mcp@0.19.9
CAUTION· startup onlyPublisher claims: "Deploy and operate multiplayer game servers on Edgegap's global edge network." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Startup only; authenticated tool behavior not tested.- Tested
- server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
- Not tested
- the tool list (the server exited before answering) · what the tools do when called with real credentials
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
EDGEGAP_API_TOKENEDGEGAP_REGISTRY_TOKENConfiguration, not credentials: EDGEGAP_APP_ALLOWLIST, EDGEGAP_BASE_URL, EDGEGAP_ENABLE_SKILLS_UPDATE, EDGEGAP_MAX_AGENT_DEPLOYMENTS, EDGEGAP_MAX_DEPLOYS_PER_RUN, EDGEGAP_MAX_DURATION_MINUTES, EDGEGAP_READ_ONLY, EDGEGAP_SKILLS_CHECK, EDGEGAP_SKILLS_RAW_URL, EDGEGAP_STOP_OWN_ONLY, EDGEGAP_TIMEOUT_MS- Where it talks to
- edgegap.com (its own)Also githubusercontent.com, amazonaws.com — not the publisher's domain; check what for before connecting.
- What we saw when it ran
- no credential files read · no network
- What we could not test
- the tool list (the server exited before answering) · what the tools do when called with real credentials
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names githubusercontent.com, amazonaws.com as well as its own vendor. The guard injects EDGEGAP_API_TOKEN, EDGEGAP_REGISTRY_TOKEN only into requests to edgegap.com (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.
What changed since 0.17.0
vet-skills diff mcp 0.17.0 0.19.9In the sandbox, this version reached for nothing 0.17.0 did not. Compared from the two recorded runs, not from source.
The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-20 and scored it 77/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Prompting user to enter sensitive credentials. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 37 s · 2 scenarios · 0 deviationsStatic findings
'pass through Edgegap infrastructure — the local server (npx @edgegap/mcp) ' +Fix: Pin versions and sources; vendor anything fetched at install time.' apt-get clean && rm -rf /var/lib/apt/lists/*',Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.: `RUN chmod +x /app/${executable}`,Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.? `RUN sed -i 's/\\r$//' /app/${executable} && chmod +x /app/${executable}`Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.? `# Scripts saved on Windows carry CRLF endings, which break the shebang.\nRUN sed -i 's/\\r$//' /app/${executable} && chmod +x /app/${executable}`Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.description: 'Paste the token value. It is held in one variable in memory for ' +Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.: `RUN chmod +x /app/${executable}`,Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.* hosted server refuses it as foreign. The stdio server (npx @edgegap/mcp)Fix: Pin versions and sources; vendor anything fetched at install time.`RUN chmod +x /root/build/${executable}`,Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.'so prefer the local server (npx -y @edgegap/mcp), which keeps the token ' +Fix: Pin versions and sources; vendor anything fetched at install time.// commands and neither named `mcp`, `npx @edgegap/mcp` cannot pick one andFix: Pin versions and sources; vendor anything fetched at install time.'runs each outdated install\'s own updater (npx ' + SKILLS_CLI + ' update for the Edgegap skills only, or claude plugin ' +Fix: Pin versions and sources; vendor anything fetched at install time.'(npx -y @edgegap/mcp) with EDGEGAP_API_TOKEN instead.'Fix: Pin versions and sources; vendor anything fetched at install time.'(npx -y @edgegap/mcp).\n\n' +Fix: Pin versions and sources; vendor anything fetched at install time.const howToUpdate = 'Reinstall the way they were installed: npx skills add edgegap/llm-agent-skills --skill "*" (add -g for user-wide); ' +Fix: Pin versions and sources; vendor anything fetched at install time.warn('no-chmod', 'No "chmod +x" on the server binary. Builds copied from Windows often lose the executable bit, giving "permission denied" at start.', 'Add "RUN chmod +x /path/to/ServerBuild".');Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.…LS_PACK.version}. A newer @edgegap/mcp may carry it: restart the server through npx -y @edgegap/mcp@latest, or update the Edgegap extension. The bundled skills are st…Fix: Pin versions and sources; vendor anything fetched at install time.// npx and claude are .cmd shims on Windows, which only start through aFix: Pin versions and sources; vendor anything fetched at install time.'returned config to a file (e.g. matchmaker-config.json) and have the developer upload it ' +Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.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
Not checked against OSV yet; the weekly check has not reached this version.
What would make this Verified
3 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 edgegap: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- npx invocation of non-well-known package≈ +2 on the totalPin versions and sources; vendor anything fetched at install time.
- Destructive file system command detected≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
- File type combined with upload action≈ +2 on the totalRemove the call, or declare the host in frontmatter and explain in the README what is sent and why.