Steam Review and Forum MCP
Provides tools for fetching and analyzing Steam store reviews, community forums, and game metadata, enabling querying, filtering, aggregation, and temporal analysis of player feedback.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Steam Review and Forum MCPsummarize major complaints in recent negative reviews for Elden Ring"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Steam Review and Forum MCP
Languages: English | 简体中文
MCP server for exploring Steam store reviews and community discussion threads, built for the question behind most Steam research:
What do players actually think about this game once you get past the store page noise?
It can answer questions like:
"Summarize the biggest complaints in recent negative reviews for Steam game XXX."
"Query reviews that talks about performance, and tell me whether the issue sounds widespread."
"Compare launch-period negative reviews with recent positive reviews. What changed?"
"Show month-by-month sentiment since release."
"List recent Events & Announcements threads, and summarize both the patch notes and player reaction."
"How many negative reviews come from players with at least 10 hours at review time?"
"What do long-playtime negative reviewers complain about?"
"Compare English reviews with all-language reviews and tell me what differs."
"Read recent forum threads and tell me whether controller support is broken."
"Based on the reviews from the DLC pages, which DLCs of this game are worth buying?"
Key Features
Server-side filtering and aggregation
Instead of pulling thousands of reviews into the model at once, you can create a saved review dataset once and let the server do the heavy lifting. Query only the reviews that mention
"performance"or"crash", filter by sentiment, date range, language, or playtime, and keep chat context focused on the signal.Temporal precision
This MCP is good at time-based analysis. You can isolate launch-period noise, compare it against a later period, and see how player priorities shift over time. Monthly and weekly trend buckets make that easy to quantify. This is especially effective when the real question is not just "what are people saying?" but "what changed, when did it change, and which players are saying it?"
Metadata context in addition to raw review text
With metadata like
timestamp_created,voted_up,author.playtime_at_review,author.playtime_forever, it's possible to separate quick bounce-offs from long-term players and identify which complaints were genuinely influential.Reviews, forums, and official announcements in one workflow
The same server can inspect Steam reviews, public discussion sections, multi-page threads, and Events & Announcements.
Related MCP server: Steam MCP Server
Quick Start
Requirements
Node.js
22.19+npm
Use from npm
npx -y steam-review-and-forum-mcpThat command starts the MCP server over stdio. Most MCP clients will run it for you from their config, so you usually do not need to launch it manually.
Storage
The server writes saved review and forum datasets to local disk. If STEAM_REVIEW_EXPORT_DIR and STEAM_FORUM_EXPORT_DIR are not set, the defaults are package-relative:
.steam-review-exports/.steam-forum-exports/
When using npx, that means the npm/npx-installed package copy. When running from a source checkout, that means the checkout root. The package-relative default works, but for durable storage you should set explicit absolute paths in your MCP client config.
For JSON-based MCP configs, add env:
{
"mcpServers": {
"steam-review-and-forum": {
"command": "npx",
"args": ["-y", "steam-review-and-forum-mcp"],
"env": {
"STEAM_REVIEW_EXPORT_DIR": "<absolute-path-to-review-exports>",
"STEAM_FORUM_EXPORT_DIR": "<absolute-path-to-forum-exports>"
}
}
}
}For Codex config.toml, add an environment table:
[mcp_servers.steam-review-and-forum]
command = "npx"
args = ["-y", "steam-review-and-forum-mcp"]
[mcp_servers.steam-review-and-forum.env]
STEAM_REVIEW_EXPORT_DIR = "<absolute-path-to-review-exports>"
STEAM_FORUM_EXPORT_DIR = "<absolute-path-to-forum-exports>"Claude Desktop
Open Claude Desktop Settings > Developer > Edit Config, add this server to claude_desktop_config.json, then restart Claude Desktop:
{
"mcpServers": {
"steam-review-and-forum": {
"command": "npx",
"args": ["-y", "steam-review-and-forum-mcp"]
}
}
}Codex CLI
codex mcp add steam-review-and-forum -- npx -y steam-review-and-forum-mcpCodex App
In the Codex app, open Settings > Integrations & MCP and add a custom server, or edit ~/.codex/config.toml:
[mcp_servers.steam-review-and-forum]
command = "npx"
args = ["-y", "steam-review-and-forum-mcp"]Claude Code
claude mcp add steam-review-and-forum -- npx -y steam-review-and-forum-mcpCursor
Create or update .cursor/mcp.json:
{
"mcpServers": {
"steam-review-and-forum": {
"type": "stdio",
"command": "npx",
"args": ["-y", "steam-review-and-forum-mcp"]
}
}
}Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"steam-review-and-forum": {
"command": "npx",
"args": ["-y", "steam-review-and-forum-mcp"]
}
}
}Other Stdio-Compatible MCP Clients
Most local MCP clients accept a config shaped like this:
{
"mcpServers": {
"steam-review-and-forum": {
"command": "npx",
"args": ["-y", "steam-review-and-forum-mcp"]
}
}
}MCP Inspector
npx -y @modelcontextprotocol/inspector -- npx -y steam-review-and-forum-mcpRun the Server Manually
npx -y steam-review-and-forum-mcpMost MCP clients will start the server for you, so manual launch is mainly useful for debugging.
Development from Source
If you want to run a local checkout instead of the published npm package:
npm install
npm run build
node build/server.jsTools at a Glance
Note that the mcp server should already work out of the box, and you don't need to know the technical details below to use it.
For the exact MCP tool input schemas, see docs/TOOL_SCHEMAS.md.
Reviews
get_steam_game_info: cleaned Steam store metadata in Englishget_steam_review: interactive review fetch for one page or a small bounded batchcreate_steam_review_corpus: background fetch for a large review dataset you want to save on the serverget_steam_review_corpus_status: progress and metadata for a saved review datasetquery_steam_review_corpus: filtered review retrieval from a saved review dataset by date, sentiment, language, playtime, text, and sort orderaggregate_steam_review_corpus: server-side counts, trends, playtime averages, and language breakdowns from a saved review dataset
Forums
list_steam_forum_sections: discover available discussion forum sectionslist_steam_forum_topics: list topics in a sectionget_steam_forum_topic: fetch a topic and its repliescreate_steam_forum_topic_corpus: background fetch for a long multi-page thread you want to save on the serverget_steam_forum_topic_corpus_status: progress and metadata for a saved forum thread datasetread_steam_forum_topic_corpus_chunk: read one stored reply chunk from a saved long thread
Operational Notes
Saved review datasets are stored in
STEAM_REVIEW_EXPORT_DIRwhen set; otherwise they are stored in.steam-review-exports/next to the installed package.Saved forum thread datasets are stored in
STEAM_FORUM_EXPORT_DIRwhen set; otherwise they are stored in.steam-forum-exports/next to the installed package.Stored exports are cleaned up automatically after
24hours by default.Review and forum fetches retry transient failures and
429responses with backoff.If the process restarts mid-fetch, later status or chunk reads can restart resumable jobs automatically.
Environment Variables
Use these only if you need to tune storage or fetch behavior:
STEAM_REVIEW_EXPORT_DIRSTEAM_FORUM_EXPORT_DIRSTEAM_REVIEW_EXPORT_TTL_HOURSSTEAM_FORUM_EXPORT_TTL_HOURS
License
This project is licensed under the BSD 3-Clause License. See LICENSE.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/icue/SteamReviewAndForumMcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server