Grades are a hygiene check, not a security guarantee.

MCP, MCP server

io.github.H129hj/checkmcp

io.github.H129hj/checkmcp, an MCP server listed in the PlugVet directory.

Compatibility: Not stated.

Grade D

Two or more Medium findings, or a High finding. Only try it in a sandbox.

Graded by PlugVetRubric v2

Not a guarantee. A hygiene check of the source at the commit shown, not a security audit.

Rescanned at commit 6d272b1cf568, unchanged since the check below.

AGENT_IGNORE_INSTRUCTIONS NETWORK_HOST_FROM_INPUT

Why this grade

  • AGENT_IGNORE_INSTRUCTIONS High

    Text tells the agent to ignore earlier instructions.

    web/lib/learn.ts, line 778 "Tool poisoning is the headline risk: hidden, imperative instructions planted in tool metadata or outputs (\"ignore previous instructions\", \"do not tell the user\", \"also forwar

  • NETWORK_HOST_FROM_INPUT Medium

    A file takes a URL or endpoint as input and has no fixed host list.

    import_mcpizy.py, line 21 return json.loads(urllib.request.urlopen(req, timeout=30).read().decode())

Any High, or two or more Medium findings, give D. One Medium gives C. Low findings or a security Note give B. Each code names a finding behind the letter.

If you think this grade is wrong, email [email protected] with the entry link and your evidence. We aim to acknowledge within two working days and decide within seven. If we confirm an error, we aim to correct it within 24 hours of confirming it. These are aims, not guarantees.

Grades are a hygiene check of the source at the commit shown, on the date checked, not a security audit or a guarantee. The free directory grades are automated. Before release, a PlugVet engineer reviews every paid report and any grade we hold back, including all F grades, against the source code and the evidence we quote.

What was checked

Plain notes from the hygiene checks.

  • Permissions

    Note

    requests shell access. The manifest names this permission.

    PERMISSION_SHELL_DECLARED

    bench.py, line 74

    out = subprocess.run(

    bench_agentic.py, line 53

    out = subprocess.run(cmd, capture_output=True, text=True, timeout=300)

    checkmcp/scanners.py, line 59

    r = subprocess.run(cmd, capture_output=True, text=True, timeout=timeout)

  • Install and update

    Pass

    No install-time script found, and dependencies are pinned or absent.

  • Network

    Serious

    accepts a host from input, with no fixed list in this file.

    NETWORK_HOST_FROM_INPUT

    import_mcpizy.py, line 21

    return json.loads(urllib.request.urlopen(req, timeout=30).read().decode())

    import_mcpizy.py, line 38

    d = json.loads(urllib.request.urlopen(u, timeout=10).read().decode())

    mcpizy_autopublish.py, line 34

    return json.loads(urllib.request.urlopen(req, timeout=30).read().decode())

    Show more

    mcpizy_autopublish.py, line 46

    urllib.request.urlopen(req, timeout=20).read()

    monitor_run.py, line 23

    urllib.request.urlopen(f"https://api.telegram.org/bot{TG_TOKEN}/sendMessage", data=data, timeout=10)

    monitor_run.py, line 34

    urllib.request.urlopen(req, timeout=10)

    benchmark/outcomes.py, line 38

    r = urllib.request.urlopen(req, timeout=60)

    checkmcp/probe.py, line 39

    r = urllib.request.urlopen(req, timeout=timeout)

    checkmcp/probe.py, line 90

    self._resp = urllib.request.urlopen(urllib.request.Request(self.url, headers=h), timeout=self.timeout)

    checkmcp/probe.py, line 151

    urllib.request.urlopen(urllib.request.Request(self.endpoint, data=json.dumps(body).encode(), headers=h, method="POST"), timeout=timeout).read()

    web/app/policy/page.tsx, line 107

    <CopyButton text={`curl -H "X-CheckMCP-Key: <YOUR_KEY>" "https://checkmcp.dev/policy/check?url=https://mcp.example.com/mcp"`} />

    web/app/report/ReportClient.tsx, line 27

    const c = await fetch(`/api/eval-job?url=${encodeURIComponent(url)}`, { cache: "no-store", ...hdr });

    web/app/report/ReportClient.tsx, line 36

    const p = await fetch(`/api/eval-job?id=${encodeURIComponent(cj.id)}`, { cache: "no-store" });

    web/app/report/ReportClient.tsx, line 55

    const r = await fetch(`${endpoint}?url=${encodeURIComponent(url)}`, {

    Dockerfile, line 22

    CMD python -c "import urllib.request,sys; sys.exit(0 if urllib.request.urlopen('http://127.0.0.1:8080/health',timeout=3).status==200 else 1)"

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_DOTENV

    import_mcpizy.py, line 13

    SB_KEY = os.environ.get("MCPIZY_SUPABASE_KEY")

    import_mcpizy.py, line 14

    GH = os.environ.get("GITHUB_TOKEN")

    mcpizy_autopublish.py, line 13

    KEY = os.environ.get("MCPIZY_SUPABASE_KEY")

    Show more

    monitor_run.py, line 14

    TG_TOKEN = os.environ.get("TELEGRAM_BOT_TOKEN")

    server.py, line 44

    INTERNAL_SECRET = os.environ.get("CHECKMCP_INTERNAL_SECRET") # partagé avec le front pour les audits privés

    server.py, line 126

    GH_TOKEN = os.environ.get("GITHUB_TOKEN")

    benchmark/outcomes.py, line 21

    KEY = os.environ.get("CHECKMCP_LLM_KEY", "")

    benchmark/outcomes.py, line 141

    DP = os.environ.get("DROP_PILOT_MCP_TOKEN")

    .gitignore, line 8

    .next/

    .gitignore, line 9

    .env

    web/.gitignore, line 2

    .next/

    web/.gitignore, line 3

    .env.local

    web/README.md, line 12

    cp .env.example .env.local # remplir si besoin (URL + anon key déjà préremplis)

  • Agent instructions

    Serious

    instructions tell the agent to ignore earlier instructions.

    AGENT_IGNORE_INSTRUCTIONS

    web/lib/learn.ts, line 778

    "Tool poisoning is the headline risk: hidden, imperative instructions planted in tool metadata or outputs (\"ignore previous instructions\", \"do not tell the user\", \"also forwar

    web/lib/learn.ts, line 1247

    "On each tools/call response, it runs the same output analysis a behavioral eval uses: multilingual injection patterns (\"ignore previous instructions\", hidden \"do not tell the u

  • Obfuscation

    Note

    a file was over the read cap and was not fully read.

    OBFUSCATION_FILE_OVER_READ_CAP

    web/lib/learn.ts, line 1

    // Educational "learn" hub - plain-English guides to the MCP-security/quality concepts

  • Licence

    Pass

    A licence file is present (LICENSE).

  • Maintainer activity

    Pass

    The latest commit is within the last 365 days.

Get the full report on this tool, £49

Note

Hygiene check of commit 6d272b1cf568. requests shell access. The manifest names this permission. accepts a host from input, with no fixed list in this file. reads a credential or environment file. instructions tell the agent to ignore earlier instructions. a file was over the read cap and was not fully read.

Date and source

Date checked:

View source

github.com

Commit graded: 6d272b1cf5684c0e83afd8c5f01eb8765edf2e9a

This link goes to the original host. PlugVet does not host or check what you download from there, and installing it is your own decision and risk.

Read how this letter is chosen