Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Sourcemap & Env Leak Check

    1

    Audit websites for exposed source maps, .env files, Git metadata, and missing security headers.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Sourcemap & Env Leak Check

    Sourcemap & Env Leak Check

    Example session with this skill installed

    Run a leak check on https://dev.example.com and tell me what needs fixing. I own this domain.

    • Read your context and instructions
    • Compiled the sourcemap & env

    I have completed the passive exposure audit for https://dev.example.com. The scan identified 2 high-severity incidents and 3 missing security headers.

    Exposure check: https://dev.example.com (2023-10-27)

    Incidents / fix now

    [high] https://dev.example.com/.env — reachable (HTTP 200) —

    fix: Block access to all dotfiles in Nginx/Apache config and rotate all credentials (DB_PASSWORD, API_KEYS) immediately.

    [high] https://dev.example.com/static/js/main.d82a.js.map — sourcesContent embedded — 142 sources public —

    fix: Disable source map generation in your production build pipeline (e.g., GENERATE_SOURCEMAP=false).

    Should fix

    [medium] https://dev.example.com/.DS_Store — reachable (HTTP 200) —

    fix: Add to .gitignore and run find . -name .DS_Store -delete on the server.

    [medium] Missing Content-Security-Policy —

    fix: Define a CSP header to mitigate XSS and data injection risks.

    Info

    • [info] Missing Strict-Transport-Security — fix: Enable HSTS header to enforce HTTPS connections.
    • [info] Missing X-Frame-Options — fix: Set to SAMEORIGIN to prevent clickjacking.

    Not covered

    • https://analytics.external.com/script.js — skipped (cross-origin).
    • /admin/config.bak — blocked (HTTP 403).

    Full report: leak-report.json

    Next steps

    1. Review leak-report.json for the full list of 50 probed paths.
    2. Confirm if /debug (HTTP 200) is a real endpoint or a catch-all redirect.
    3. Update your CI/CD pipeline to run this script on every staging deploy.

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

    About this skill

    This skill identifies passive security exposures on websites by detecting publicly reachable source maps, sensitive configuration files, and missing security headers. It operates as a read-only auditor, sending standard HTTP requests to identify high-risk assets like .env files, .git directories, and backup scripts that should never be public. The tool prioritizes findings by severity, distinguishing between metadata leaks and full source code exposure.

    What it does

    • Detects exposed source maps by probing for .map files and identifying embedded source code that reveals original logic and comments.
    • Identifies sensitive paths including .env, /.git/HEAD, /wp-config.php~, and database backups.
    • Audits security headers to flag the absence of Content-Security-Policy, HSTS, X-Frame-Options, and Referrer-Policy.
    • Triages information disclosure by finding metadata files like .DS_Store or exposed status endpoints like /server-status.
    • Differentiates catch-all routes to reduce false positives where a server returns a 200 status code for every path.

    How it works

    1. Initialize the audit by running the bundled Node.js script against a specific URL you are authorized to test.
    2. Review the JSON report generated by the tool, which categorizes findings into high, medium, and info severity levels.
    3. Analyze incidents starting with high-priority leaks like configuration files or source maps with embedded content.
    4. Apply remediation steps using the provided prioritized fix list, tailored to the specific server or framework in use.

    Frameworks & tools

    This skill requires Node.js (version 18.17 or higher) for its global fetch capability. It is a zero-dependency script that works against any web origin regardless of the underlying backend language or frontend framework.

    Why this beats prompting it yourself

    Manual prompting cannot reliably probe network endpoints or parse JSON reports for specific security flags like hasSourcesContent. This skill provides a repeatable, scriptable process that accurately distinguishes between a valid file exposure and a generic 404-redirect, ensuring you don't miss hidden source maps that a browser wouldn't normally show.

    Use cases

    • Auditing a production site after a new deployment to ensure build artifacts were cleaned up.
    • Identifying leaked credentials in .env files before they are exploited by automated scanners.
    • Verifying that sensitive Git metadata is not accessible via the public web server.
    • Checking that security headers meet basic hardening requirements for compliance.

    Known limitations

    This is a passive scanner and not a full penetration test. It only checks specific known paths and does not perform brute-force attacks or authenticate to private areas.

    Requires Node >=18.17. Passive only; no brute-force or auth bypass. Does not audit third-party scripts. Findings limited to specific probed paths.

    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 any payment.

    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

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 21 days ago

    Requires

    Node.js 18.17+

    Permissions

    Terminal / Shell
    Write Files
    Network Access

    Passive probes against the target origin only — HTTP status and metadata, never file contents. Third-party script URLs are skipped, never fetched.

    Listed1 month ago
    Updated27 days ago

    What's inside

    Frequently Asked Questions

    This skill scans your web application for high-risk files that developers often forget to delete, such as .env files containing credentials, Git metadata that reveals code history, and source maps that expose your original source code to competitors or attackers.

    You will receive the audited Node.js script, a README with deployment instructions, and a prioritized remediation guide to help you fix any vulnerabilities found during the scan.

    The skill requires Node.js 18.17 or higher and is designed to run in any environment that supports a standard Node runtime, making it compatible with most AI agents that have CLI or script-execution capabilities.

    No, this is a passive auditor that only sends standard HTTP requests to common sensitive paths; it does not perform invasive brute-force attacks or attempt to bypass authentication, ensuring a safe scan for production environments.

    Unlike a general prompt, this skill uses a specialized script to handle complex network logic, such as distinguishing between real file exposures and 'false positive' 200 OK responses from servers that redirect all 404 errors to a homepage.

    The script is a zero-dependency Node.js tool, meaning you can run it immediately without installing external packages—simply provide the target URL to the AI agent to begin the audit.

    New in Security