hydrusnetwork/hydrus

Hydrus is a personal media tagger that organizes and manages image and video files with a booru-style tagging system. It runs as a desktop application and lets you build a searchable database of media from your local drives and popular websites, then optionally share collections with others through peer-to-peer servers.

The tool solves the problem of managing large personal media libraries without relying on centralized cloud services. It works by importing files into a local database where you assign tags, ratings, and metadata to each item. The tagging system supports hierarchical tag relationships, allowing you to organize content by multiple dimensions simultaneously. Search queries combine tags with boolean logic to find exactly what you need. The application handles duplicate detection and can fetch tags automatically from external sources when importing.

Hydrus suits anyone building a personal archive who wants full control over their data and doesn't want to depend on third-party services. It works well for managing collections too large or sensitive for cloud storage, or for users who prefer self-hosted solutions. The peer-to-peer sharing model means you can exchange curated collections with trusted users without uploading to a central server. This approach is fundamentally different from cloud-based alternatives that store your media on company servers.

Development on the project shows consistent, focused activity with regular updates addressing both core functionality and user-reported issues. The maintainer actively incorporates user feedback and implements feature requests that align with the tool's scope. Bug fixes and improvements arrive in a steady cadence rather than sporadic bursts. The codebase receives ongoing maintenance to handle edge cases and improve stability across different operating systems and file types.