PhishGuard is a high-performance, Cyberpunk Dark themed web application designed to detect phishing and scam attempts in real-time using Google's gemini-3-flash-preview neural model.
- AI Backend: Node.js + Express +
@google/genai - Frontend Matrix: React 18 + Vite + Tailwind CSS v4
- Animations: Anime.js + Lenis Smooth Scroll
- Security: Helmet, CORS origin-locking, IP-based rate limiting, and strict schema validation.
- Rate Limiting: 15 requests per minute per IP to prevent matrix flooding.
- Strict Validation: All analysis requests are sanitized and strictly validated against the matrix schema (only
message10-5000 chars allowed). - Vault Management: No secrets are exposed to the frontend; all API communication is routed through the secure Node gateway.
- OWASP Alignment: Implements protection against common web vulnerabilities using industry-standard middlewares.
PHISHGUARD/
├── backend/ # Express API Server
│ ├── routes/ # Security-hardened endpoints
│ ├── services/ # Gemini AI integration
│ ├── middleware/ # Security & Error layers
│ └── .env # Secret configuration (Git ignored)
└── frontend/ # React + Tailwind v4 UI
├── src/hooks/ # Neural API orchestration
├── src/components/ # Cyberpunk components (Glows, Scans)
└── src/index.css # Matrix design tokens
cd backend
npm install
# Configure .env with GEMINI_API_KEY
npm startcd frontend
npm install
npm run dev- Node.js: v18+
- API Access: Google Gemini API Key
- Browser: Modern browser with Hardware Acceleration (for animations)
© 2026 Cybershield Neural Labs // Protocol: PHISHGUARD-WAVE-FINAL