Bear Notes RAG Server

Example Use Cases

• 1

  Semantically search a user's personal Bear Notes (meaning-based retrieval rather than simple keyword matching).

• 2

  Context for responses via retrieval-augmented generation, returning relevant note content formatted for RAG.

• 3

  Fetch specific notes or list tags from a local Bear Notes database while keeping all processing and data local.

Description

This project indexes Bear Notes into vector embeddings and exposes an MCP-compatible server that lets AI assistants perform semantic search and RAG on your personal notes. It uses a local transformer model (all-MiniLM-L6-v2 via transformers.js/Xenova) so processing and data remain on your machine. The server offers tools to search notes, fetch individual notes, list tags, and retrieve context-formatted results for RAG workflows. It's read-only, macOS-specific, and intended for private, local use.