Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    code security vulnerability scanner

    1

    Deterministic SAST scanner for OWASP Top 10, SQL injection, and hardcoded secrets with CVSS scoring.

    Free

    Security scanned
    code security vulnerability scanner

    code security vulnerability scanner

    Example session with this skill installed

    Scan this for vulnerabilities

    const user = req.body.username;
    const query = "SELECT * FROM accounts WHERE user = '" + user + "'";
    const secret = 'sk_live_51MzBy8L2vQ9x1';
    db.execute(query);
    
    • Read your context and instructions
    • Compiled the code security vulnerability

    I have completed the security audit of the provided source code. Two CRITICAL vulnerabilities were identified, resulting in a blocked deployment gate.

    Security Audit Report

    Audit Hash: e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
    Security Posture Score: 50/100 (C_AT_RISK)
    Deploy Gate: BLOCKED_CRITICAL_VULNERABILITIES

    LineIDCWESeverityCategoryDescription
    2INJ-SQL-001CWE-899.8 (CRITICAL)OWASP A03:2021SQL Injection via string concatenation in query construction.
    3SEC-KEY-001CWE-7989.1 (CRITICAL)OWASP A07:2021Hardcoded Stripe API Secret Key detected.

    Remediation Patches

    // Fix for Line 2: Use parameterized queries
    const query = "SELECT * FROM accounts WHERE user = ?";
    db.execute(query, [user]);
    
    // Fix for Line 3: Move secrets to environment variables
    const secret = process.env.STRIPE_SECRET_KEY;
    

    Next steps

    1. Replace string concatenation with parameterized queries or an ORM to prevent SQL injection.
    2. Revoke the exposed Stripe API key immediately and migrate to an environment variable manager.
    3. Re-run the scan after applying patches to clear the deployment gate.

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

    About this skill

    Manual security audits are slow, inconsistent, and prone to human oversight. This skill provides a deterministic static security scanner (SAST) for AI agents to audit code for vulnerabilities before they hit production.

    What it does

    • OWASP Auditing scans for the Top 10 flaws including SQL injection, XSS, SSRF, and prototype pollution.
    • Secret Detection identifies hardcoded API keys, PEM keys, and database URIs using high-precision regex.
    • Severity Scoring calculates a Security Posture Score using CVSS 3.1 metrics and CWE identifiers.
    • Remediation Patching generates specific code fixes for every detected vulnerability.
    • Deployment Gating blocks releases automatically when HIGH or CRITICAL severity findings are present.

    How it works

    1. Pass Code by providing the source code snippet or file contents to the source_code argument.
    2. Set Context by specifying the language and the environment_tier to adjust scanning strictness.
    3. Analyze Results by reviewing the line-by-line findings, CVSS scores, and the generated audit hash.
    4. Apply Patches using the provided remediation blocks to resolve identified CWE exposures.

    Frameworks & tools

    This skill is built for TypeScript, JavaScript, Python, SQL, Go, and Java. It maps findings to OWASP Top 10 categories and CWE standards using CVSS 3.1 scoring logic.

    Why this beats prompting it yourself

    Generic prompts often miss subtle injection vectors or fail to provide consistent severity scores. This skill uses a deterministic formula to calculate a 0-100 Security Posture Score, ensuring your AI auditor applies the same strictness to every pull request.

    Use cases

    • Reviewing AI-generated backend routes for injection flaws.
    • Scanning legacy codebases for hardcoded credentials.
    • Enforcing security gates in automated CI/CD pipelines.
    • Auditing third-party SQL scripts before execution.

    Known limitations

    Does not perform dynamic penetration testing against live URLs. Performance is limited to static analysis of the provided text.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.

    Static analysis only; no dynamic execution or live pen-testing. Requires source code to be passed as a string. Audit patches require human review before deployment.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Free to use

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 day ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables
    Listed1 day ago

    What's inside

    Frequently Asked Questions

    Unlike generic LLM prompts that can be inconsistent, this skill uses a deterministic SAST engine to provide standardized CVSS 3.1 severity scores and specific CWE mappings for every finding.

    The skill supports static analysis for TypeScript, JavaScript, Python, SQL, Go, and Java, making it ideal for auditing both modern web apps and backend database scripts.

    Yes, the skill includes a 'Remediation Patching' feature that not only identifies the flaw but also generates the specific code fix needed to resolve the vulnerability.

    By using the `environment_tier` parameter, you can configure the skill to act as a deployment gate that automatically flags or blocks releases containing HIGH or CRITICAL severity findings.

    The scanner is built to be platform-agnostic; it can be integrated into any AI agent or workflow that supports tool-calling and can pass source code as a text argument.

    New in Security

    Free