steamworks-mcp@0.5.2
CAUTION· not startedPublisher claims: "Release assistant for Steam games, as an MCP server for Claude and ChatGPT: store page, achievements, checks against Valve's rules, market data, and Steamworks only after you approve." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Not started: dependency install failed; the server itself was not observed.- Tested
- the dependency install step · every file read, file written and host contacted by what ran
- Not tested
- server startup · the tool list · tool behavior
What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
STEAM_WEB_API_KEYSTEAMWORKS_MCP_TOKENConfiguration, not credentials: SANITIZE_DENYLIST, STEAMWORKS_MCP_ROOT- Where it talks to
- no vendor host named in the codeAlso steampowered.com, steamgames.com, steam-api.com, chatgpt.com, claude.ai, claude.com, steamcommunity.com, cloudflare.com — not the publisher's domain; check what for before connecting.
- What the honeypot saw
- no credential files read · no network
- What we could not test
- server startup · the tool list · tool behavior
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names steampowered.com, steamgames.com, steam-api.com, chatgpt.com, claude.ai, claude.com, steamcommunity.com, cloudflare.com as well as its own vendor. The guard injects STEAM_WEB_API_KEY, STEAMWORKS_MCP_TOKEN only into requests to the host you bind (and its subdomains) and refuses the rest.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; not started: dependency install failed; the server itself was not observed) on 2026-11-19 and scored it 68/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Access to sensitive file paths; Access to sensitive file paths; Hidden zero-width characters detected. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenarios · 0 deviationsStatic findings
scripts: keep the key on a server or in an untracked .env, log steamcmd in once by hand, and never commitFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.for path in (settings, dotenv if dotenv is not None else Path.cwd() / ".env"):Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.return "" + buf.getvalue()Fix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.code = (run or (lambda a: subprocess.run(a, check=False).returncode))(args)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out."""Configuration from the environment and an optional ``.env`` file (a small parser of our own, no dependency).Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.parser.add_argument("--env-file", help=".env file to read (default: ./.env if present)")Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.TUNNEL_URL = re.compile(r"https://[a-z0-9-]+\.trycloudflare\.com")Fix: Replace encoded or assembled strings with plain literals and remove undeclared network calls."""The real environment wins over a ``.env`` (in the working folder, or ``dotenv``), which wins over the user'sFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."terminal (they type the password and Steam Guard code), then try again."Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.proc = subprocess.Popen(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.proc = subprocess.run(Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.key = load_config(dotenv=ROOT / ".env").publisher_keyFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Web API key and put it in .env."Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.- SANITIZE_DENYLIST terms (from .env): app names -> "ExampleGame", everything else -> "Redacted"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.ROOT_GITIGNORE_SUGGESTIONS = (".env",)Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.if name.startswith(".env"):Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.env = ROOT / ".env"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.else "Ask the user to sign in to Steamworks in the browser window that opened (they type the password and "Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.fix: Leave steam_appid.txt out of the build folders and depots you upload.Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.quote: If you use Root Overrides, you must specify [All OSes] in the Root OS drop-down above.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.if "(" in around: # expanded on first use: "time to kill (TTK)" or "TTK (time to kill)"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.- **ChatGPT on the web**: Settings > Apps & Connectors > Advanced settings: turn on Developer mode, then create aFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.rule: By default, Auto-Cloud files sync only to the root's OS. For cross-platform saves, use one root plus Root Overrides.Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.- **ChatGPT on the web**: Settings > Apps & Connectors > Advanced settings: turn on Developer mode, then create aFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.Keep the terminal from step 2 open while you work. Custom connectors in ChatGPT need developer mode; check that yourFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.res = subprocess.run(["claude", "mcp", "get", SERVER_NAME], capture_output=True, text=True, check=False)Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.rule: "With Root Overrides, the original root's OS must be [All OSes]. An override has five parts: Original Root, OS, New Root, Add/Replace Path, Replace Path."Fix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.Keep the terminal from step 2 open while you work. Custom connectors in ChatGPT need developer mode; check that yourFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.Cost to the agent
Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.
Known vulnerabilities
No advisories in OSV for steamworks-mcp@0.5.2 as of 5 h ago. This is the package itself, not its dependencies.
What would make this Verified
12 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Prompting user to enter sensitive credentials≈ +9 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Unclaimed publisher≈ +8 on the totalClaim wazzapsenk: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Access to sensitive file paths≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Hidden zero-width characters detected≈ +4 on the totalKeep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.
- Tunneling service URL≈ +4 on the totalReplace encoded or assembled strings with plain literals and remove undeclared network calls.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.