steam-mcp

steam-mcp

An MCP server that connects AI assistants to the Steam Web API and storefront, enabling game library analysis, mood-based backlog recommendations, wishlist evaluation, and storefront searches.

Category
访问服务器

README

steam-mcp

An MCP (Model Context Protocol) server that connects AI assistants to the Steam Web API and storefront services. Enables game library analysis, backlog recommendations based on mood, wishlist evaluation, and storefront searches. Works alongside itad-mcp and opencritic-mcp.

npm version

What this does

This server acts as the foundational layer of a 3-server gaming stack:

  • steam-mcp (this one) — understands your Steam library, playtime, backlog, wishlist, and tastes. It surfaces game AppIDs.
  • opencritic-mcp — takes game names/AppIDs and provides detailed critic scores and reviews.
  • itad-mcp — takes game names/AppIDs and finds the best deals across legitimate storefronts and historical lows.

Example workflow: User: "I want something atmospheric and chill to play tonight."

  1. steam-mcp (recommend_from_library) finds unplayed "Atmospheric" + "Relaxing" games in your backlog.
  2. opencritic-mcp validates which of those games have great critic scores.
  3. User: "These look great, any of them on sale?"
  4. itad-mcp takes the AppIDs and checks if any are on sale right now.

Wishlist deal check: get_wishlist → take AppIDs → itad-mcp get_historical_low → surface any games at or near their historical low price.

Tools

Tool Parameters Description Requires API Key?
get_library steam_id (string, optional)<br>sort ("playtime" | "recent" | "name")<br>filter ("all" | "unplayed" | "played")<br>limit (number, default 50) Retrieve the user's owned Steam game library with playtime statistics. Yes
get_recently_played steam_id (string, optional) Games played in the last 2 weeks. Yes
get_wishlist steam_id (string, optional)<br>sort ("priority" | "date_added" | "name")<br>limit (number, default 50) Retrieve the user's Steam wishlist to find interesting games. Yes
get_app_details appid (number)<br>country_code (string, default 'us') Full Steam Store metadata for a specific game (description, genres, categories, price, metacritic). No
get_app_reviews appid (number)<br>filter ("all" | "recent" | "positive" | "negative") Steam user review sentiment for a game. No
search_games query (string)<br>limit (number, default 10) Resolve a game title to a Steam AppID. No
recommend_from_library steam_id (string, optional)<br>tags (string[])<br>unplayed_only (boolean, default true)<br>limit (number, default 10) Find games in the user's library that match specified tags/mood. Yes
discover_games tags (string[])<br>steam_id (string, optional)<br>max_results (number, default 20)<br>country_code (string, default 'us') Find well-reviewed games on Steam matching specific tags that the user does NOT own. Optional

Prerequisites

  • Node 18+
  • Steam account with a public library
  • Steam Web API key

Getting a Steam API Key

  1. Go to https://steamcommunity.com/dev/apikey
  2. Log in with your Steam account.
  3. Enter any domain name (e.g. localhost).
  4. Click "Register" and copy the generated 32-character API key.

Finding your Steam ID

You can use either your vanity name or your 17-digit SteamID64.

  • If your profile URL is https://steamcommunity.com/id/your_vanity_name/, your Steam ID is your_vanity_name.
  • If your profile URL is https://steamcommunity.com/profiles/76561198012345678/, your Steam ID is 76561198012345678.

Make your profile public

For the API to read your library and playtime, your Game Details privacy setting must be Public.

  1. Go to your Steam Profile -> Edit Profile -> Privacy Settings.
  2. Set "My Profile" to Public.
  3. Set "Game details" to Public.
  4. Set "Inventory" to Public if you want wishlist access.

Installation

Option A: Using npx (Recommended)

Add this to your MCP configuration JSON (mcp_config.json):

{
  "mcpServers": {
    "steam": {
      "command": "npx",
      "args": ["-y", "steam-user-mcp"],
      "env": {
        "STEAM_API_KEY": "your_api_key_here",
        "STEAM_ID": "your_vanity_name"
      }
    }
  }
}

Option B: Clone & Build

git clone https://github.com/brandikun/steam-mcp
cd steam-mcp
npm install
npm run build

Then add this to your MCP config:

{
  "mcpServers": {
    "steam": {
      "command": "node",
      "args": ["/path/to/steam-mcp/dist/index.js"],
      "env": {
        "STEAM_API_KEY": "your_api_key_here",
        "STEAM_ID": "your_vanity_name"
      }
    }
  }
}

MCP Configuration variables

  • STEAM_API_KEY (Required): Your 32-character Steam Web API key.
  • STEAM_ID (Optional): The default user to query for library and playtime tools. If not provided, it must be passed per tool call.

Usage with itad-mcp and opencritic-mcp

For the ultimate gaming assistant, combine all three:

{
  "mcpServers": {
    "steam": {
      "command": "npx",
      "args": ["-y", "steam-user-mcp"],
      "env": { "STEAM_API_KEY": "your_api_key", "STEAM_ID": "your_id" }
    },
    "itad": {
      "command": "npx",
      "args": ["-y", "itad-mcp"],
      "env": { "ITAD_API_KEY": "your_itad_key" }
    },
    "opencritic": {
      "command": "npx",
      "args": ["-y", "opencritic-mcp"],
      "env": { "RAPIDAPI_KEY": "your_rapidapi_key" }
    }
  }
}

Troubleshooting

  • "No Steam ID provided": You must set STEAM_ID in your environment or provide it in the tool call.
  • "Steam API key is required": The STEAM_API_KEY environment variable is missing.
  • "Vanity URL not found": The Steam ID provided couldn't be resolved. Ensure it's your exact vanity name or a valid SteamID64.
  • Library returns empty: Your profile is private. See the "Make your profile public" section above.
  • Tags not matching: Steam tags are very specific. The server does its best with fuzzy matching, but if a tag isn't found, try a broader synonym (e.g., "RPG" instead of "Role-playing").

Privacy

Credentials are only read from environment variables. No data is logged or sent anywhere except Steam's own APIs.

License

MIT

推荐服务器

Baidu Map

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。

官方
精选
JavaScript
Playwright MCP Server

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。

官方
精选
TypeScript
Magic Component Platform (MCP)

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。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

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。

官方
精选
本地
TypeScript
VeyraX

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。

官方
精选
本地
graphlit-mcp-server

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。

官方
精选
TypeScript
Kagi MCP Server

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。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

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。

官方
精选