📝 AI Sticky Notes – MCP Project

This project is a simple and functional application that uses the MCP (Model Context Protocol) to create a sticky notes system powered by AI. It leverages FastMCP, the mcp[cli] interface, and integrates with Claude Desktop.

🚀 Features

• Add new notes
• Read all existing notes
• Remove all notes
• Retrieve the latest note
• Generate a prompt to summarize all notes

🧰 Technologies Used

• Python
• FastMCP
• uv for environment management
• Claude Desktop for execution with tools, resources, and prompts

🛠️ Setup and Usage

1. Create and initialize the MCP environment

uv install
uv init project-folder-name
cd project-folder-name
uv add "mcp[cli]"

2. Install and use Claude Desktop

• Download and install Claude Desktop
• Run and interact with the defined tools, resources, and prompts

📁 File Overview

main.py

This file contains all the tools and logic to manage notes using the MCP framework.

🔧 Tools

add_note(message: str) -> str

Appends a new note to the notes.txt file.

read_notes() -> str

Returns all saved notes. If no notes exist, returns a default message.

remove_notes() -> str

Clears all notes in the notes.txt file.

🔗 Resource

get_latest_note() -> str

Returns the most recently added note.

💬 Prompt

note_summary_prompt() -> str

Generates a prompt asking the AI to summarize all current notes.

📄 Notes File

All notes are stored in a local notes.txt file, which is automatically created if it doesn't exist.

✅ Requirements

• Python 3.10+
• uv installed
• Claude Desktop installed

🧠 Inspiration

This project demonstrates how to use MCP to build intelligent agents that interact with local files and structured data. It’s easily extendable with new functionalities.