🛡️ KAVACH

Digital Armour Against Modern Cyber Threats

Digital Armour Against Modern Cyber Threats

🚀 Overview

KAVACH (Digital Armour) is an AI-powered cybersecurity platform designed to detect, analyze, explain, and mitigate modern cyber threats in real time.

Unlike traditional security tools that only generate alerts, KAVACH provides explainable threat intelligence, personalized risk assessments, psychological vulnerability analysis, prompt injection detection, phishing analysis, steganography detection, deepfake detection, and incident management through a unified platform.

Built for organizations, security teams, and everyday users, KAVACH transforms complex cybersecurity threats into actionable insights.

---

🌐 Live Demo

🔗 https://kavach-indianext-hackathon.vercel.app/

---

✨ Key Features

🎣 Advanced Phishing Detection

• 35-signal heuristic evaluation
• Psychological manipulation detection
• Sender reputation analysis

🔗 Malicious URL Analysis

• WHOIS intelligence
• Levenshtein domain similarity detection
• Brand impersonation detection
• Suspicious domain scoring

🛡️ Prompt Injection Detection

• 41 attack pattern signatures
• Jailbreak detection
• Prompt manipulation analysis

🔍 StegoScan Engine

• LSB pixel analysis
• Unicode zero-width character detection
• Hidden payload discovery
• EXIF metadata inspection

🎭 Deepfake Detection

• AI-generated media analysis
• Image manipulation detection
• Video authenticity verification
• Content trust scoring

🧠 Explainable AI (XAI)

• Human-readable threat explanations
• Psychological vulnerability scoring
• Consequence simulation
• Executive-friendly reports

📊 Incident Dashboard

• Real-time monitoring
• Historical threat records
• Risk categorization
• Threat analytics

🌍 ThreatPulse Intelligence

• Security news aggregation
• Community threat monitoring
• Emerging attack trend analysis
• Real-time cybersecurity updates

---

🏗️ Architecture

┌──────────────────────────────────────────┐
│           INPUT & FRONTEND               │
├──────────────────────────────────────────┤
│ Web Dashboard                            │
│ Mail Scanner                             │
│ Chrome Extension                         │
│ ThreatPulse                              │
│ Deepfake Detector                        │
└──────────────────────────────────────────┘
                    │
                    ▼
┌──────────────────────────────────────────┐
│           ANALYSIS ENGINES               │
├──────────────────────────────────────────┤
│ Phishing Engine                          │
│ URL Engine                               │
│ Injection Engine                         │
│ StegoScan                                │
│ Deepfake Engine                          │
└──────────────────────────────────────────┘
                    │
                    ▼
┌──────────────────────────────────────────┐
│             FUSION ENGINE                │
├──────────────────────────────────────────┤
│ Phishing Weight  : 45%                   │
│ URL Weight       : 35%                   │
│ Injection Weight : 20%                   │
│ Threshold        : 0.45                  │
└──────────────────────────────────────────┘
                    │
                    ▼
┌──────────────────────────────────────────┐
│        INTELLIGENCE & OUTPUT             │
├──────────────────────────────────────────┤
│ XAI Engine                               │
│ Safe Word Challenge                      │
│ Incident Dashboard                       │
│ ThreatPulse                              │
└──────────────────────────────────────────┘

---

🔄 System Workflow

1. User submits content through dashboard, extension, scanner, or detector.
2. Inputs are processed by specialized detection engines.
3. Each engine produces an independent threat score.
4. Fusion Engine combines results using weighted risk scoring.
5. XAI Engine generates explanations and impact simulations.
6. Safe Word Challenge is triggered for high-risk events.
7. Incident Dashboard logs and visualizes findings.
8. Threat intelligence is enriched through ThreatPulse.

---

🧠 Explainable AI Layer

KAVACH focuses on explainability rather than black-box decisions.

Generated outputs include:

• Psychological Vulnerability Score
• Consequence Simulator
• Executive Summary
• Threat Confidence Score
• Recommended Actions
• Detailed Threat Report

---

🔐 Safe Word Challenge

When risk score exceeds 0.70, KAVACH triggers a human verification mechanism.

Purpose:

• Prevent social engineering attacks
• Detect account compromise
• Verify user authenticity
• Reduce automated exploitation

---

📡 ThreatPulse Intelligence

ThreatPulse continuously gathers intelligence from trusted cybersecurity sources:

• Hacker News
• Krebs on Security
• BleepingComputer
• Dark Reading
• Security Week

This allows KAVACH to stay updated with emerging threats and attack patterns.

---

⚙️ Tech Stack

Frontend

• React
• Vite
• Tailwind CSS
• Vercel

Backend

• FastAPI
• Python 3.11
• SQLite

AI Models

• Groq API

Deployment

• Vercel
• Hugging Face Spaces

---

🚀 Getting Started

Clone Repository

git clone https://github.com/YOUR_USERNAME/kavach.git

cd kavach

Install Dependencies

npm install

Start Development Server

npm run dev

# or

yarn dev

# or

pnpm dev

# or

bun dev

---

🔮 Future Roadmap

• Mobile Application
• SOC Integration
• SIEM Connectors
• Threat Intelligence Sharing
• Real-time Collaboration
• Enterprise SSO
• Automated Incident Response
• Deepfake Voice Detection

---

🤝 Contributing

Contributions, feature requests, and bug reports are welcome.

1. Fork the repository
2. Create a feature branch
3. Commit your changes
4. Submit a pull request

---

👥 Team — Algorithm Avengers

Rudra Parmar

Daksh Gopani

Rishi Vejani

Samyak Chheda