NoteKeeper MCP

An MCP (Model Context Protocol) server, built with FastMCP, that gives an AI assistant tools to capture, search, tag, and manage notes. Notes are stored in a local SQLite database.

Built to explore the Model Context Protocol: how to expose real, stateful functionality to an LLM through a clean, testable, deployable server.

Features

• Four MCP tools for full note management (create, list, search, delete)
• Persistent storage in SQLite — notes survive across sessions
• Tag support — organize notes with comma-separated tags and search by them
• Safe database access using parameterized queries (SQL-injection safe)
• Fully tested with an isolated pytest suite
• Installable package with a clean src/ layout

Tools

Tool	Description
add_note	Save a new note with optional tags
list_notes	List all notes, newest first
search_notes	Find notes by keyword in content or tags
delete_note	Delete a note by its id

Installation

Requires Python 3.10+.

# Clone the repository
git clone https://github.com/VRurs1606/notekeeper-mcp.git
cd notekeeper-mcp

# Create and activate a virtual environment
python -m venv .venv
# Windows:
.venv\Scripts\activate
# macOS/Linux:
source .venv/bin/activate

# Install the package and its dependencies
pip install -e .

Usage

Run the tests

pytest

Try the demo

A demo client that exercises every tool in-memory:

python demo.py

Connecting to Claude Desktop

To use NoteKeeper with Claude Desktop, add it to your MCP config file:

Windows: %APPDATA%\Claude\claude_desktop_config.json macOS: ~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "notekeeper": {
      "command": "python",
      "args": ["/absolute/path/to/notekeeper-mcp/src/notekeeper_mcp/server.py"]
    }
  }
}

Restart Claude Desktop, and you can ask it to save and search notes in natural language.

Project Structure

notekeeper-mcp/ ├── src/notekeeper_mcp/ │ ├── init.py │ ├── server.py # MCP server and tool definitions │ └── database.py # SQLite storage layer ├── tests/ │ ├── conftest.py # Shared test fixtures │ └── test_database.py # Storage layer tests ├── demo.py # In-memory demo client ├── pyproject.toml # Package configuration └── README.md

Tech Stack

• Python 3.10+
• FastMCP — MCP server framework
• SQLite — persistent storage (Python standard library)
• pytest — testing