Enterprise Security Platform with Conversational AI
Automated security scanning pipeline with intelligent analysis, real-time conversations, and comprehensive reporting.
SentinelHub is a next-generation security platform that revolutionizes how teams approach application security. Built with enterprise-grade architecture, it combines automated scanning, conversational AI, and intelligent reporting to provide comprehensive security analysis.
Key Innovations:
- Conversational AI Security Expert - Chat with AI about your security findings in natural language
- Automated Security Pipeline - Complete CI/CD-style workflow from scan to report
- Real Security Tools Integration - Semgrep (1000+ rules), TruffleHog (700+ detectors), GitLeaks
- MongoDB Atlas Database - Persistent storage for scans, conversations, and analytics
- Intelligent Recommendations - AI-powered risk assessment and remediation guidance
Complete workflow automation that orchestrates:
- Security Scanning → Multiple tools (Semgrep, TruffleHog, Pattern Analysis)
- AI Analysis → Risk assessment and intelligent recommendations
- Conversation Initialization → AI ready to answer questions about findings
- Database Storage → Persistent MongoDB storage for all results
- Report Generation → Dashboard-ready analytics and exportable reports
Revolutionary chat interface powered by HuggingFace Gemma and OpenAI:
- Natural Language Security Discussions - Ask questions about vulnerabilities in plain English
- Context-Aware Responses - AI understands your specific scan results and codebase
- Actionable Remediation Steps - Get specific fix instructions and AWS CLI commands
- Compliance Guidance - GDPR, PCI-DSS, HIPAA implications explained
- Best Practices Education - Learn security concepts through conversation
Enterprise-grade security scanning with actual industry tools:
Static Analysis:
- Semgrep - 1000+ community security rules, OWASP Top 10, CWE Top 25
- ESLint Security - JavaScript/TypeScript vulnerability detection
- Pattern Matching - Multi-language vulnerability patterns
Secret Detection:
- TruffleHog - 700+ verified secret detectors with confidence scoring
- GitLeaks - Git history scanning with entropy analysis
📖 Complete Security Tools Guide - Detailed documentation of all security tools, their purposes, and when they're used.
- AWS/GitHub/API Key Detection - Real credential verification
AI Intelligence:
- HuggingFace Gemma - Advanced language model for security analysis
- OpenAI GPT - Enhanced conversational capabilities (optional)
- Risk Assessment - Automated threat level calculation
Comprehensive AWS S3 bucket security analysis:
- Bucket Configuration Analysis - Encryption, versioning, logging checks
- Permission Auditing - Public access, ACL, and policy analysis
- Content Scanning - Pattern-based secret detection in stored files
- Compliance Checking - GDPR, PCI-DSS, HIPAA violation detection
- AI-Powered Recommendations - Conversational guidance for security improvements
MongoDB Atlas integration for enterprise data management:
- Pipeline Results Storage - Complete scan history and metrics
- Conversation Sessions - AI chat history and context preservation
- Security Analytics - Trend analysis and dashboard data
- Report Generation - Exportable security reports and audit logs
- Search & Filtering - Advanced querying of historical scan data
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐
│ Frontend │ │ Conversation │ │ Dashboard │
│ (Next.js) │ │ AI │ │ Analytics │
└─────────────────┘ └─────────────────┘ └─────────────────┘
│ │ │
└───────────────────────▼───────────────────────┘
│
┌───────────────────────▼───────────────────────┐
│ Security Pipeline Orchestrator │
│ (Automated Workflow Management) │
└───────────────────────┬───────────────────────┘
│
┌────────────────────────────┼────────────────────────────┐
│ │ │
▼ ▼ ▼
┌─────────┐ ┌─────────────┐ ┌─────────────┐
│ Code │ │ S3 │ │ GitHub │
│Scanner │ │ Scanner │ │ Scanner │
│(Semgrep)│ │(AWS + AI) │ │(TruffleHog) │
└─────────┘ └─────────────┘ └─────────────┘
│ │ │
└────────────────────────────▼────────────────────────────┘
│
┌───────────────────────▼───────────────────────┐
│ MongoDB Atlas Database │
│ (Scans, Conversations, Reports) │
└───────────────────────────────────────────────┘
- Node.js 18+
- Docker (for Redis, SonarQube, Semgrep, TruffleHog)
- MongoDB Atlas account (provided)
- OpenAI API key (required for chat)
- AWS credentials for S3 scanning (optional)
1. Start API Gateway (All Backend Services)
cd "api-gateway"
npm install # First time only
npm run devPort: 5000 | Services: Scanning, GitHub, AWS, Docker, Reports, Chat AI (Integrated)
- Chat Service: Automatically starts on port 4000 (no separate command needed!)
- AI Responses: Now shorter, better formatted, and more focused
- Single Command: All backend services run with one command
2. Start Client (Frontend)
cd "client"
npm install # First time only
npm run devPort: 3000 | Interface: Web UI
- 🌐 Main Application: http://localhost:3000
- 🔧 API Health: http://localhost:5000/api/health
- 💬 Chat Health: http://localhost:4000/api/health (auto-started)
Client (3000) ←→ API Gateway (5000) ←→ Security Services
↓
Chat Service (4000) ←→ OpenAI API (Integrated)
Benefits:
- Single Command Startup - All backend services with one command
- Process Isolation - Chat service runs as child process, prevents crashes
- Auto-Recovery - Chat service automatically restarts if it fails
Port Already in Use
# Windows - Kill process on port
netstat -ano | findstr ":5000"
taskkill /PID [PID_NUMBER] /F
# Mac/Linux - Kill process on port
lsof -ti:5000 | xargs kill -9Services Not Starting
- Check
npm installcompleted in each directory - Ensure Docker is running (for Redis)
- Verify ports 3000, 5000, 4000 are available
Chat Not Responding
- Chat service auto-starts with API Gateway
- Check OpenAI API key in
services/.env - Chat only responds to security-related questions
- Look for "[CHAT SERVICE]" logs in API Gateway output
# Clone repository
git clone https://github.com/Sarahenia20/SentinelHub.git
cd SentinelHub
# Install dependencies
npm install
# Install security tools
pip install semgrep
# Download TruffleHog from: https://github.com/trufflesecurity/trufflehog/releases
# Configure environment (already set up)
# MongoDB Atlas, HuggingFace, AWS credentials are pre-configured
# Test the complete system
node test-pipeline-system.jsYour environment is pre-configured with:
const pipeline = new SecurityPipelineOrchestrator();
// Analyze JavaScript code with complete pipeline
const result = await pipeline.executePipeline({
type: 'code-analysis',
input: {
code: yourCode,
language: 'javascript'
}
});
console.log(`Risk Level: ${result.scanResults.summary.riskLevel}`);
console.log(`Findings: ${result.scanResults.findings.vulnerabilities.length}`);
console.log(`Chat Ready: ${result.conversation.canChat}`);// Chat about security findings
const chatResponse = await pipeline.chatAboutResults(
result.pipelineId,
"What are the most critical security issues I need to fix immediately?"
);
console.log(chatResponse.message);
// AI Response: "I found 3 critical security issues in your code.
// The most urgent is the SQL injection vulnerability on line 42..."// Comprehensive S3 security scan with AI recommendations
const s3Result = await pipeline.executePipeline({
type: 's3-bucket',
input: {
bucketName: 'my-bucket',
credentials: { /* AWS creds */ }
}
});
// Chat about S3 security
const s3Chat = await s3Result.conversation.chat(
"How do I secure my S3 bucket against public access?"
);const db = new MongoDBManager();
// Store and retrieve scan results
await db.storePipelineResults(pipelineData);
const results = await db.getPipelineResults(pipelineId);
// Get security metrics for dashboard
const metrics = await db.getSecurityMetrics();
console.log(`Total Scans: ${metrics.totalScans}`);
console.log(`Critical Issues: ${metrics.criticalIssues}`);- Semgrep: 1000+ community security rules, SAST analysis
- ESLint Security: JavaScript/TypeScript vulnerability detection
- Pattern Matcher: Multi-language security pattern detection
- TruffleHog: 700+ verified secret detectors
- GitLeaks: Git history secret scanning
- Custom Patterns: API keys, tokens, credentials
- HuggingFace Gemma: Conversational security expert
- OpenAI GPT: Enhanced natural language processing
- CVE Database: Real-time vulnerability intelligence
- AWS S3 Scanner: Bucket security analysis
- IAM Assessment: Permission and policy analysis
- Compliance Checking: GDPR, PCI-DSS, HIPAA
SentinelHub/
├── services/
│ ├── security-pipeline/
│ │ └── pipeline-orchestrator.js # Main automation pipeline
│ ├── real-security-scanner.js # Code analysis orchestrator
│ ├── cloud-security/
│ │ └── s3-bucket-scanner.js # AWS S3 security scanner
│ ├── ai-intelligence/
│ │ ├── conversation-ai.js # Conversational AI system
│ │ └── huggingface-gemma.js # Gemma AI integration
│ ├── database/
│ │ └── mongodb-manager.js # MongoDB Atlas operations
│ └── real-security-tools/
│ ├── semgrep-integration.js # Semgrep SAST integration
│ └── trufflehog-integration.js # TruffleHog secret detection
├── client/ # Next.js frontend
├── api-gateway/ # Express.js API gateway
├── .env # Environment configuration
├── package.json # Dependencies and scripts
└── test-pipeline-system.js # Complete system test
Run the comprehensive system test:
node test-pipeline-system.jsThis tests:
- Complete pipeline execution
- Database storage and retrieval
- Conversational AI functionality
- Security tool integration
- S3 scanner (if AWS configured)
// Execute security pipeline
POST /api/pipeline/execute
{
"type": "code-analysis|s3-bucket|github-repository",
"input": { /* scan input */ },
"options": { /* scan options */ }
}
// Chat about results
POST /api/pipeline/chat/:pipelineId
{
"message": "How do I fix the SQL injection vulnerability?"
}// Get recent scans
GET /api/database/pipelines?limit=10
// Get security metrics
GET /api/database/metrics
// Search scans
GET /api/database/search?q=critical&type=code-analysis# Build for production
npm run build
# Start production server
npm run start
# Or use Docker
docker build -t sentinelhub .
docker run -p 3001:3001 sentinelhubThe platform is designed for enterprise CI/CD integration:
- Automated security scanning in build pipelines
- Conversation AI for development teams
- MongoDB Atlas for persistent scan history
- Dashboard integration for security metrics
- Automated security pipeline orchestration
- Conversational AI with Gemma and OpenAI
- MongoDB Atlas database integration
- Real security tools (Semgrep, TruffleHog)
- AWS S3 security scanning
- Complete testing suite
- Voice Assistant Integration - Eleven Labs text-to-speech
- API Gateway - RESTful endpoints for frontend integration
- Advanced Reporting - Executive dashboards and exports
- GitHub Actions Integration - CI/CD pipeline scanning
- Multi-tenant enterprise deployment
- Advanced compliance frameworks (SOC2, PCI-DSS)
- Custom security rule creation
- Mobile security monitoring app
We welcome contributions! Areas where you can help:
- Security Tool Integrations - Add new scanning tools
- AI Model Integration - Enhance conversation capabilities
- Cloud Provider Support - Azure, GCP security scanning
- Frontend Development - Dashboard and UI improvements
- Documentation - Setup guides and tutorials
- Local Processing - All analysis performed locally, no data sent to third parties
- Secure Storage - MongoDB Atlas with encryption at rest
- API Security - Rate limiting and authentication on all endpoints
- Secret Management - Secure handling of API keys and credentials
- Audit Logging - Complete tracking of security activities
- Issues: GitHub Issues
- Documentation: Complete setup and usage guides included
- Testing: Comprehensive test suite for validation
MIT License - see LICENSE file for details.
Built with enterprise security in mind by Sarah Henia
SentinelHub - Where AI meets enterprise security scanning