Searches Steam, retrieves reviews and announcements, and analyzes sentiment without requiring an API key.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent ā or use 1-click editor setup below.
We haven't yet run this listing's install command through our automated sandbox check. This isn't a red flag ā we're steadily working through the catalog.
š” Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Steam Reviews MCP.
The jhomen368/steam-reviews-mcp MCP server exposes Steam game discovery and review analysis through MCP tools. It can search for games by name, including batches of up to five queries, and return AppID and price information. A separate game-information tool retrieves regional store data, purchase notices, language support, store categories, system requirements, downloadable content, and Steam Deck compatibility when Steam provides those details.
Review retrieval supports pagination and filters for recent or all-time reviews, positive or negative review types, date windows, Steam Deck-related reviews, and off-topic activity associated with review bombing. Sentiment analysis can process a sample of reviews generally or focus on a supplied topic such as performance. Results can include representative review quotes with links to the Steam community.
The server also reads official Steam app announcements. Responses may include publication details, Steam markup, and cursors for loading older posts. This makes the server suitable for questions that combine store metadata, community feedback, and developer-published updates.
The jhomen368/steam-reviews-mcp MCP server uses Steam's public Store API and does not require an API key. MCP clients call the exposed tools with structured arguments, and the server validates inputs with Zod schemas before making requests. Steam data is returned with context about regional country and language requests rather than being silently converted or interpreted.
Search accepts either one query or a batch of up to five. Game-information requests can specify an ISO two-letter store country code and a Steam language code; defaults are the United States and English. Optional criteria can filter results by review score, review count, Metacritic availability, or free status. Steam Deck results preserve Valve's compatibility evidence, including unknown states when no published result is available.
Caching is included to reduce repeated API requests, with variable time-to-live behavior. The implementation also validates malformed or incomplete upstream data and can retain warnings when Steam does not return usable store, language, or compatibility information.
Install the jhomen368/steam-reviews-mcp MCP server globally with npm, then configure Claude Desktop to launch it through npx. The documented configuration uses the package @jhomen368/steam-reviews-mcp and the -y option so the package runner does not wait for interactive confirmation. Node.js 18 or newer and npm, or a compatible package manager, are required.
The project can also be built from source: clone the repository, install dependencies, build the TypeScript project, and run node build/index.js. A Docker image is available at ghcr.io/jhomen368/steam-reviews-mcp:latest. In HTTP mode, the example maps port 8086 and exposes MCP over Server-Sent Events at http://localhost:8086/mcp; the health endpoint is http://localhost:8086/health.
No environment variables or credentials are documented as required. The Docker image is designed to run as a non-root user and uses a minimal Alpine-based build.
search_steam_games: searches by one name or a batch of up to five names, with limits and AppID lookup support.get_game_info: returns regional pricing, purchase notices, language declarations, store features, requirements, DLC data, and Deck compatibility.fetch_reviews: retrieves reviews with pagination, time, review-type, and off-topic activity filters.analyze_reviews: performs sentiment analysis, supports a sample size, drills into topics, and returns example quotes.fetch_app_announcements: retrieves official app posts, preserves available Steam markup, and paginates backward with cursors.Returned store prices are regional Steam quotes and are not converted between currencies. Missing price fields do not establish that a game is free, and an unavailable quote does not identify the specific reason for unavailability. Steam may also fall back from a requested language without reporting which language was actually served.
Purchase notices are preserved from Steam but are not treated as proof that a game has or lacks a launcher, DRM, account requirement, or online-only behavior. Likewise, store category declarations are not independent tests of actual features. Announcement bodies can be truncated or malformed, and the response status should be checked before treating them as complete. The server relies on public Steam data, so availability and completeness depend on what Steam returns.
Factual signals from GitHub, npm, and our automated checks ā not a rating.
No reviews yet ā be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/jhomen368-steam-reviews-mcp)<a href="https://allmcps.com/mcp/jhomen368-steam-reviews-mcp"><img src="https://allmcps.com/api/badge/jhomen368-steam-reviews-mcp?style=directory" alt="Steam Reviews MCP on AllMCPs" /></a>