Grades are a hygiene check, not a security guarantee.

Claude Code, Skill

Ai Tooling

LaunchDarkly agent skills and MCP servers for feature flag management, AgentControl, and metrics

Compatibility: Not stated.

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.

PERMISSION_DELETE_FILES NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST SECRETS_PATH_DOTENV SECRETS_ENV_READ OBFUSCATION_FILE_OVER_READ_CAP

Why this grade

  • PERMISSION_DELETE_FILES Low

    Contains a call that can delete files.

    evals/providers/claude-skill-agent-sdk.js, line 495 fs.rmSync(cwd, { recursive: true, force: true });

  • NETWORK_NO_FIXED_HOST Low

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

    skills/agentcontrol/built-in-metrics/references/langchain-tracking.md, line 36 curl -X PATCH \

  • NETWORK_UNNAMED_HOST Low

    Calls a fixed host that the description does not name.

    skills/agentcontrol/built-in-metrics/references/langchain-tracking.md, line 36 curl -X PATCH \

  • SECRETS_PATH_DOTENV Low

    Mentions a .env file.

    evals/scripts/aggregate.js, line 70 ...(fs.existsSync(path.join(EVALS_DIR, ".env")) ? ["--env-file", path.join(EVALS_DIR, ".env")] : []),

  • SECRETS_ENV_READ Note

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

    evals/providers/claude-skill-agent-sdk.js, line 198 const apiKey = process.env.ANTHROPIC_API_KEY;

  • OBFUSCATION_FILE_OVER_READ_CAP Note

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

    evals/package-lock.json, line 1 {

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

    evals/providers/claude-skill-agent-sdk.js, line 495

    fs.rmSync(cwd, { recursive: true, force: true });

  • 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 app.launchdarkly.com. The description does not name that host. can reach api.tavily.com. The description does not name that host.

    NETWORK_NO_FIXED_HOST NETWORK_UNNAMED_HOST

    skills/agentcontrol/built-in-metrics/references/langchain-tracking.md, line 36

    curl -X PATCH \

    skills/agentcontrol/built-in-metrics/references/metrics-api.md, line 48

    response = requests.get(url, headers=headers, params=params)

    skills/agentcontrol/configs-create/SKILL.md, line 166

    curl -s "https://app.launchdarkly.com/api/v2/projects/$PROJECT/ai-configs/$CONFIG_KEY/targeting?env=$ENV" \

    Show more

    skills/agentcontrol/configs-create/SKILL.md, line 173

    curl -X PATCH "https://app.launchdarkly.com/api/v2/projects/$PROJECT/ai-configs/$CONFIG_KEY/targeting?env=$ENV" \

    skills/agentcontrol/configs-targeting/SKILL.md, line 58

    curl -X GET "https://app.launchdarkly.com/api/v2/projects/{projectKey}/ai-configs/{configKey}/targeting" \

    skills/agentcontrol/configs-targeting/SKILL.md, line 74

    curl -X PATCH "https://app.launchdarkly.com/api/v2/projects/{projectKey}/ai-configs/{configKey}/targeting" \

    skills/agentcontrol/configs-targeting/SKILL.md, line 92

    curl -X PATCH "https://app.launchdarkly.com/api/v2/projects/{projectKey}/ai-configs/{configKey}/targeting" \

    skills/agentcontrol/configs-targeting/SKILL.md, line 115

    curl -X PATCH "..." \

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_DOTENV

    evals/providers/claude-skill-agent-sdk.js, line 198

    const apiKey = process.env.ANTHROPIC_API_KEY;

    evals/scripts/aggregate.js, line 70

    ...(fs.existsSync(path.join(EVALS_DIR, ".env")) ? ["--env-file", path.join(EVALS_DIR, ".env")] : []),

    evals/scripts/run-models.js, line 8

    * still pass on Haiku, or did I overfit to Sonnet?" - without juggling .env edits.

    Show more

    evals/scripts/run-models.js, line 108

    "--env-file", ".env",

    skills/agentcontrol/built-in-metrics/references/gemini-tracking.md, line 94

    const genAI = new GoogleGenAI({ apiKey: process.env.GEMINI_API_KEY! });

    skills/agentcontrol/built-in-metrics/references/openai-tracking.md, line 40

    const ldClient = init(process.env.LD_SDK_KEY!);

    skills/agentcontrol/built-in-metrics/references/openai-tracking.md, line 108

    const client = new OpenAI({ apiKey: process.env.OPENAI_API_KEY });

    skills/agentcontrol/custom-metrics/SKILL.md, line 333

    SDK_KEY = os.environ.get("LAUNCHDARKLY_SDK_KEY")

    skills/agentcontrol/migrate/SKILL.md, line 176

    sdk_key = os.environ.get("LD_SDK_KEY")

    skills/agentcontrol/migrate/SKILL.md, line 200

    if (!process.env.LD_SDK_KEY) {

    skills/agentcontrol/migrate/SKILL.md, line 203

    const ldClient = init(process.env.LD_SDK_KEY ?? 'sdk-offline');

    skills/agentcontrol/projects/references/admin-tooling.md, line 39

    @click.option('--save-env/--no-save-env', default=True, help='Save SDK keys to .env')

    skills/agentcontrol/projects/references/admin-tooling.md, line 50

    click.echo(f"✓ Saved SDK keys to .env")

    skills/agentcontrol/projects/references/env-config.md, line 20

    env_file: str = ".env",

    skills/agentcontrol/projects/references/env-config.md, line 23

    """Save SDK key to .env file."""

    skills/agentcontrol/projects/references/env-config.md, line 31

    # Read existing .env content

  • 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.

    OBFUSCATION_FILE_OVER_READ_CAP

    evals/package-lock.json, line 1

    {

    tests/package-lock.json, line 1

    {

  • 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 01bc9468b886. can delete files. can reach a host that is not written as a fixed address. can reach app.launchdarkly.com. The description does not name that host. can reach api.tavily.com. The description does not name that host. reads a credential or environment file. a file was over the read cap and was not fully read.

Date and source

Date checked:

View source

github.com

Commit graded: 01bc9468b8860b91f2ca340047a83e5a418340fe

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