Show HN: MetaBrain – A local document memory for AI agents
9 hours ago
- #document management
- #AI tools
- #local storage
- metaBrain is a local-first, open-source tool for storing and managing document memory for AI agents and human users.
- It provides a durable, searchable store for notes, source snippets, task context, metadata, tags, links, and version history.
- The tool includes a CLI (`mb`), an optional daemon (`mbd`), and an embeddable Swift library (`MetaBrainCore`).
- Designed for AI agents, it offers stable local storage with predictable paths, search capabilities, and retained document versions.
- Key features include storing and retrieving data by words, paths, tags, or metadata, and applying unified diffs to documents.
- The default store is located at `.metabrain/store.leveldb` for easy discovery without setup.
- It supports macOS, Linux, and Windows, and is BSD 3-Clause licensed, available via Homebrew installation.