cve-poc-generator

Generate a standalone Python PoC and markdown vulnerability report for a CVE ID.

Updated May 27, 2026
One-click install
npx skills add https://github.com/hoanghn61/.agents --skill cve-poc-generator-hoanghn61
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cve-poc-generator
Source: https://github.com/hoanghn61/.agents/tree/main/skills/cve-poc-generator
Command: npx skills add https://github.com/hoanghn61/.agents --skill cve-poc-generator-hoanghn61

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill helps you research a CVE and produce a safe, standalone proof-of-concept plus a detailed vulnerability report, saving time versus manual research and writing while reducing the risk of harmful PoCs.

Core Features & Use Cases

  • CVE research from NVD and advisories: Pulls CVSS v3.1 scoring, CWE IDs, CPE matches, timelines, and source links from NVD, then deep-dives vendor and public advisories to document root cause and affected versions.
  • Deterministic, safe PoC script generation: Generates a standalone Python poc.py using standard libraries plus requests, designed for least-harm validation with a read-only default and an explicit --confirm for any potentially impactful action.
  • Comprehensive report generation: Produces report.md with metadata, root cause analysis, risk assessment, affected product ranges, remediation guidance, and fully cited sources.

Quick Start

Run the skill with a target CVE ID to generate a poc.py and a report.md in the outputs/ processed and reports folders.

Frequently Asked Questions about cve-poc-generator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate a safe CVE proof-of-concept script for vulnerability research?

A safe CVE proof-of-concept is generated by providing a CVE ID to automate NVD lookup and advisory research. The process outputs a standalone Python poc.py script using standard libraries and requests, designed for least-harm validation with a read-only default and an explicit confirmation flag for impactful actions.

Can I extract CVSS v3.1 scoring and CPE matches directly from NVD for a vulnerability report?

Yes, extracting CVSS v3.1 scoring, CWE IDs, CPE matches, and timelines directly from NVD is supported. The skill deep-dives into vendor advisories to document root causes and affected versions, compiling findings into a comprehensive markdown vulnerability report with fully cited sources.

What is included in an automated vulnerability report for a CVE?

An automated vulnerability report includes metadata, root cause analysis, risk assessment, affected product ranges, and remediation guidance. It reuses exact NVD scoring vectors and ensures every claim is fully cited with source links from public advisories.

Do I need external Python packages to run the generated proof-of-concept for CVE validation?

You only need the requests package as an external dependency to run the generated proof-of-concept. The poc.py script is built using standard Python libraries alongside requests, ensuring strict rules for least harm and safe non-destructive verification.

What are the limitations of using automated PoC generation for security reporting?

The limitation of automated PoC generation is its strict adherence to non-destructive verification, requiring an explicit --confirm flag for any potentially impactful actions. It applies to security research workflows needing accurate CVSS v3.1 extraction and source-attributed analysis.