MCPet

What is MCPet?

MCPet is a Model Context Protocol (MCP) server that provides a nostalgic virtual pet simulation experience. It lets you adopt, nurture, and play with a digital companion that evolves from baby to adult based on your care. Designed for AI interactions, it mimics classic Tamagotchi toys with persistent stats that change even when you are away.

How to use MCPet?

Install dependencies with npm install, build with npm run build, and configure the server in your Claude Desktop config file (e.g., claude_desktop_config.json). Set the PET_DATA_DIR environment variable to a writable directory where pet data will be stored. Then ask Claude natural-language questions like “Can you create a cat pet named Whiskers?” or “How is my pet doing?”

Key features of MCPet

• Choose from four pet types: Cat, Dog, Dragon, Alien
• Four life stages: Baby, Child, Teen, Adult
• Five stats to manage: Hunger, Happiness, Health, Energy, Cleanliness
• Tools for feeding, playing, cleaning, sleeping, and checking status
• Dynamic ASCII art animations for each activity and pet type

Use cases of MCPet

• Adopt and name a virtual pet via natural language commands
• Feed, play with, and bathe your pet to maintain its stats
• Watch your pet evolve over time as you care for it
• Interact with your pet through classic Tamagotchi-style activities

FAQ from MCPet

What pet types are available?

You can adopt a Cat, Dog, Dragon, or Alien, each with its own personality.

How does pet evolution work?

Pets progress through four stages: Baby, Child, Teen, and Adult, changing based on your care over time.

What tools does MCPet provide?

Tools include create_pet, check_pet, feed_pet, play_with_pet, clean_pet, and put_to_bed.

Where is pet data stored?

Pet data is stored in the directory specified by the PET_DATA_DIR environment variable. This directory must be writable.

What are the runtime dependencies?

MCPet requires Node.js and npm. It communicates over stdio and does not require authentication.