Grades are a hygiene check, not a security guarantee.

Claude Code, Skill

skillEd

Test your agent’s Skills.

Compatibility: Not stated.

Grade D

Two or more Medium findings, or a High. 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.

NETWORK_HOST_FROM_INPUT SECRETS_PATH_CREDENTIAL

Why this grade

  • NETWORK_HOST_FROM_INPUT Medium

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

    install.sh, line 32 requested_version="$(curl -fsSL --retry 3 "$SKILLED_LATEST_VERSION_URL")"

  • SECRETS_PATH_CREDENTIAL Medium

    Mentions a credential store path (~/.ssh, id_rsa, .npmrc, .aws/credentials, credentials.json, .netrc).

    skills/skilled-improve/evals/test_regression_scripts.py, line 380 for command in ('grep token ~/.codex/auth.json', 'rg token ~/.claude/.credentials.json',

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

    Concern

    can delete files.

    PERMISSION_DELETE_FILES

    skills/skilled-improve/evals/test_regression_scripts.py, line 32

    result = subprocess.run(

    skills/skilled-improve/evals/test_regression_scripts.py, line 372

    result = subprocess.run(['bash', str(judge)], cwd=workspace,

    skills/skilled-improve/evals/test_regression_scripts.py, line 533

    and in ("python3 -c 'import subprocess; subprocess.run([\"skilled\",\"run\",\"evals/eval.yaml\

    Show more

    skills/skilled-improve/evals/fixtures/scripts/assert-credential-presence.sh, line 601

    result = subprocess.run(argv, cwd=root, env=env, text=True, capture_output=True, timeout=10)

    skills/skilled-improve/evals/fixtures/scripts/assert-localized-regressions.sh, line 310

    validation = subprocess.run(["skilled", "validate", "evals/eval.yaml"], text=True, capture_output=True, timeout=20)

    skills/skilled-improve/evals/fixtures/scripts/assert-localized-regressions.sh, line 312

    result = subprocess.run(["skilled", "list-cases", "evals/eval.yaml"], text=True, errors="replace", capture_output=True, timeout=20)

    src/agent/codex.ts, line 439

    const actualTurn = state.commands.get(item.id ?? "") ?? turn;

    src/evaluator/artifacts-collect.ts, line 71

    await fs.rm(root, { recursive: true, force: true });

    src/evaluator/artifacts.ts, line 117

    await fs.rm(temporaryPath, { recursive: true, force: true });

    src/evaluator/artifacts.ts, line 202

    await fs.rm(state.temporaryPath, { recursive: true, force: true });

    src/evaluator/artifacts.ts, line 239

    await fs.rm(target, { recursive: true, force: true });

    src/evaluator/artifacts.ts, line 308

    await fs.rm(dir, { recursive: true, force: true });

    src/evaluator/artifacts.ts, line 312

    await fs.rm(dir, { recursive: true, force: true });

    src/evaluator/fixtures.ts, line 135

    await fs.rm(tmp, { recursive: true, force: true });

    src/runtime/docker.ts, line 198

    await fs.rm(root, { recursive: true, force: true });

    assets/completion/powershell-no-descriptions.txt, line 92

    Invoke-Expression -OutVariable out "$RequestComp" 2>&1 | Out-Null

  • Install and update

    Note

    README documents an installer that pipes a download into a shell. The docs document an installer that pipes a download into a shell.

    INSTALL_DOCUMENTED_INSTALLER

    README.md, line 26

    curl -fsSL https://github.com/DeadHead04/skillEd/releases/latest/download/install.sh | bash

    instruction.md, line 14

    curl -fsSL https://github.com/DeadHead04/skillEd/releases/latest/download/install.sh | bash

    docs/index.md, line 56

    curl -fsSL https://github.com/DeadHead04/skillEd/releases/latest/download/install.sh | bash

    Show more

    docs/guide/installation.md, line 16

    curl -fsSL https://github.com/DeadHead04/skillEd/releases/latest/download/install.sh | bash

    skills/skilled-improve/references/install.md, line 10

    curl -fsSL https://raw.githubusercontent.com/DeadHead04/skillEd/main/install.sh | bash

  • Network

    Serious

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

    NETWORK_HOST_FROM_INPUT

    install.sh, line 32

    requested_version="$(curl -fsSL --retry 3 "$SKILLED_LATEST_VERSION_URL")"

    install.sh, line 34

    latest_url="$(curl -fsSLI --retry 3 -o /dev/null -w '%{url_effective}' "https://github.com/$repository/releases/latest")"

    install.sh, line 37

    requested_version="$(curl -fsSL --retry 3 "${release_base%/}/latest-version.txt")"

    Show more

    install.sh, line 57

    curl -fL --retry 3 "$base_url/$archive" -o "$tmp_dir/$archive"

    install.sh, line 58

    curl -fL --retry 3 "$base_url/$archive.sha256" -o "$tmp_dir/checksums.txt"

    action/main.ts, line 395

    const response = await fetch(url, { signal: AbortSignal.timeout(120000) });

    scripts/package.mjs, line 34

    const sumsResponse = await fetch(`${base}SHASUMS256.txt`);

    scripts/package.mjs, line 55

    const response = await fetch(base + nodeFile);

    scripts/services/opensandbox-local.mjs, line 124

    const res = await fetch(environment.OPENSANDBOX_BASE_URL + "/health", {

    src/agent/custom.ts, line 641

    const response = await fetch(f.url!, {

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV

    skills/skilled-improve/evals/test_regression_scripts.py, line 221

    for source in ('import os; open("credential-check.sh", "w").write(os.getenv("OPENAI_API_KEY"))',

    skills/skilled-improve/evals/test_regression_scripts.py, line 222

    'from pathlib import Path; import os; Path("credential-check.sh").write_text("prefix" + os.getenv("OPENAI_API_KEY"))',

    skills/skilled-improve/evals/test_regression_scripts.py, line 380

    for command in ('grep token ~/.codex/auth.json', 'rg token ~/.claude/.credentials.json',

    Show more

    skills/skilled-improve/evals/test_regression_scripts.py, line 381

    'awk \'{print}\' ~/.skilled/credentials.yaml', 'grep key .env', 'cat ~/.codex/auth.json>/tmp/auth-copy', 'grep token<~/.codex/auth.json'):

    skills/skilled-improve/evals/test_regression_scripts.py, line 474

    'from pathlib import Path; import os; Path("credential-check.sh").write_text(os.getenv("OPENAI_API_KEY"))',

    skills/skilled-improve/evals/fixtures/scripts/assert-credential-presence.sh, line 341

    assert not re.search(r"(?:^|/)(?:\.?credentials\.(?:yaml|yml|json)|auth\.json|oauth_creds\.json|\.env)$", target), "Author read a credential file"

    skills/skilled-improve/evals/fixtures/scripts/assert-credential-presence.sh, line 400

    protected_names = {"credentials.yaml", "credentials.yml", "credentials.json", ".credentials.yaml", ".credentials.yml", ".credentials.json", "auth.json", "oauth_creds.json", ".env"}

    skills/skilled-improve/evals/fixtures/scripts/assert-credential-presence.sh, line 557

    assert not re.search(r"(?:^|[/\s\"'])(?:\.?credentials\.(?:yaml|yml|json)|auth\.json|oauth_creds\.json|\.env)(?=$|[\s\"';|&<>()])", command), "Author accessed a credential file"

    src/agent/custom.ts, line 669

    for (const secret of Object.values(this.cfg.custom?.env ?? {}))

    src/credentials/index.ts, line 108

    options.dotenvPath ?? path.resolve(".env"),

    src/credentials/index.ts, line 117

    this.warn(`Failed to load .env file: ${(e as Error).message}`);

    src/runtime/opensandbox.ts, line 121

    const apiKey = process.env.OPENSANDBOX_API_KEY;

  • Agent instructions

    Pass

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

  • Obfuscation

    Note

    a file was over the read cap and was not fully read. the file cap was reached before every file was read.

    OBFUSCATION_FILE_OVER_READ_CAP OBFUSCATION_FILE_CAP_REACHED

    tests/config/validator-reference.json, line 1

    [

    tests/integration/phase4-helper-reference.json, line 1

    {

    tests/judge/scoring-contract.test.ts, line 1

    import test from "node:test";

  • 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 c92850f89aad. can delete files. README documents an installer that pipes a download into a shell. The docs document an installer that pipes a download into a shell. accepts a host from input, with no fixed list in this file. reads a credential or environment file. a file was over the read cap and was not fully read. the file cap was reached before every file was read.

Date and source

Date checked:

View source

github.com

Commit graded: c92850f89aad0466b061de5be0a653fa05536b14

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