Lutris MCP Server
Enables AI assistants to browse, search, and manage a Lutris Linux gaming library. It provides tools for modifying game details, organizing categories, and accessing collection statistics directly from the Lutris SQLite database.
README
lutris-mcp
MCP server for managing your Lutris game library. Gives AI assistants direct access to browse, search, organize, launch, and install games in your Linux gaming collection.
Examples
Here are some things you can ask your AI assistant once this server is connected:
- "What games do I have installed?"
- "Show me my most played games"
- "Search my Steam library for Hades"
- "Launch Balatro"
- "Install this game from
/path/to/setup.exe" - "Add all my roguelike games to a 'Roguelike' category"
- "Are there any duplicate games in my library?"
- "Show me the log for my last Hades II session"
- "Export my installed games as JSON"
- "Set the cover art for Timberborn from this image"
Installation
Quick Start (npx)
No install needed — just add to your MCP client config:
{
"mcpServers": {
"lutris": {
"command": "npx",
"args": ["-y", "lutris-mcp"]
}
}
}
Global Install
npm install -g lutris-mcp
Then configure your MCP client:
{
"mcpServers": {
"lutris": {
"command": "lutris-mcp"
}
}
}
From Source
git clone https://github.com/Praeses0/lutris-mcp.git
cd lutris-mcp
npm install
npm run build
{
"mcpServers": {
"lutris": {
"command": "node",
"args": ["/path/to/lutris-mcp/dist/index.js"]
}
}
}
Requirements
- Lutris installed and set up
- Node.js 20+
Tools
Library
| Tool | Description |
|---|---|
list_games |
List and filter games with pagination, sorting, and smart search |
get_game |
Get full game details including categories, config, and media paths |
add_game |
Add a new game to the library |
update_game |
Update fields on an existing game |
remove_game |
Remove a game from the database (does not delete files) |
get_library_stats |
Aggregate stats: totals, playtime, breakdowns by runner/platform/service |
find_duplicates |
Find potential duplicate games (same directory or similar slugs) |
export_library |
Export full library as JSON with optional filters (category, runner, installed) |
Categories
| Tool | Description |
|---|---|
list_categories |
List all categories with game counts |
create_category |
Create a new category |
assign_category |
Add a game to a category |
unassign_category |
Remove a game from a category |
bulk_assign_category |
Add multiple games to a category at once |
Services
| Tool | Description |
|---|---|
search_service_games |
Search games synced from Steam, GOG, etc. |
import_service_game |
Import a service game into your Lutris library |
Launch & Install
| Tool | Description |
|---|---|
launch_game |
Launch an installed game via Lutris |
install_game |
Install a game from a Lutris installer slug or local setup executable |
check_game_running |
Check if a game is currently running (multi-strategy detection) |
Configuration
| Tool | Description |
|---|---|
read_game_config |
Read a game's YAML configuration file |
write_game_config |
Update or create a game's YAML configuration (deep-merged) |
Media
| Tool | Description |
|---|---|
set_game_cover |
Set a game's cover art, banner, or icon from a local file |
get_game_media |
Get all media paths for a game with existence check |
System
| Tool | Description |
|---|---|
list_runners |
List available Lutris runners or Wine versions |
view_game_log |
Read the last launch log for a game (for troubleshooting) |
Bulk Operations
| Tool | Description |
|---|---|
bulk_update_games |
Update a field on multiple games at once |
Environment Variables
| Variable | Default | Description |
|---|---|---|
LUTRIS_DB_PATH |
~/.local/share/lutris/pga.db |
Path to Lutris SQLite database |
LUTRIS_GAMES_CONFIG_DIR |
~/.config/lutris/games |
Path to game YAML configs |
LUTRIS_DATA_DIR |
~/.local/share/lutris |
Path to Lutris data directory (media, logs) |
Development
npm install # install dependencies
npm run build # compile TypeScript
npm test # run tests (219 tests)
npm run dev # run with tsx (hot reload)
License
MIT
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。