Bear Notes

Connect your Bear app via MCP (Model Context Protocol) allowing AI assistants to search and retrieve your notes

javascript

GitHub Stars

javascript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "ruanodendaal-bear-mcp-server": {
      "command": "node",
      "args": [
        "/absolute/path/to/bear-mcp-server/src/bear-mcp-server.js"
      ],
      "env": {
        "BEAR_DATABASE_PATH": "/Users/yourusername/Library/Group Containers/9K33E3U3T4.net.shinyfrog.net.bear/Application Data/database.sqlite"
      }
    }
  }
}

Bear Notes MCP Server with RAG connects your Bear Notes data to AI assistants through semantic search and Retrieval-Augmented Generation, enabling contextual, meaning-based answers while keeping everything on your machine.

How to use

You run the local MCP server and connect it to an MCP client. Your AI assistant can search your Bear Notes semantically, fetch relevant notes, and use them as context to generate informed responses. Start by indexing your notes, then run the server so your assistant can query and retrieve note content efficiently.

How to install

Prerequisites: Node.js version 16 or higher and Bear Notes for macOS. You also need an MCP-compatible AI assistant client.

Step 1: Install dependencies locally.

Step 2: Build and index your notes (this creates the semantic vector index). You will run the indexing script after installing dependencies.

Step 3: Configure the MCP server entry point and environment variable for Bear Notes storage.

Configuration and usage notes

This server runs entirely locally. It uses semantic search to match queries to your notes by meaning, and it can pull in relevant context for AI assistants via RAG. If semantic search is unavailable for any reason, it will fall back to traditional search without failing.

Troubleshooting

If you encounter issues, verify the Bear Notes database path is correct, ensure indexing has completed, check file permissions on the database, confirm the MCP server script is executable, and inspect logs for error messages.

Docker (optional)

You can run the server in Docker. First build the image, then index your notes, and finally start the container with the appropriate volume mounts and environment variables.

Available tools

search_notes

Find notes that match a query using semantic relevance, with optional limit and semantic toggles.

get_note

Fetch a specific note by its ID.

get_tags

List all tags used in Bear Notes.

retrieve_for_rag

Get notes semantically similar to a query, formatted for RAG integrations.

Built by

VeilStrat

AI signals for GTM teams