Grades are a hygiene check, not a security guarantee.

MCP, MCP server

AgentAegis

Every cybersecurity service, callable by any AI agent, billed per use.

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.

Pattern found, check before installing

The checks matched a pattern that needs a closer look. It does not mean the code is harmful. Read the quoted lines below.

  • NETWORK_HOST_FROM_INPUT High

    host from input plus a credential path in operative code

NETWORK_HOST_FROM_INPUT OBFUSCATION_RUNS_ENCODED_PAYLOAD INSTALL_REMOTE_SCRIPT SECRETS_PATH_CREDENTIAL

Why this grade

  • NETWORK_HOST_FROM_INPUT High

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

    audit/run-tool.sh, line 19 INIT_RESP=$(curl -sS -X POST "$API/mcp" \

  • OBFUSCATION_RUNS_ENCODED_PAYLOAD High

    Decodes and runs an encoded payload (for example eval with atob or base64).

    pnpm-lock.yaml, line 1 lockfileVersion: '9.0'

  • INSTALL_REMOTE_SCRIPT Medium

    Fetches a script from the network and runs it (a pipe to a shell, or an install hook with a URL).

    docker/Dockerfile, line 21 RUN curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b /usr/local/bin

  • SECRETS_PATH_CREDENTIAL Medium

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

    src/tools/trustLayer/scanMcpPlugin.ts, line 51 const SECRET_READ_RE = /process\.env\b|os\.environ|\.ssh\/|\.aws\/credentials|id_rsa|\bAKIA[0-9A-Z]{16}\b|readFileSync?\([^)]*\.(env|pem|key)/i;

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

    src/utils/sandbox.ts, line 62

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

  • Install and update

    Serious

    install fetches a script from the network.

    INSTALL_REMOTE_SCRIPT

    docker/Dockerfile, line 21

    RUN curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b /usr/local/bin

    docker/Dockerfile.railway, line 40

    RUN curl -sSfL https://raw.githubusercontent.com/trufflesecurity/trufflehog/main/scripts/install.sh | sh -s -- -b /usr/local/bin

    docker/Dockerfile.railway, line 43

    RUN curl -sfL https://raw.githubusercontent.com/aquasecurity/trivy/main/contrib/install.sh | sh -s -- -b /usr/local/bin

  • Network

    Serious

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

    NETWORK_HOST_FROM_INPUT

    audit/run-tool.sh, line 19

    INIT_RESP=$(curl -sS -X POST "$API/mcp" \

    audit/run-tool.sh, line 35

    curl -sS -X POST "$API/mcp" \

    audit/run-tool.sh, line 53

    curl -sS -X POST "$API/mcp" \

    Show more

    audit/run-tool.sh, line 62

    curl -sS -X DELETE "$API/mcp" \

    audit/red-team/auto/a-x402-challenge-inspection.ts, line 60

    const initRes = await fetch(`${TARGET}/mcp`, {

    audit/red-team/auto/a-x402-challenge-inspection.ts, line 83

    await fetch(`${TARGET}/mcp`, {

    audit/red-team/auto/a-x402-challenge-inspection.ts, line 94

    const callRes = await fetch(`${TARGET}/mcp`, {

    audit/red-team/auto/a-x402-challenge-inspection.ts, line 120

    const initRes = await fetch(`${TARGET}/mcp`, {

    src/tools/account/help.ts, line 76

    example: `curl -X POST https://agentaegis-mcp-production.up.railway.app/v1/customers \\

    src/tools/account/help.ts, line 80

    curl -X POST https://agentaegis-mcp-production.up.railway.app/v1/customers/<id>/api-keys \\

    src/tools/account/help.ts, line 119

    example: `curl -X POST https://agentaegis-mcp-production.up.railway.app/v1/customers/<id>/checkout-session \\

    src/tools/trustLayer/scanMcpPlugin.ts, line 49

    // match between "(" and a quote, so /fetch\s*\(\b/ would miss fetch('url').

  • Secrets

    Concern

    reads a credential or environment file.

    SECRETS_ENV_READ SECRETS_PATH_CREDENTIAL SECRETS_PATH_DOTENV

    audit/red-team/auto/b1-balance-race.ts, line 28

    const API_KEY = process.env.RED_TEAM_API_KEY;

    audit/red-team/auto/b1-balance-race.ts, line 31

    console.error("Set RED_TEAM_API_KEY in .env.red-team. Use a TEST customer with $0.20 balance.");

    src/tools/trustLayer/scanMcpPlugin.ts, line 51

    const SECRET_READ_RE = /process\.env\b|os\.environ|\.ssh\/|\.aws\/credentials|id_rsa|\bAKIA[0-9A-Z]{16}\b|readFileSync?\([^)]*\.(env|pem|key)/i;

    Show more

    audit/red-team/auto/e1-e3-webhook-ssrf.ts, line 28

    const SUPABASE_ANON_KEY = process.env.SUPABASE_ANON_KEY;

    evals/tool-routing.mjs, line 72

    const API_KEY = process.env.ANTHROPIC_API_KEY || keyFromCredentialManager();

    src/server.ts, line 261

    includeCredentialCheck: !!process.env.HIBP_API_KEY,

    src/server.ts, line 308

    if (process.env.HIBP_API_KEY) {

    src/apis/abusech.ts, line 25

    return !!process.env.ABUSECH_API_KEY;

    src/apis/abusech.ts, line 29

    const key = process.env.ABUSECH_API_KEY;

    src/apis/abuseipdb.ts, line 63

    const apiKey = process.env.ABUSEIPDB_API_KEY;

    .gitignore, line 2

    dist/

    .gitignore, line 10

    audit/x402-test/.env*

    AGENTS.md, line 77

    - **Do not commit secrets.** `cdp_api_key.json`, `wallets-mainnet.json` and `.env` are gitignored and must stay that way. The admin token, API keys and the CDP key belong in Window

    README.md, line 19

    cp .env.example .env # add API keys

    ROADMAP.md, line 300

    - **Local dev `.env`**: `agentaegis-mcp/.env` (gitignored)

    audit/REPORT.md, line 143

    - **No exposed `.env` or `.git`** on either web property - verified via direct curl

  • Agent instructions

    Pass

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

  • Obfuscation

    Serious

    runs an encoded payload.

    OBFUSCATION_RUNS_ENCODED_PAYLOAD

    pnpm-lock.yaml, line 1

    lockfileVersion: '9.0'

    src/tools/trustLayer/scanMcpPlugin.ts, line 173

    if (obfExec.length) { hardBlock = true; reasons.push(`Obfuscated decode-then-execute (e.g. eval(atob(...))) in ${obfExec.length} place(s).`); }

  • 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 081ed56b268d. can delete files. install fetches a script from the network. accepts a host from input, with no fixed list in this file. reads a credential or environment file. runs an encoded payload.

Date and source

Date checked:

View source

github.com

Commit graded: 081ed56b268dcf99c32a3cc5d710b4f5d9fe2a94

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