give Claude or Cursor live screen context for more accurate code suggestions without manual copy-pasting
- Remote workers building a searchable archive of all meetings, documentation, and browser research without cloud storage or manual note-taking during calls
- Researchers reviewing every web page, paper, and note visited during a research session using the natural language search interface the next day
- Engineers building custom pipe automations that trigger actions - Slack messages, calendar events, or code snippets - based on patterns in screen capture output
Pros
- 21,000+ GitHub stars and a publicly readable codebase mean the local-only privacy claims can be checked against the source rather than taken on trust
- MCP server integration is the only screen memory tool that natively feeds context to Claude, Cursor, and other AI coding tools without manual copy-pasting
- Free plan captures screen, audio, and meetings on one device and shares context with Claude, Codex, and other AI tools, with no payment method required to start
Cons
- Source-available under the Screenpipe Commercial License, not OSI open source - the code can be read and audited, but commercial use requires a paid license and the usual rights to fork or redistribute do not apply
- Continuous screen recording generates large local storage requirements - a full 8-hour workday can consume multiple gigabytes of capture data per day
- Running local AI models for search and summarization requires a machine with enough RAM to run Ollama models alongside other active applications
- Windows support is newer and has fewer tested Pipe plugins than Mac - some automations documented in the community are Mac-only at this stage
Screenpipe Alternatives
Explore similar tools and alternatives
Looking for alternatives to Screenpipe? Here are some similar tools you might like:
Rewind AI
AI-powered personal memory tool for Mac that records your screen, mic, and camera to make everything searchable via natural language chat.
Limitless AI
AI memory device and desktop app that captures, transcribes, and recalls conversations using a wearable Pendant and always-on computer recording.
Mem
AI-powered note-taking app that automatically organizes your notes by surfacing connections to people, projects, and past ideas - no folders required.
Ready to try Screenpipe?
Visit the official website to explore all features and get started with Screenpipe today.
Reviews
0 reviews for Screenpipe
Based on 0 reviews
Share your experience
Log in to write a review for Screenpipe
Notion AI
AI writing and organization assistant built into Notion for summarizing, drafting, brainstorming, and editing within your workspace.
Otter.ai
AI meeting transcription and note-taking tool that records, transcribes, and summarizes conversations in real-time.
Zapier
Automation platform connecting 7,000+ apps with AI-powered workflow creation from plain English descriptions.
Have an AI Tool?
List your AI tool for free, or go featured for top placement in your category - and reach thousands of potential users.
Submit Your Tool