screenpipe/screenpipe

What it solves

Screenpipe provides a local, 24/7 recording of your computer's screen and audio, transforming your digital activity into a searchable, AI-powered memory. It eliminates the need to manually take notes or remember exactly where a piece of information was seen or heard, while keeping all data strictly local and private.

How it works

The system uses event-driven capture to record screenshots and the OS accessibility tree (falling back to OCR when necessary) only when meaningful changes occur, such as app switches or typing. It captures system and microphone audio, transcribing it locally using Whisper. All data is stored in a local SQLite database with full-text search capabilities. This data is then exposed via a REST API and a Model Context Protocol (MCP) server, allowing AI agents and assistants (like Claude or Cursor) to query your history.

Who it’s for

• Knowledge workers and researchers who need to recall specific details from their digital workflow.
• Developers who want to provide AI coding assistants with real-time context of their current work.
• People with ADHD who struggle to keep track of tabs, documents, and conversations.
• Remote workers who require automatic meeting transcriptions and summaries.
• Enterprises looking to deploy AI memory tools with deterministic, OS-level data permissions.

Highlights

• 100% Local-First: Data is stored on the user's device, with optional local AI (via Ollama) for maximum privacy.
• AI Pipes: Scheduled AI agents defined as markdown files that can automate tasks like daily recaps or meeting summaries based on screen data.
• MCP Server: Native integration with AI assistants like Claude Desktop and Cursor for direct history querying.
• Event-Driven Capture: Reduces CPU and storage overhead by capturing only when activity changes rather than recording every second.
• Multi-Monitor Support: Captures all connected monitors simultaneously.