This project provides an open-source macOS application that locally records and indexes everything a user views on their computer, enabling keyword search and retrieval of past content. It targets users who want to recall information, capture context for LLMs, or maintain a searchable history of their digital interactions, with a strong emphasis on local-only processing for privacy and security.

How It Works

The application operates by taking screenshots every two seconds, utilizing macOS's Live Text feature to recognize and index text content. This data is stored locally in an SQLite database alongside the screenshots. Users can navigate through their recorded history via a timeline scrubber, search for specific keywords or applications, and copy text directly from past views. The core advantage is its fully local, privacy-focused design, avoiding any cloud synchronization or telemetry.

Quick Start & Requirements

• Download the latest release from the project's GitHub repository.
• Requires macOS. An Intel build is available, but it's primarily tested on Apple Silicon.
• Grant "Screen Recording" permissions to the application.

Highlighted Details

• Automatic screenshotting every 2 seconds with text recognition.
• Full-screen timeline scrubber for navigating recorded history.
• Keyword search functionality, filterable by application.
• "Copy Recent Context" feature for LLM prompting.

Maintenance & Community

The project is actively seeking contributions and help, with the developer expressing a need for assistance due to limited Swift experience. Issues and feature requests can be logged on GitHub.

Licensing & Compatibility

The project is open source, but the specific license is not explicitly stated in the README. Compatibility for commercial use or linking with closed-source projects would require clarification of the license.

Limitations & Caveats

The project is described as "very alpha" and is primarily tested on Apple Silicon, with an Intel build available but needing more testing. Natural language search and advanced search filters are listed as future features, indicating current limitations in search capabilities.