What is Tik Tak Toe Mcp

tik-tak-toe-mcp is an MCP extension that provides an interactive user interface for playing the classic game Tic-Tac-Toe. It serves as a showcase for building MCP servers with UI components.

Use cases

Use cases include creating interactive learning tools, developing AI-driven game applications, and enhancing user interfaces in software that requires simple game mechanics.

How to use

To use tik-tak-toe-mcp, you can install it via PyPI using ‘uvx install tik-tak-toe-mcp’ or clone the repository and install it from source. Once installed, you can run the server or integrate it with Goose for a more interactive experience.

Key features

Key features include an interactive UI for Tic-Tac-Toe, separation of AI interaction from the user interface, tools for game management, and comprehensive game rules. It also serves as a template for creating custom MCP extensions.

Where to use

tik-tak-toe-mcp can be used in educational settings for teaching programming concepts, in game development for prototyping, and in interactive applications where simple games can enhance user engagement.

Tic-Tac-Toe MCP Extension

A showcase example of building Model Context Protocol (MCP) servers with interactive UIs. This project demonstrates how to create an MCP extension that launches a separate UI window for playing Tic-Tac-Toe.

Demo

Watch a demo of the Tic-Tac-Toe MCP Extension in action:

https://github.com/user-attachments/assets/a6ae4ba7-7a9c-42ed-bccc-56e6c988c266

Direct link to video

Features

• Demonstrates MCP server implementation with interactive UI components
• Showcases separation of AI interaction and user interface
• Play Tic-Tac-Toe through Goose AI with a separate visual interface
• Provides tools to start a game, make moves, and check the board state
• Includes game rules as a resource
• Serves as a template for building your own MCP extensions with UIs

Installation

Option 1: Install from PyPI

Install directly from PyPI using uvx (recommended):

uvx install tik-tak-toe-mcp

Or using pip:

pip install tik-tak-toe-mcp

Option 2: Install from source

1. Clone this repository:

   git clone https://github.com/Kvadratni/tik-tak-toe-mcp.git
   cd tik-tak-toe-mcp
   

2. Install the package:

   pip install -e .
   

Usage

As a standalone MCP server

Run the server:

tik-tak-toe-mcp

With Goose

Quick Install (One-Click)

Click the link below if you have Goose installed:

goose://extension?cmd=uvx&arg=tik-tak-toe-mcp&id=tik_tak_toe&name=Tic-Tac-Toe%20Game&description=Play%20Tic-Tac-Toe%20with%20a%20separate%20UI

Option 1: Using Goose CLI (recommended)

Start Goose with your extension enabled:

# If you installed via PyPI
goose session --with-extension "uvx tik-tak-toe-mcp"

# Or if you want to use a local development version
goose session --with-extension "python -m tik_tak_toe_mcp"

Option 2: Manual setup in Goose

1. Run goose configure
2. Select “Add Extension” from the menu
3. Choose “Command-line Extension”
4. Enter a name (e.g., “Tic-Tac-Toe Game”)
5. For the command, enter: uvx tik-tak-toe-mcp
6. Follow the prompts to complete the setup

Tools

• start_game(): Start a new Tic-Tac-Toe game and launch the UI
• make_move(row, col): Make a move at the specified position (row and col are 0-2)
• get_board_state(): Get the current state of the board

Resources

• game_rules: The rules of Tic-Tac-Toe and how to use this extension

MCP Architecture Overview

This project showcases the Model Context Protocol (MCP) architecture for building AI-powered applications with separate UIs:

Key Components

1. MCP Server: Handles communication between the AI model and the UI

   • Defines tools that the AI can use to interact with the game
   • Manages game state and logic
   • Provides a protocol for the UI to connect and receive updates

2. Interactive UI: A separate process that provides visual interaction

   • Renders the game board
   • Captures user input
   • Updates based on AI actions

3. Integration with Goose AI:

   • The AI uses the provided tools to interact with the game
   • Tool calls are processed by the MCP server
   • Results are reflected in both the AI conversation and the UI

Benefits Demonstrated

• Separation of Concerns: UI logic is separate from AI interaction
• Enhanced User Experience: Visual interface alongside AI conversation
• Extensibility: Pattern can be applied to more complex applications
• Reusability: Same MCP server can work with different AI models

This architecture can be adapted for various applications requiring both AI interaction and rich user interfaces.