MCP-Security-Review

    2

    Specialized static security scanner for MCP servers and Python tool handlers to prevent injection and data leaks.

    $5

    Secure checkout via Stripe

    30-day refund guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI tools you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    MCP-Security-Review

    Example session with this skill installed

    Run a security audit on my new sqlite-mcp-server to make sure there are no injection or path traversal vulnerabilities.

    • Read your context and instructions
    • Compiled the mcp-security-review

    [HIGH] SQL Injection Found

    • File: server.py:42
    • Code: db.execute(f"SELECT * FROM logs WHERE id = {tool_input}")
    • Fix: Use parameterized queries: db.execute("SELECT * FROM logs WHERE id = ?", (tool_input,))

    [MED] Path Traversal Risk

    • File: files.py:12
    • Manual Check: Ensure base_path is sanitized.

    Connects securely to your tools. The creator never sees your data.

    What you get

    Audit MCP servers before publishing to public registriesDetect command injection risks in AI tool handlersIdentify SSRF vulnerabilities in tools that fetch external URLsFind hardcoded secrets or sensitive data exposure in server logs

    About this skill

    What it does

    This skill provides a specialized security audit for Model Context Protocol (MCP) servers and Python-based tool servers. It performs a targeted static analysis focused specifically on the "input-to-sink" path—where AI-generated or user-controlled input meets dangerous system functions.

    Why use this skill

    AI tools are uniquely vulnerable because they often bridge the gap between natural language and local system execution. Standard scanners often miss the nuances of MCP tool handlers. This skill prioritizes findings in tool-definition files and identifies high-risk patterns like command injection, SSRF, and unsafe deserialization that could lead to full system compromise if an agent is manipulated.

    Supported checks

    • Injection: Command, Code, SQL, and Server-Side Template Injection.
    • Data Safety: Unsafe deserialization (pickle/yaml) and path traversal.
    • Network: SSRF via outbound HTTP calls to input-derived URLs.
    • Secrets: Scanning for hardcoded credentials or accidental logging of sensitive data.

    The output provides a prioritized list of vulnerabilities with exact file paths, line numbers, and actionable remediation steps, followed by a custom manual review checklist for the tool's architecture.

    How to install

    Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.

    ~30 seconds
    1. 1

      Download the ZIP

      Free skills download straight away. Paid skills unlock right after purchase.

    2. 2

      Unzip into your skills folder

      Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.

    3. 3

      Ask your agent to use it

      Restart the agent if it was already running. It picks the skill up automatically - no config needed.

    Skills folder by agent

    Click the path to copy it. Create the folder if it does not exist yet.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill passes our trust checks below.

    Security scanned

    Passed our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    30-day refund

    Not a fit? Get your money back

    Trust & safety

    Security scanned

    Verified clean 3 months ago

    • Passed all security checks, Safe to install

    Listed3 months ago

    What's inside

    Frequently Asked Questions