steam-mcpSAFE
Read-only Steam MCP server for AI agents: 41 tools + prompts + resources across library, friends, achievements, store, reviews, sales, discovery, recommendations, and co-op planning. Official Steam Web API, bring-your-own-key, open source. On PyPI + the MCP Registry.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://pypi.org/project/steam-mcp/) [](https://pypi.org/project/steam-mcp/) [](https://github.com/Sarg338/steam-mcp/actions/workflows/ci.yml) [](LICENSE) [](https://registry.modelcontextprotocol.io)
A read-only Model Context Protocol server for the public Steam Web API and storefront — 41 tools, 5 prompts, and 2 resources that let any MCP client (Claude Desktop, Claude Code, Cursor, ...) answer questions about Steam: your friends, games, playtime, and achievements, plus account-independent things like sales, reviews, live player counts, Steam Deck compatibility, discovery, recommendations, and co-op planning.
Read-only · official Steam APIs only · open source. Nobody logs in, and the server never writes, trades, posts, launches games, or makes purchases.
Quick start — no API key needed
Install `uv`, then:
Claude Code
claude mcp add steam -- uvx steam-mcp
That's the whole setup. 19 of the 41 tools work with no credential at all — anything about the store or a game itself:
"Is Baldur's Gate 3 worth buying, and how are its recent reviews trending?" "What co-op games are on sale under £20 right now?" "How many people are playing Helldivers 2 this minute?" "Will Hades II run properly on my Steam Deck?"
The three game-finders (steam_discover, steam_should_i_buy, steam_recommend) work without a key too, as long as you don't personalize them.
Adding your own account
A [free Steam Web API key](https://steamcommunity.com/d
b7c6df1524e4OBSERVED · 2026-10-10Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add steam-mcp --env STEAM_API_KEY=${STEAM_API_KEY} -- uvx steam-mcp==1.19.0Exposed tools (20)
18 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
steam_analyze_app_reviews | read | Analyze thousands of a game |
steam_compare_games | read | Compare 2-5 games side by side — price, reviews and their trend, players now, Deck, co-op, tags — for |
steam_discover | read | Discover games by criteria — tags, max price, on-sale, platform — optionally personalized to a user |
steam_get_app_details | read | Get comprehensive store details for a game — the best |
steam_get_app_news | write | Get recent news/update posts for a game (patch notes, announcements). |
steam_get_app_regional_pricing | read | Compare a game |
steam_get_app_reviews | read | Get the review score and sample reviews for a game (lifetime and/or recent). |
steam_get_app_tags | read | Get a game |
steam_get_current_players | read | Get the number of players currently in-game for a title (live concurrency). |
steam_get_deck_compatibility | read | Steam Deck rating for a game: Verified, Playable, Unsupported, or Unknown. |
steam_get_dlc | read | List a game |
steam_get_featured_specials | read | List games currently ON SALE (featured specials) on the Steam store. |
steam_get_game_schema | read | Get the achievement and stat definitions for a game (not user-specific). |
steam_get_global_achievement_percentages | read | Get the global unlock percentage (rarity) of each achievement in a game. |
steam_get_market_price | read | Get the Community Market price for a single item, with rarity and condition. |
steam_get_package_details | read | Get store details for a Steam package (a sub/bundle of one or more games). |
steam_get_store_highlights | read | List a Steam storefront section: top sellers, new releases, or coming soon. |
steam_get_update_impact | write | Show how a game |
steam_get_workshop_item | read | Get metadata for a Steam Workshop item (mod, map, guide, collection, ...). |
steam_search_apps | read | Look up a game |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (3 observation(s))
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (9)
.mcpbignore
"http://169.254.169.254/latest/meta-data"):
for bad in ("http://169.254.169.254/latest/meta-data", # cloud metadata"https://steamcommunity.com/profiles/[email protected]"))
"http://169.254.169.254/latest/meta-data"):
for bad in ("http://169.254.169.254/latest/meta-data", # cloud metadatamcp, httpx2, pydantic
read from the `STEAM_API_KEY` environment variable. It is never written to disk,
- macOS / Linux: `curl -LsSf https://astral.sh/uv/install.sh | sh`
Gates applied: no_behavioural_pass.
b7c6df1524e4full audit observations/trust-audit/mcp-server/sarg338__steam-mcp.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-10 | b7c6df1524e4 | SAFE | B | 89 | first audit |
Questions
What is the steam-mcp MCP server?
Read-only Steam MCP server for AI agents: 41 tools + prompts + resources across library, friends, achievements, store, reviews, sales, discovery, recommendations, and co-op planning. Official Steam Web API, bring-your-own-key, open source. On PyPI + the MCP Registry.
What tools does steam-mcp expose?
20 in total: 18 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is steam-mcp safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does steam-mcp need?
It reads STEAM_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does steam-mcp run?
It speaks stdio, so it runs as a local process your client starts. It is published on PyPI as steam-mcp.
How current is this page?
The grade is for one exact copy of the source (b7c6df1524e4), read on 2026-10-10. The repository is watched and re-audited when it changes.