Grades are a hygiene check, not a security guarantee.

Cursor, Plugin

ken

Thompson-mode systems discipline for AI agents.

Compatibility: Portable across standard formats.

Grade B

Minor issues only, no Medium or High. Worth a quick look at the findings below before installing.

Graded by PlugVetRubric v2

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

Rescanned at commit bee7f2cf3b48.

PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST SECRETS_PATH_DOTENV SECRETS_ENV_READ

Why this grade

  • PERMISSION_DELETE_FILES Low

    Contains a call that can delete files.

    benchmarks/agentic/run.py, line 150 return subprocess.run([shutil.which("git") or "git", *args], cwd=str(workdir),

  • NETWORK_NO_FIXED_HOST Low

    Makes a network call whose host is not written as a fixed address.

    benchmarks/agentic/judge.py, line 72 with urllib.request.urlopen(req, timeout=60) as r:

  • SECRETS_PATH_DOTENV Low

    Mentions a .env file.

    benchmarks/agentic/complete.py, line 20 Judge: claude-sonnet-4-6, key from ../../.env (shared with judge.py). ~$0.003/cell.

  • SECRETS_ENV_READ Note

    Reads an environment variable whose name contains KEY, SECRET, TOKEN, PASSWORD or CREDENTIAL.

    benchmarks/agentic/judge.py, line 48 return os.environ.get("ANTHROPIC_API_KEY")

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.

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

    Concern

    can delete files.

    PERMISSION_DELETE_FILES

    benchmarks/agentic/run.py, line 150

    return subprocess.run([shutil.which("git") or "git", *args], cwd=str(workdir),

    benchmarks/agentic/run.py, line 228

    subprocess.run(["taskkill", "/F", "/T", "/PID", str(proc.pid)],

    benchmarks/agentic/run.py, line 236

    p = subprocess.Popen([sys.executable, "-c", "import time; time.sleep(30)"],

    Show more

    benchmarks/agentic/run.py, line 321

    # open forever, so subprocess.run(timeout=) never fires and the worker freezes. Writing to a file

    benchmarks/agentic/run.py, line 326

    proc = subprocess.Popen(cmd, cwd=str(workdir), stdout=so, stderr=se,

    benchmarks/agentic/run.py, line 397

    try: return subprocess.run([shutil.which("claude"), "--version"], capture_output=True, text=True).stdout.strip()

    benchmarks/agentic/tasks.py, line 541

    proc = subprocess.Popen([node, str(srv)], cwd=str(workdir), env=env,

    benchmarks/maintenance/run.py, line 63

    subprocess.run(["taskkill", "/F", "/T", "/PID", str(proc.pid)],

    benchmarks/maintenance/run.py, line 174

    shutil.rmtree(ws)

    hooks/ken-runtime.js, line 38

    try { fs.unlinkSync(statePath); } catch (e) {}

    scripts/uninstall.js, line 16

    fs.unlinkSync(filePath);

  • Install and update

    Pass

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

  • Network

    Concern

    can reach a host that is not written as a fixed address.

    NETWORK_NO_FIXED_HOST

    benchmarks/agentic/judge.py, line 72

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

    benchmarks/agentic/tasks.py, line 538

    with urllib.request.urlopen(r, timeout=4) as resp:

    benchmarks/maintenance/tickets.py, line 363

    assert json.load(urllib.request.urlopen(req)) == {"balance": 1050}

    Show more

    benchmarks/maintenance/tickets.py, line 381

    last = json.load(urllib.request.urlopen(req))

    docs/platform-native.md, line 172

    | `requests` (simple GET) | `urllib.request.urlopen(url)`, `requests` for anything real |

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_DOTENV

    benchmarks/agentic/judge.py, line 48

    return os.environ.get("ANTHROPIC_API_KEY")

    benchmarks/agentic/complete.py, line 20

    Judge: claude-sonnet-4-6, key from ../../.env (shared with judge.py). ~$0.003/cell.

    benchmarks/agentic/complete.py, line 146

    if not key: sys.exit("no ANTHROPIC_API_KEY (.env or env)")

    Show more

    benchmarks/agentic/judge.py, line 13

    Judge: claude-sonnet-4-6 via the Anthropic Messages API (key from ../../.env). Scores the SOURCE

    benchmarks/agentic/judge.py, line 43

    for line in (ROOT / ".env").read_text(encoding="utf-8").splitlines():

    benchmarks/agentic/judge.py, line 177

    if not key: sys.exit("no ANTHROPIC_API_KEY (.env or env)")

    skills/ken-gain/SKILL.md, line 27

    (see benchmarks/README.md; needs API keys in .env) to produce the first

    .env.example, line 1

    # Copy to .env (gitignored) and fill in. promptfoo reads this automatically.

  • Agent instructions

    Pass

    No instruction to ignore earlier instructions or to hide a step.

  • Obfuscation

    Pass

    No encoded or minified payload found.

  • 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 bee7f2cf3b48. can delete files. can reach a host that is not written as a fixed address. reads a credential or environment file.

Date and source

Date checked:

View source

github.com

Commit graded: bee7f2cf3b48d08d1f91e2ec2d25909d3073164c

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