AI Sticky Notes MCP Server

What is AI Sticky Notes MCP Server?

AI Sticky Notes MCP Server is a lightweight note-taking system built with the Model Context Protocol (MCP) that allows AI assistants to create, read, and summarize notes. It uses FastMCP for easy integration with AI assistants like Claude Desktop or Cursor IDE.

How to use AI Sticky Notes MCP Server?

Clone the repository, install dependencies with uv pip install -e ., and run the server with uv run mcp dev main.py for development. To integrate with Claude Desktop, run uv run mcp install main.py and configure the claude_desktop_config.json. For Cursor IDE, add a new MCP server in settings with stdio transport and the command uv run --with mcp[cli] mcp run /full/path/to/your/main.py.

Key features of AI Sticky Notes MCP Server

• add_note tool to append new notes
• read_notes tool to return all stored notes
• notes://latest resource to retrieve the most recent note
• note_summary_prompt to generate AI summaries of notes
• Built with FastMCP for easy AI assistant integration

Use cases of AI Sticky Notes MCP Server

• Adding reminders via natural language (e.g., "Add a reminder to buy groceries")
• Reading all existing notes to review tasks or ideas
• Retrieving the latest note quickly
• Generating a summary of stored notes through AI

FAQ from AI Sticky Notes MCP Server

What are the prerequisites for AI Sticky Notes MCP Server?

Python 3.8+, the uv package manager, and optionally Claude Desktop or Cursor IDE.

How do I install dependencies for AI Sticky Notes MCP Server?

Run uv pip install -e . in the cloned repository directory.

Where are notes stored?

Notes are stored in a notes.txt file in the project directory. The path can be customized by modifying NOTES_FILE in main.py.

Why is the notes://latest resource not available in Cursor?

Cursor only supports MCP tools, not resources. Therefore the notes://latest resource will not work in Cursor; only tools add_note and read_notes are available there.

How can I debug AI Sticky Notes MCP Server?

Use uv run mcp dev main.py to run the server in development mode, which provides helpful debugging information for stdio-based MCP servers.