An open-source AI memory tool that records your screen and audio locally, enabling search and automation agents based on your computer activity.
Screenpipe is an open-source AI memory tool that turns your computer into a personal AI by continuously recording your screen and audio locally. It creates a searchable memory of everything you've seen, heard, or done, enabling you to recall past activities and run automated agents (pipes) based on your work patterns. The tool solves the problem of forgetting important information and manual context switching by providing AI-powered search and automation directly from your activity history.
Knowledge workers, developers, researchers, remote workers, people with ADHD, and teams who need to recall computer-based information or automate workflows based on screen activity. It's particularly valuable for developers using AI coding assistants who want to provide context about their work.
Developers choose Screenpipe because it's the leading open-source, privacy-focused alternative to commercial tools, offering 100% local data storage, deterministic AI data permissions, and a flexible plugin system for automation. Its event-driven capture minimizes resource usage while providing rich context for AI assistants through MCP integration.
Run agents that work for you based on what you do. AI finally knows what you are doing
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Captures screenshots only on meaningful OS events like app switches or clicks, reducing CPU usage to 5-10% and storage to ~20GB/month compared to continuous recording.
All data is stored locally by default with support for local AI models like Ollama, ensuring no data leaves the machine without user consent, as emphasized in the privacy section.
Runs as an MCP server for AI assistants like Claude Desktop and Cursor, and allows scheduled AI agents (pipes) defined as markdown files for automation based on screen activity.
Offers per-pipe AI data permissions via YAML frontmatter with OS-level enforcement, enabling centralized config management for organizations without exposing employee data.
The desktop app requires a one-time $400 purchase, which is steep compared to free or subscription-based alternatives and may barrier entry for individual users.
Creating pipes involves writing markdown files with YAML frontmatter for scheduling and permissions, which can be cumbersome for non-technical users despite the plugin flexibility.
Linux support requires building from source, unlike macOS and Windows with native installers, making setup less straightforward for some users.
Screenpipe is an open-source alternative to the following products:
Granola is a macOS app for managing window layouts and organizing desktop workspaces with keyboard shortcuts.
Limitless is an AI-powered productivity tool that provides meeting recording, transcription, and summarization features to help users capture and organize information.
Rewind.ai is a platform that captures and organizes everything you see, say, or hear on your computer to create a searchable memory of your work.