Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    legacy stack vuln audit

    3

    Defensive vulnerability audit for PHP and COBOL stacks using live CISA KEV and CVE threat data.

    Free

    6 uses4.0 (1 review)
    legacy stack vuln audit

    legacy stack vuln audit

    Example session with this skill installed

    Audit this PHP 7.4 server running on Ubuntu 20.04. Check against CISA KEV and find EOL components. I am authorized to scan this host at 192.168.1.50.

    • Read your context and instructions
    • Compiled the legacy stack vuln

    I have completed the vulnerability audit for the PHP environment on host 192.168.1.50. Live data was retrieved from CISA KEV, NVD, and Ubuntu Security trackers on 2026-11-05.

    Legacy Stack Vulnerability Audit

    IDComponentVersion FoundStatusCVSSKEVFix / MitigationPriority
    CVE-2024-XXXXPHP-FPM7.4.3Confirmed9.8YesUpgrade to PHP 8.2+; Apply WAF rule1
    EOL-PHP-7.4PHP Core7.4.3ConfirmedN/ANoMigrate to supported PHP 8.x branch1
    MISC-01php.iniexpose_php=OnConfirmed5.3NoSet expose_php = Off3
    COMP-04Guzzle (Composer)6.5.5Likely7.5NoUpdate to Guzzle 7.4.5+2

    Executive Summary

    • Critical Findings: 2 (1 in CISA KEV)
    • High Findings: 1
    • EOL Components: 1 (PHP 7.4 reached EOL on 2022-11-28)
    • Authorization: Confirmed by user for IP 192.168.1.50

    Remediation Plan

    Immediate (Critical): The PHP 7.4 branch is end-of-life and contains a KEV-listed vulnerability. Because Ubuntu 20.04 backports are limited, migrate to a supported PHP 8.x container or instance immediately.
    2.

    Configuration: Disable expose_php and display_errors in /etc/php/7.4/fpm/php.ini to prevent version leakage and path disclosure during migration.
    3.

    Dependencies: Run composer update to resolve known vulnerabilities in the Guzzle library and other vendor packages.

    Next steps

    • Verify if the PHP-FPM process was restarted after the last security patch.
    • Review web server logs for indicators of compromise related to the KEV-listed bug.
    • Schedule a full migration to a supported PHP 8.2+ environment.

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

    About this skill

    Legacy environments often harbor unpatched vulnerabilities that standard scanners miss due to version backporting or end-of-life (EOL) status. This skill provides a defensive, read-only audit of PHP and COBOL stacks against live threat intelligence, including CISA KEV and vendor-specific advisories. ## What it does - **Live vulnerability matching** by querying CISA KEV, NVD CVE, and vendor-specific security trackers for PHP, IBM, and OpenText. - **Environment inventory** across PHP versions, Composer dependencies, COBOL runtimes, and mainframe middleware like CICS or MQ. - **EOL status tracking** to identify components no longer receiving security updates, even if no specific CVE is currently flagged. - **Configuration hardening** checks for `php.ini` directives and mainframe resource security gaps in RACF, ACF2, or Top Secret. - **Static code review** for high-risk patterns like `unserialize()` in PHP or missing bounds checks and dynamic SQL in COBOL. - **Prioritized remediation** that ranks findings based on reachability, KEV status, and exploitability. ## How it works 1. **Scope verification**: Confirms authorization and defines the target boundaries including LPARs, repositories, or paths. 2. **Inventory collection**: Analyzes server outputs, config files, and dependency locks to build a component map. 3. **Threat intelligence pull**: Queries live APIs for the most recent disclosures and exploited-in-the-wild bugs. 4. **Verification and matching**: Compares inventory against advisories while accounting for OS-level backports and running process memory. 5. **Report generation**: Delivers a structured audit with severity counts, KEV status, and a phased remediation plan. ## Frameworks & tools PHP (7.x, 8.x), Composer, WordPress, Laravel, Symfony, COBOL (IBM Enterprise, GnuCOBOL, Micro Focus), z/OS, CICS, Db2, JCL, and Zowe. ## Why this beats prompting it yourself Generic prompts often rely on stale training data, missing critical zero-days or recent CISA KEV additions. This skill enforces a structured verification workflow that distinguishes between disk-level patches and vulnerable code still running in memory, preventing false sense of security. ## Use cases - Auditing a legacy PHP server for vulnerabilities listed in the current CISA KEV catalog. - Identifying end-of-life components in a mainframe COBOL environment. - Reviewing a WordPress or Drupal stack for risky configurations and abandoned plugins. - Assessing the security posture of web-to-mainframe API gateways and wrappers. ## Known limitations This is a read-only audit and does not execute exploit code or provide functional PoCs. Mainframe assessments require specific environment data or source code access to be effective.

    Read-only audit. Does not execute exploit code or PoCs. Requires user-provided inventory or command access. Mainframe checks depend on available JCL or source files.

    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

    4.0
    1 review
    5
    0
    4
    1
    3
    0
    2
    0
    1
    0

    This skill has been used 6 times.

    Trust & safety

    Security scan

    Security scan passed 20 days ago

    • Free with an Agensi account

    Permissions

    Network Access
    Read Files
    Write Files
    Listed21 days ago
    Updated20 days ago

    What's inside

    Frequently Asked Questions

    Standard scanners often fail to account for version backporting or end-of-life (EOL) status in older systems. This skill solves that by cross-referencing your specific inventory with live CISA KEV data and vendor-specific trackers to identify risks that generic tools miss.

    The skill is compatible with any AI agent or platform that supports external data integration and file analysis. It works best when provided with server outputs, dependency locks (like composer.json), or COBOL source code snippets via your preferred LLM interface.

    No complex installation is required; once added to your agent, you simply provide the target environment's metadata or configuration files. For mainframe environments, you may need to provide outputs from Zowe or specific JCL logs to facilitate the inventory collection phase.

    This skill performs a non-intrusive, read-only analysis of configurations and code patterns. It identifies vulnerabilities and ranks them by exploitability without ever executing active exploit code or compromising the stability of your production environment.

    Unlike static prompts, this skill uses a structured verification workflow and live API queries to fetch the most recent threat intelligence. This ensures your audit includes the latest zero-days and CISA 'Exploited-in-the-Wild' updates that occurred after an LLM's training cutoff.

    You receive the full agent skill logic designed for PHP and COBOL stacks, including the specialized workflows for threat intelligence matching, EOL tracking, and prioritized remediation reporting.

    New in LLM Ops