This project implements an AI dungeon-style RPG game system for the Roo Code platform, including two MCP service modules: game state management and combat engine. It supports functions such as character creation, item management, combat dice rolling, and task tracking, and uses SQLite for persistent storage.

This project includes two MCP servers: OnePiece for querying anime character information and IP address geographical location query services. It is developed using TypeScript and can be integrated with AI clients through standard input and output.

VRChat MCP OSC is a bridge that connects an AI assistant with VRChat via the OSC protocol, enabling AI-driven virtual character control and interaction

Gamemaster MCP is an AI-assisted Dungeons & Dragons (D&D) game management server built on FastMCP, providing complete campaign management, character control, NPC interaction, map building, and quest tracking functions, supporting multiplayer collaboration and single-player immersive gaming experiences.

This project designs and implements an architecture that connects the data space with generative AI. Through the Model Context Protocol (MCP) server as a secure middle layer, the language model can indirectly access the DuckDB database. It has the characteristics of modularity, extensibility, and log tracking, laying the foundation for future upgrades to the RAG architecture.

DollhouseMCP is a server based on the Model Context Protocol (MCP) that provides dynamic AI character management functions and supports sharing and commercializing characters through the GitHub community. It includes 23 tools, covering the full lifecycle management of character creation, activation, editing, validation, etc., and has enterprise-level security measures and an automatic update system.

A secure MCP server that allows AI assistants to execute TypeScript and JavaScript code through the Deno runtime, with configurable permission control and lightweight characteristics.

Implementation of AI character model context protocol server

A Spotify playlist management server based on the MCP protocol that helps users optimize playlists by analyzing the characteristics of existing tracks and using Claude AI to make intelligent song recommendations.