🧠 AI Humanizer: The Ultimate AI-to-Human Text Transformer

GitHub Repository

A high-performance, research-grade platform designed to bypass AI detection and transform machine-generated content into high-quality, undetectable human-like prose.

📋 Table of Contents

• Overview
• Architecture
• Key Features
• Quick Start
• Installation
• Usage
• Development
• Testing
• Contributing
• FAQ
• License

---

🎯 Overview

AI Humanizer is an advanced, full-stack application that solves the growing challenge of AI content detection and machine-like writing. Built with Node.js, React 19, and PostgreSQL, it provides users with a comprehensive Iterative Detection-Refinement Loop, ensuring AI-generated text consistently passes audits (like GPTZero) while maintaining impeccable flow.

What Makes AI Humanizer Different

• Evidence-Based Refinement: Grounded in linguistic analysis and adversarial detection theory.
• AI-Powered Analysis: Real-time sentiment preservation and pattern detection via Gemini Pro.
• Recursive Optimization: Rewrites content until the AI Probability Score drops below 15%.
• Privacy-First: Secure JWT-based authentication and modular processing of sensitive documents.
• Research-Driven: Incorporates findings on machine-sounding syntax to deliver human-like prose.

---

🏗️ Architecture

graph TB
    A[React 19 + Vite<br/>Presentation Layer] --> B[Express.js<br/>Business Logic]
    B --> C[Prisma ORM<br/>Data Integrity]
    C --> D[(PostgreSQL Server<br/>Data Storage)]

    B --> E[Refinement Engine<br/>Google Gemini AI]
    B --> F[Detection Engine<br/>GPTZero API]
    B --> G[Payment Integration<br/>Stripe API]

    H[File System] --> C
    I[External APIs] --> B

    subgraph "Core Components"
        J[User Management]
        K[Iterative Loop Logic]
        L[Document Extractor]
        M[Analytics Engine]
    end

    B --> J
    B --> K
    B --> L
    B --> M

Loading

System Components

Component	Technology	Purpose
Frontend UI	React 19 + Vite	Ultra-fast, responsive web interface
Styling	Tailwind CSS	Modern, premium utility-first styling
Animations	Framer Motion	Fluid micro-interactions and transitions
Backend Engine	Node.js, Express	Core business logic and API delivery
Database	PostgreSQL	Robust relational data persistence
ORM Layer	Prisma	Type-safe database management
ML Content Core	Google Gemini	Advanced text re-humanization services
Detection Core	GPTZero API	Real-time AI detection scoring
Auth System	bcrypt, JWT	Secure user authentication
Migration System	Prisma CLI	Database schema evolution

Data Flow

Input Text → Pattern Analysis → Refinement → AI Probability Audit → [Reprocess if > 15%] → Output → UI Update

---

✨ Key Features

Core Humanization

• ✅ Iterative Loop: Up to 3 recursive passes for maximum undetectability.
• ✅ Style Casting: Specialized modes for Academic, Professional, and Creative needs.
• ✅ Grammar Guard: Automatic syntax correction while preserving tone.
• ✅ Tone Consistency: Ensures the core message remains indistinguishable from human writing.

AI & Analytics

• Real-time Scoring: Integrated GPTZero audits with live reporting.
• Plagiarism Guard: Built-in verification to ensure original content.
• Pattern Recognition: Detects machine-like sentence structures and repetitive syntax.
• Trend Analysis: Visualize your "Humanization Journey" with interactive charts.
• ML Integration: Context-aware re-writing that understands cultural nuances.

User Experience

• File Support: Drag and drop extraction for .pdf, .docx, and .txt files.
• Multi-Export: One-click download of humanized results in multiple formats.
• Premium Dashboard: Retractable sidebars, glassmorphism UI, and dark/light modes.
• Secure Authentication: Fully encrypted user data and session management.
• Data Management: Full export/delete capabilities for project history.

Developer Experience

• 🧪 Comprehensive Testing: Dedicated test suites for both Frontend and Backend.
• 🔄 Type-Safe Stack: Full TypeScript integration across the entire application.
• 🐳 Docker Ready: Standardized environments for consistent development.
• 📖 Self-Documenting API: Clean, RESTful endpoint architecture.

---

🚀 Getting Started

1. Setup Environment

# Clone the repository
git clone https://github.com/Sappymukherjee214/AI-Humanizer.git
cd AI-Humanizer

# Initialize Backend
cd backend
npm install
# Create .env based on the template
cp .env.example .env

# Initialize Frontend
cd ../frontend
npm install

2. Launch Application

A. Backend API (Primary)

cd backend
npx prisma db push
npm run dev

B. Frontend Client (Web)

cd frontend
npm run dev

The application will be available at http://localhost:5173.

---

🛠️ Developer Workflow

Tip

Prisma Studio: Use npx prisma studio in the backend directory to instantly visualize and edit your local database.

---

Tip

Contributing Workflow: If you are contributing specifically to the Web frontend, ensure the Backend API is running so the dashboard can fetch state.

Note

For detailed architecture, sidecar management, and GSSoC'26 guidelines, see CONTRIBUTING.md and SECURITY.md.

---

🎮 Usage

For Users

1. Launch: Open both the frontend and backend servers.
2. Setup: Create your profile via the secure Sign In modal.
3. Analyze: Paste your AI text or upload a document for pattern analysis.
4. Transform: Select your specialization (e.g., "Academic") and click "Run Humanizer".
5. Results: View your AI score and export the humanized prose.

For Developers

API Usage

// Request for humanization
const response = await fetch("/api/humanize", {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({ 
    text: "AI text...", 
    mode: "professional" 
  })
});

const data = await response.json();
console.log(data.humanized);

CLI Tools

# Database seeding
npx prisma db seed

# Type generation
npx prisma generate

---

🧪 Testing

Run Test Suite

# All tests (Root)
npm run test

# Frontend specific
cd frontend && npm test

# Backend specific
cd backend && npm test

Test Categories

• Unit Tests: Core function/component testing.
• Integration Tests: Database and service integration (Prisma).
• Service Tests: External API mocking (Gemini/GPTZero).
• UI Tests: Headless component and accessibility verification.

---

🤝 Contributing

We welcome contributions! Please see our Contributing Guide.

Development Workflow

1. Fork the repository
2. Create a feature branch: git checkout -b feature/amazing-feature
3. Make your changes with tests
4. Commit your changes: git commit -m 'Add amazing feature'
5. Push to the branch: git push origin feature/amazing-feature
6. Open a Pull Request

---

❓ FAQ

General Questions

Is this a plagiarism tool? No. It is a refinement tool for AI-generated text to restore human flow and tone.

Are my documents stored? Data is stored securely in your private project history. You can delete it at any time.

How are the results calculated? Results are based on an adversarial loop between the Refinement Engine (Gemini) and the Detection Audit (GPTZero).

Technical Questions

What are the system requirements?

• Node.js v18.0.0+
• PostgreSQL v14+
• Modern Browser (Chrome, Firefox, Safari)

How do I backup my data? Database backups are available via standard pg_dump or Prisma CLI export strategies.

---

🛡️ License

This project is licensed under the ISC License - see the LICENSE file for details.

---

🙏 Acknowledgments

• Research Core: Based on NLP findings on adversarial AI-detection patterns.
• Open Source: Built with React, Vite, Prisma, and PostgreSQL.
• GSSoC 2026: Special thanks to the GirlScript community for supporting this project.

---

Built with ❤️ for authentic human expression and personal growth by Saptarshi Mukherjee