Bear MCP Server is an MCP server designed to connect AI models with your Bear Notes application. It empowers AI to access and utilize the knowledge stored within your notes, enhancing their ability to provide contextually relevant and informed responses. This server allows AI models to read all your notes, search for specific information within them, and list all available tags, effectively turning your Bear Notes into a dynamic knowledge base for AI.

By running SQL commands on Bear's SQLite database, the server seamlessly integrates with your existing notes. To use it, you'll need Node.js and access to your Bear database. Once configured, AI models can leverage tools like get_notes, get_tags, and get_notes_like to interact with your notes. The core value lies in unlocking the potential of your personal knowledge base, enabling AI to provide more personalized and insightful assistance. You can easily integrate it by configuring your claude_desktop_config.json file.

Note Retrieval via SQL

Bear-mcp-server leverages direct SQL queries on the Bear Notes SQLite database to retrieve note content. This core functionality allows AI models to access the complete text of all notes stored within Bear. The server executes pre-defined SQL commands, abstracting away the complexities of database interaction from the AI model. This ensures that the AI can reliably and efficiently access the information stored in Bear Notes. For example, an AI assistant can use this feature to summarize all meeting notes stored in Bear, providing a quick overview of past discussions. The technical implementation involves establishing a connection to the SQLite database and executing SELECT queries to extract the note content.

Text-Based Note Search

This feature enables AI models to search for specific notes within Bear based on textual content. The bear-mcp-server implements this by executing SQL LIKE queries against the Bear Notes database. This allows for flexible searching, including partial matches and wildcard characters. The AI model can specify the search term, and the server returns a list of notes that contain the specified text. This is particularly useful for AI applications that need to find relevant information quickly, such as a chatbot answering questions based on information stored in Bear Notes. For instance, a user could ask the AI "What did I learn about MCP?", and the server would return all Bear Notes containing the word "MCP".

Tag Listing

Bear-mcp-server provides the ability to list all tags used within Bear Notes. This functionality allows AI models to understand the organization and categorization of notes within the user's Bear environment. By retrieving a list of all tags, the AI can then use this information to filter and prioritize notes based on their relevance to a specific task or query. For example, an AI model could use the list of tags to identify all notes related to a specific project or client. This feature is implemented by querying the Bear Notes database for all unique tags. The AI can then use these tags to refine its search and retrieval of notes, providing more accurate and relevant results.

Technical Implementation

Bear-mcp-server is built using Node.js and TypeScript, providing a modern and efficient platform for interacting with the Bear Notes database. The server utilizes the sqlite3 Node.js package to establish a connection to the SQLite database and execute SQL queries. The server exposes its functionality through a set of MCP tools (get_notes, get_tags, get_notes_like), which can be accessed by AI models via the MCP protocol. The use of TypeScript ensures type safety and code maintainability, while Node.js provides a lightweight and scalable runtime environment. The server is designed to be easily installed and configured, requiring only Node.js and access to the Bear Notes database.