PatchPilot Security Management System

A Flask web application for vulnerability management and penetration testing using NodeZero MCP, Apify, Claude AI, and Redis Cloud.

Overview

PatchPilot implements a Find-Fix-Verify (FFV) security cycle that:

1. Find: Scrapes web vulnerabilities using Apify
2. Fix: Uses Claude AI to correlate vulnerabilities with support tickets
3. Verify: Launches penetration tests via NodeZero MCP integration

Architecture

• Frontend: Bootstrap 5 with vanilla JavaScript
• Backend: Flask with Python 3.13
• AI Integration: Claude Messages API with MCP support
• Web Scraping: Apify MCP server
• Penetration Testing: NodeZero MCP server
• Vector Database: Redis Cloud for ticket similarity search

Setup

Prerequisites

• Python 3.13+
• NodeZero API account
• Apify API account
• Claude API account
• Redis Cloud instance

Installation

1. Clone the repository

2. Create virtual environment:

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   pip install flask anthropic redis python-dotenv requests

4. Create .env.local file:

   APIFY_API_KEY=your_apify_key
   ANTHROPIC_API_KEY=your_claude_key
   NODEZERO_API_KEY=your_nodezero_key
   REDIS_HOST=your_redis_host
   REDIS_PORT=your_redis_port
   REDIS_PASSWORD=your_redis_password
   

Running

Start the application:

source venv/bin/activate
python app.py

Access at http://localhost:5001

Features

Vulnerability Intelligence

• Web scraping for security vulnerabilities
• Support ticket vector database
• AI-powered correlation and prioritization

Dashboard

• System overview with metrics
• Recent activity monitoring
• Quick navigation

Penetration Testing

• AI-generated test scenarios
• Real-time progress tracking
• Log viewing and reporting

Reports

• Vulnerability assessments
• Penetration test results
• Executive summaries

Usage

1. Scrape Vulnerabilities: Use the first tab to collect security data from web sources
2. Create Vector Database: Build searchable database of support tickets
3. AI Analysis: Let Claude correlate vulnerabilities with tickets and prioritize targets
4. Launch Pentests: Start penetration tests on priority targets
5. Monitor Progress: Track test execution and view logs
6. Generate Reports: Export findings and recommendations

Configuration

Redis Connection

The application connects to Redis Cloud without SSL due to connection issues. See REDIS_FIX.md for troubleshooting.

MCP Servers

• NodeZero MCP: Handles penetration testing operations
• Apify MCP: Manages web scraping tasks

Development

The application uses Flask's debug mode for development. File changes trigger automatic reloads.

Key files:

• app.py: Main Flask application
• templates/: HTML templates
• static/: CSS and JavaScript files
• .env.local: Environment variables