Grades are a hygiene check, not a security guarantee.

Claude Code, Skill

Skills

Twenty-eight research-backed specialists for the work between “I have an idea” and “it is live in production.” Each skill owns one job, carries its own restraints and verification contract, and loads only when it is relevant.

Compatibility: Not stated.

Grade C

One Medium finding. Read the findings 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 bf81e844e662.

PERMISSION_SHELL_UNDECLARED

Why this grade

  • PERMISSION_SHELL_UNDECLARED Medium

    Runs shell commands that the manifest or description does not mention.

    docs/upgrade/evidence/forward/run.py, line 5 p=subprocess.run(cmd,cwd=Path('/private/tmp/skills-forward')/slug,text=True,capture_output=True)

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

    requests shell access. No manifest field names shell or command execution.

    PERMISSION_SHELL_UNDECLARED

    docs/upgrade/evidence/forward/run.py, line 5

    p=subprocess.run(cmd,cwd=Path('/private/tmp/skills-forward')/slug,text=True,capture_output=True)

    docs/upgrade/evidence/intelligence-forward/observation-runner.py, line 22

    r = subprocess.run(args, cwd=cwd, text=True, capture_output=True)

    docs/upgrade/evidence/specialist-forward/execution-sources/finalize.py, line 37

    p=subprocess.run(cmd,cwd=ROOT,text=True,capture_output=True)

    Show more

    docs/upgrade/evidence/specialist-forward/execution-sources/presentation_tasks.py, line 116

    p=subprocess.run(command,shell=True,text=True,capture_output=True)

    docs/upgrade/evidence/specialist-forward/execution-sources/record.py, line 41

    p=subprocess.run(args,cwd=cwd or w/'work',text=True,input=input,capture_output=True)

    docs/upgrade/evidence/specialist-forward/frontpage/output/exercise_readme.py, line 6

    p=subprocess.run(command,shell=True,text=True,capture_output=True)

    scripts/check-claims, line 78

    context = json.loads(subprocess.check_output(

    scripts/check-demos, line 44

    result = subprocess.run([sys.executable, "-m", "unittest", "-v", test],

  • 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. can reach requests.readthedocs.io. The description does not name that host.

    NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST

    benchmarks/fixtures/ts-dashboard/src/App.tsx, line 66

    axios.post("/api/orders", payload);

    docs/upgrade/evidence/intelligence-forward/observation-runner.py, line 184

    - Separate top-level `requests.get()` calls create temporary Sessions (`src/requests/api.py:70-71`); do not assume they establish cross-call Session pooling.

    docs/upgrade/evidence/intelligence-forward/masterplan/snapshot/src/App.tsx, line 66

    axios.post("/api/orders", payload);

    Show more

    docs/upgrade/evidence/specialist-forward/blueprint/input/src/App.tsx, line 66

    axios.post("/api/orders", payload);

    docs/upgrade/evidence/specialist-forward/blueprint/output/src/App.tsx, line 66

    axios.post("/api/orders", payload);

    docs/upgrade/evidence/intelligence-forward/masterplan/commands.jsonl, line 4

    {"argv": ["cat", "src/App.tsx"], "cwd": "/private/tmp/skills-forward/intelligence/masterplan", "exit_code": 0, "stdout": "// TaskBoard app \u2014 main component\nimport React, { us

    docs/upgrade/evidence/intelligence-forward/scout-web-raw.json, line 1

    "Advanced Usage — Requests 2.34.2 documentation (https://requests.readthedocs.io/en/latest/user/advanced/)\nciteturn33view0 [wordlim: 200] Crawled: today; Content type: text/htm

    docs/upgrade/evidence/intelligence-forward/scout/artifact.md, line 22

    - Separate top-level `requests.get()` calls create temporary Sessions (`src/requests/api.py:70-71`); do not assume they establish cross-call Session pooling.

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_DOTENV

    docs/upgrade/evidence/forward/harden/notes.py, line 23

    secret = os.environ.get("NOTES_SECRET")

    benchmarks/fixtures/janitor-mess/setup.sh, line 12

    printf 'NOTES_KEY=FIXTURE-FAKE-KEY-do-not-use\n' > .env

    benchmarks/fixtures/janitor-mess/setup.sh, line 13

    git add .env

    Show more

    CHANGELOG.md, line 173

    `pilot` **NO LIFT on available evidence** (`.env` read auto-rejection

    CHANGELOG.md, line 180

    path-existence grading, preserves leading-dot filenames (`.env`,

    CHANGELOG.md, line 624

    caught and fixed two harness bugs — bare `.env`-style tokens counted

    benchmarks/answer-keys/ts-dashboard.json, line 5

    {"id": "S1", "severity": "critical", "category": "secrets", "location": "src/App.tsx API_KEY + .env tracked with example placeholder keys (intentionally committed); .gitignore omit

    docs/benchmarks/janitor.md, line 41

    1. Bare dotfile tokens (`.env`) counted as hallucinated paths, so the

    docs/benchmarks/janitor.md, line 42

    required finding about the committed `.env` false-failed. Fixed:

  • Agent instructions

    Pass

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

  • Obfuscation

    Concern

    contains a minified payload.

    OBFUSCATION_MINIFIED_PAYLOAD

    docs/upgrade/evidence/specialist-forward/execution-sources/analytical.py, line 50

    (w/'model.py').write_text('import json\np=json.load(open("capacity-case.json"))\nr,e=p["workloads"]\nprint(json.dumps({"basis":"synthetic assumptions, not measured capacity", "read

    docs/upgrade/evidence/specialist-forward/execution-sources/final_receipt_corrections.py, line 1

    from record import *

  • Licence

    Pass

    A licence file is present (LICENSE).

  • Maintainer activity

    Pass

    The latest commit is within the last 365 days.

Note

Hygiene check of commit bf81e844e662. requests shell access. No manifest field names shell or command execution. can reach a host that is not written as a fixed address. can reach requests.readthedocs.io. The description does not name that host. reads a credential or environment file. contains a minified payload.

Date and source

Date checked:

View source

github.com

Commit graded: bf81e844e6623ab27e4ed7022a7ba9a6f8dda527

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