Game Discovery MCP Server
Discovers and manages installed games on Windows PC from platforms like Steam, Epic Games, GOG, and more. Provides tools to list, get details, and launch games via MCP.
README
Game Discovery MCP Server
<!-- mcp-name: io.github.moonolgerd/game-mcp -->
A Model Context Protocol (MCP) server that discovers and manages installed games on Windows PC from various platforms including Steam, Epic Games, GOG, Windows Store/Xbox, and other installed programs.
Features
-
Multi-platform game discovery: Automatically scans for games from:
- Steam (including custom library locations)
- Epic Games Store
- GOG (Good Old Games)
- Xbox/Windows Store games
- EA games
- Ubisoft Connect games
- Rockstar games
- Battle.Net games
- Other installed programs (via Windows registry)
-
Comprehensive game information: Provides details including:
- Game name and platform
- Installation path and executable location
- Installation date and size
- Platform-specific metadata
-
MCP Tools Available:
discover_games: Finds all installed games across all platformsget_game_info: Gets detailed information about a specific gamelaunch_game: Launches a game by name (if executable is found)
Prerequisites
- .NET 10.0 or later
- Windows operating system
Usage with VS Code
{
"servers": {
"GameMcpServer": {
"type": "stdio",
"command": "dnx",
"args": [
"GameMcpServer@1.0.8",
"--yes"
]
}
}
}
Available Tools
discover_games
Discovers all installed games from all supported platforms.
Example usage in Claude: "Can you discover all the games I have installed on my PC?"
Returns: JSON object with:
- Total game count
- Games grouped by platform
- Detailed list of all games with metadata
get_game_info
Gets detailed information about a specific game by name.
Parameters:
gameName: The name of the game to search for
Example usage: "Get information about Cyberpunk 2077"
launch_game
Launches a game by name if it has a valid executable path.
Parameters:
gameName: The exact name of the game to launch
Example usage: "Launch Steam's Half-Life 2"
How It Works
The server uses multiple discovery methods:
-
Steam Games:
- Reads Steam installation path from registry
- Parses
libraryfolders.vdffor custom library locations - Scans
steamapps/commondirectories for game folders
-
Epic Games:
- Reads game manifests from
%ProgramData%/Epic/EpicGamesLauncher/Data/Manifests - Parses JSON manifest files for game information
- Reads game manifests from
-
GOG Games:
- Queries Windows registry under
SOFTWARE\GOG.com\Games - Extracts game paths and executable information
- Queries Windows registry under
-
Xbox/Windows Store:
- Scans
%ProgramFiles%/WindowsAppsdirectory - Parses AppxManifest.xml files for application information
- Filters for gaming applications
- Scans
-
Registry Programs:
- Scans Windows uninstall registry entries
- Uses heuristics to identify game-like applications
- Filters out system utilities and non-gaming software
Error Handling
The server is designed to be robust and continue functioning even if:
- Some platforms are not installed
- Registry keys are inaccessible
- Directories have permission restrictions
- Individual games have corrupted metadata
Errors are logged to stderr and don't interrupt the discovery process for other platforms.
Security Considerations
- The server only reads system information and doesn't modify anything
- Game launching requires explicit user confirmation through the MCP client
- All file system access is read-only except for launching executables
- Registry access is read-only
Development
This project uses:
- .NET 10.0
- Model Context Protocol SDK for .NET
- Microsoft.Extensions.Hosting for application lifecycle
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。