Stickynotes MCP

🚀 Sticky Notes - MCP

A Python-based MCP (Model Context Protocol) server project - Sticky Notes - demonstrates how to create custom tools, resources, and prompts for integration with AI assistants like the Claude desktop application.

🚀 Quick Start

Large Language Models (LLMs) such as Claude and GPT-4 excel at generating text but fall short when it comes to interacting with external data or performing real-world operations. MCP addresses this issue by defining a standard, modular way for ExternalNLP tools to communicate with applications. MCP follows the client-server model and uses state-based JSON-RPC messages, ensuring that only the necessary context is exchanged - thereby enhancing scalability, security, and flexibility.

This project is a starting implementation of an MCP server using the official Python SDK. It aims to serve as a proof of concept, showing how to create and expose simple functionality (in this case, managing sticky notes) for later connection to an MCP client (such as the Claude desktop).

✨ Features

About the Model Context Protocol (MCP)

MCP is an emerging open standard introduced by Anthropic, designed to simplify the way LLMs interact with external tools and data. Its main advantages include:

• Unified Integration: Standardizes the way applications call external services by reducing the need for custom integrations.
• Modularity and Scalability: Allows developers to build once and use across multiple applications - creating a large and well-interoperable ecosystem of components.
• State-based Sessions and Context Management: Supports multi-round conversations and maintains context across multiple interactions.
• Enhanced Security: Ensures that each tool receives only the necessary context information and protects sensitive data through sandboxing and capability negotiation.

With MCP, AI applications can dynamically expand their functionality, whether it's retrieving the latest customer data or performing specific operations (such as managing sticky notes).

Specific Features of the Project

• Tool: add_note(message: str) -> str
  Appends a new note to the local file (notes.txt) and returns a confirmation message.
• Tool: read_notes() -> str
  Reads all notes from the sticky notes file and returns them as a string. If no notes exist, it returns a default message.
• Resource: get_latest_note() -> str
  Retrieves and returns the latest note in the file.
• Prompt: note_summary_prompt() -> str
  Generates a prompt for the AI to summarize all current notes.

🤝 Contribution

Contributions are welcome! Feel free to fork this repository and submit pull requests for improvements. Please ensure that your code follows the existing patterns for MCP integration.