fuel-mcp@r1
CAUTION· not startedPublisher claims: "Enables finding the cheapest nearby fuel station for diesel or petrol, returning its price, distance, and age in a single sentence. Supports simulated data and configurable real price sources, including use in voice sessions over stdio." 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
- no credential variables read in the code
- Where it talks to
- no vendor host named in the code
- What we saw when it ran
- no credential files read · no network
- What we could not test
- server startup · the tool list · tool behavior
Connect it under the guard. Not started: dependency install failed; the server itself was not observed. The code names no credential variable; the guard pins the network to what the honeypot saw and logs every call.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
Is fuel-mcp@r1 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 2026-11-22 and scored it 76/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: HTML comment carrying imperative instructions to the agent. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 0 s · 2 scenarios · 0 deviationsStatic findings
not run: the workflow runs each of these; nesting them here would run the suite inside itselfFix: Keep SKILL.md to visible, descriptive text. Remove comments and encoded text that address the agent.result = subprocess.run( # noqa: S603Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.msg = f"{folder} is open to others, so nothing private goes in it. Fix: chmod go-rwx {folder}"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.kill -9 -- -<its pid>, which this script prints; the next rule check then puts its file back.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.done = subprocess.run( # noqa: S603 - this interpreter, asked one fixed questionFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.f" another terminal with: kill -9 -- -{child.pid}\n"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.fix = f"chmod go-w {path}" if facts.st_uid in owners else f"move the private folder out of {path}"Fix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.# RULE: a kill by another test names the tests that went redFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.subprocess.run would kill it outright: a rule check would leave a file switched off,Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.``chmod go-rwx <folder>``, with one log line. The log never holds what the fileFix: Request the narrowest permission that works; replace wildcards with the exact paths or hosts.# A kill that can flip between runs has to say what flipped it.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.# A kill by some other test is not evidence about this rule.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.name = "a kill by another test names the tests that went red"Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.if importlib.util.find_spec(PACKAGE) is not None:Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.twice. Its output goes to files, not pipes: after a kill -9 of this run, a pipe wouldFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.r = subprocess.run( # noqa: S603 - this interpreter, pytest, and a folder this script madeFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.done = subprocess.run(command, cwd=ROOT, capture_output=True, encoding="utf-8", check=True) # noqa: S603Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.child = subprocess.Popen( # noqa: S603 - the commands are this script's ownFix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.A kill is not enough on its own. A mutation breaks whatever it breaks, and itFix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.result = subprocess.run( # noqa: S603Fix: Build commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.# turns colour back on, which would otherwise make every kill unattributable.Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.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
Vulnerability advisories (OSV) cover npm and PyPI packages; this registry has no advisory database.
What would make this Verified
4 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim roykollensvendsen: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
- New publisher≈ +3 on the totalTime clears this on its own after 14 and 30 days. Claiming the handle (+50) is the fastest way past it.
- Python subprocess execution detected≈ +2 on the totalBuild commands from fixed arguments, never from interpolated strings; prefer an API client over shelling out.
- File permission modification command≈ +2 on the totalRequest the narrowest permission that works; replace wildcards with the exact paths or hosts.
- Process kill command≈ +2 on the totalBuild commands from fixed arguments; do not interpolate untrusted strings into a shell.