mcp-bear

MCP.Pizza Chef: jkawamoto

mcp-bear is an MCP server designed to integrate Bear note-taking software with LLMs, allowing real-time interaction and data exchange. It exposes Bear's notes and functionalities as structured data accessible to models, facilitating advanced workflows involving note retrieval, creation, and management. Currently under development, it has a known issue where a browser opens for every API call, but it aims to streamline Bear note integration in AI-enhanced environments.

Archived · The author made this repository read-only on GitHub.

Use This MCP server To

Access Bear notes for real-time content retrieval Create and update Bear notes via LLM commands Search Bear notes using natural language queries Integrate Bear note data into AI workflows Automate note organization and tagging in Bear

README

Important

This application is currently under development. A known issue exists where a browser opens for every API call.

Bear MCP Server

A MCP server for interacting with Bear note-taking software.

Installation

For Goose CLI

To enable the Bear extension in Goose CLI, edit the configuration file ~/.config/goose/config.yaml to include the following entry:

extensions:
  bear:
    name: Bear
    cmd: uvx
    args: [--from, git+https://github.com/jkawamoto/mcp-bear, mcp-bear]
    envs: { "BEAR_API_TOKEN": "<YOUR_TOKEN>" }
    enabled: true
    type: stdio

mcp-bear FAQ

How do I install the mcp-bear server?

You can install mcp-bear by cloning its GitHub repository and following the setup instructions in the README, which include Python environment setup and dependencies installation.

What platforms does mcp-bear support?

mcp-bear primarily supports macOS, as Bear note-taking software is macOS/iOS exclusive.

Are there any known issues with mcp-bear?

Yes, currently a browser window opens for every API call, which is a known issue under active development.

How does mcp-bear handle authentication with Bear?

mcp-bear uses Bear's local APIs or scripting interfaces for authentication, typically requiring user permission on the host machine.

Can mcp-bear be used with multiple LLM providers?

Yes, mcp-bear is provider-agnostic and can be integrated with OpenAI, Anthropic Claude, and Google Gemini models.

Is mcp-bear suitable for production use?

As it is currently under development with known issues, it is recommended for experimental or development use rather than production.

How can I contribute to mcp-bear development?

Contributions are welcome via GitHub pull requests; check the repository for contribution guidelines and open issues.