# πŸ₯· StealthHumanizer v2 **A comprehensive free & open-source AI text humanizer.** [![Live Demo](https://img.shields.io/badge/🌐_Live_Demo-GitHub_Pages-22c55e?style=for-the-badge&logo=github)](https://rudra496.github.io/StealthHumanizer/) [![Deploy](https://img.shields.io/badge/πŸš€_Deploy-Vercel-000?style=for-the-badge&logo=vercel)](https://vercel.com/new/clone?repository-url=https://github.com/rudra496/StealthHumanizer) [![License: MIT](https://img.shields.io/badge/License-MIT-22c55e?style=for-the-badge)](LICENSE) [![PRs Welcome](https://img.shields.io/badge/PRs-Welcome-22c55e?style=for-the-badge)](CONTRIBUTING.md) [![Stars](https://img.shields.io/github/stars/rudra496/StealthHumanizer?style=social)](https://github.com/rudra496/StealthHumanizer/stargazers) [![Forks](https://img.shields.io/github/forks/rudra496/StealthHumanizer?style=social)](https://github.com/rudra496/StealthHumanizer/network/members) [![Issues](https://img.shields.io/github/issues/rudra496/StealthHumanizer)](https://github.com/rudra496/StealthHumanizer/issues) Transform AI-generated text into **natural, human-like writing** using 13 AI providers, 4 rewrite levels, 13 tones, multi-pass humanization, and a 12-metric detection engine. **βœ… 100% Free β€’ No Login β€’ No Limits β€’ Open Source (MIT) β€’ Private & Secure**

πŸ“Ί Demo

πŸ’‘ Try it now: Live Demo on GitHub Pages β€” no installation needed!

How It Works

πŸ“ AI Text β†’ πŸ₯· Multi-Pass Rewrite β†’ πŸ“Š AI Detection β†’ πŸ” Re-Humanize β†’ βœ… Human Score 100%
  1. Paste your AI-generated text (up to 10,000 words)
  2. Select rewrite level, style, and tone
  3. Set target human score (50%–100%)
  4. Humanize β€” AI rewrites with human patterns
  5. Auto-loop β€” re-humanizes flagged sentences until target reached
  6. Export β€” copy, download TXT/DOCX

✨ Features

πŸ₯· Rewrite Engine

Feature Details
4 Rewrite Levels Light, Medium, Aggressive, Ninja (5-pass auto-loop)
5 Writing Styles Academic, Professional, Casual, Creative, Technical
13 Tone Presets Conversational, Academic Formal/Casual, Journalistic, Creative Writing, Professional, Technical, Persuasive, Storytelling, Humorous, Emotional, Analytical
Multi-Pass Humanization Ninja: 5 passes, Aggressive: 3 passes, auto-target 100%
Target Score Control Slider from 50% to 100% human
Alternative Rewrites Click any sentence for 3 different versions
Side-by-Side Comparison Diff view with per-sentence scores
16 Languages Auto-detects and preserves input language

πŸ” 12-Metric AI Detector

Metric What It Measures
Perplexity How predictable/unexpected the word choices are
Burstiness Sentence length variation (humans vary wildly)
Vocabulary Diversity Unique word usage vs repetition
Sentence Variation Length, structure, type diversity
Sentence Start Diversity Do sentences start the same way?
Pronoun Usage First/second person pronouns (human indicator)
Transition Frequency Overuse of formal transitions
Passive Voice Ratio of passive constructions
AI Phrase Density Detection of 50+ known AI phrases
Hedging Language "It could be argued", "one might consider"
Quantifier Patterns Overuse of "numerous", "various", "multiple"
Contraction Usage Humans use contractions; AI often doesn't

Readability Scores: Flesch Reading Ease, Flesch-Kincaid Grade Level, Coleman-Liau Index

πŸ†“ 13 AI Providers (10 Free!)

Provider Free? Speed Quality Model
Google Gemini βœ… ⚑ ⭐⭐⭐⭐⭐ gemini-1.5-flash/pro
Groq βœ… ⚑⚑⚑ ⭐⭐⭐⭐ llama-3.3-70b
ZAI (GLM-5) βœ… ⚑⚑ ⭐⭐⭐⭐⭐ glm-5
OpenRouter βœ… ⚑⚑ ⭐⭐⭐⭐ 100+ models
Together AI βœ… ⚑⚑ ⭐⭐⭐⭐ llama-3-70b
Cerebras βœ… ⚑⚑⚑ ⭐⭐⭐⭐ llama3.1-70b
Mistral AI βœ… ⚑⚑ ⭐⭐⭐⭐ mistral-large
Cohere βœ… ⚑⚑ ⭐⭐⭐ command-r-plus
DeepInfra βœ… ⚑⚑ ⭐⭐⭐⭐ llama-3-70b
HuggingFace βœ… ⚑ ⭐⭐⭐ llama-3-8b
Cloudflare Workers AI βœ… ⚑⚑⚑ ⭐⭐⭐ llama-3-8b
OpenAI GPT-4 ❌ ⚑ ⭐⭐⭐⭐⭐ gpt-4o
Anthropic Claude ❌ ⚑ ⭐⭐⭐⭐⭐ claude-sonnet-4

πŸ’‘ No API key? Start with Gemini or Groq β€” both free, no credit card needed!

🎨 UX Features

  • πŸŒ™ Dark / β˜€οΈ Light mode
  • ⌨️ Keyboard shortcuts (Ctrl+Enter, Ctrl+1/2/3/4)
  • πŸ“œ History (localStorage, 50 entries)
  • πŸ“‹ Copy / πŸ’Ύ Export as TXT & DOCX
  • πŸ“± Fully mobile responsive
  • πŸ”’ Zero server data storage
  • πŸƒ No account or sign-up required

πŸš€ Quick Start

Option 1: 🌐 Use Online (Zero Install)

πŸ‘‰ Live Demo β€” add any free API key and start humanizing!

Option 2: πŸš€ Deploy on Vercel (Recommended)

Deploy with Vercel

Option 3: πŸ’» Run Locally

git clone https://github.com/rudra496/StealthHumanizer.git
cd StealthHumanizer
npm install
npm run dev

Step-by-Step Usage

  1. Get a free API key:

    • Google Gemini β€” Sign in β†’ Create API Key β†’ Copy
    • Groq β€” Sign up β†’ Create key β†’ Paste
  2. Paste your AI-generated text

  3. Configure:

    • Level: Light / Medium / Aggressive / Ninja
    • Style: Academic / Professional / Casual / Creative / Technical
    • Tone: Conversational / Formal / Journalistic / etc.
    • Target: 80% β†’ 100%
  4. Click "Humanize" β€” get human-like natural writing!


πŸ† StealthHumanizer vs Others

Feature StealthWriter QuillBot StealthHumanizer
Price $20–50/mo $19.95/mo FREE forever
Daily Limit 10–150 Limited Unlimited
Word Limit 1K–5K 1.2K 10K
AI Providers 1 1 13 (10 free)
Rewrite Levels 3 2 4 (+ Ninja)
Writing Styles 3 4 5
Tone Presets 0 0 13
Multi-Pass ❌ ❌ βœ… (up to 5)
Target Score ❌ ❌ βœ… (50–100%)
AI Detector Basic None 12 metrics
Readability ❌ ❌ 3 scores
Open Source ❌ ❌ βœ… MIT
Self-Hostable ❌ ❌ βœ…
API Keys Stored Their server Their server Your browser only
No Account ❌ ❌ βœ…

πŸ› οΈ Tech Stack

Layer Technology
Framework Next.js 14 (App Router)
Language TypeScript
Styling Tailwind CSS
AI Integration Native fetch() β€” zero SDK dependencies
Deployment Vercel / GitHub Pages
Dependencies next, react, lucide-react β€” that's it

πŸ”’ Security & Privacy

  • πŸ”‘ API keys never leave your browser β€” stored only in localStorage
  • 🚫 No server-side storage β€” no databases, no user accounts
  • πŸ” All API calls use HTTPS β€” encrypted in transit
  • πŸ“΅ No tracking, no analytics β€” zero third-party scripts
  • πŸ›‘οΈ MIT Licensed β€” fully transparent, auditable code
  • See SECURITY.md for full security policy

πŸ“ Project Structure

StealthHumanizer/
β”œβ”€β”€ app/
β”‚   β”œβ”€β”€ api/
β”‚   β”‚   β”œβ”€β”€ humanize/route.ts    # Multi-pass humanization endpoint
β”‚   β”‚   └── alternative/route.ts # Alternative rewrite suggestions
β”‚   β”œβ”€β”€ layout.tsx               # Root layout with metadata
β”‚   β”œβ”€β”€ page.tsx                 # Main application page
β”‚   └── globals.css              # Global styles + dark mode
β”œβ”€β”€ components/
β”‚   β”œβ”€β”€ Humanizer.tsx            # Main humanization interface
β”‚   β”œβ”€β”€ Detector.tsx             # AI detection dashboard
β”‚   β”œβ”€β”€ Settings.tsx             # API key management
β”‚   β”œβ”€β”€ History.tsx              # Humanization history
β”‚   β”œβ”€β”€ Navbar.tsx               # Navigation bar
β”‚   └── Toast.tsx                # Notification toasts
β”œβ”€β”€ lib/
β”‚   β”œβ”€β”€ types.ts                 # TypeScript types
β”‚   β”œβ”€β”€ providers.ts             # 13 AI provider configs
β”‚   β”œβ”€β”€ prompts.ts               # 13 tones Γ— 4 levels system prompts
β”‚   β”œβ”€β”€ detector.ts              # 12-metric AI detection engine
β”‚   β”œβ”€β”€ humanizer.ts             # Multi-pass humanization logic
β”‚   β”œβ”€β”€ readability.ts           # Readability score calculations
β”‚   └── storage.ts               # localStorage management
β”œβ”€β”€ docs/
β”‚   β”œβ”€β”€ index.html               # Standalone GitHub Pages version
β”‚   β”œβ”€β”€ sitemap.xml              # SEO sitemap
β”‚   └── robots.txt               # Crawler rules
β”œβ”€β”€ .github/
β”‚   β”œβ”€β”€ workflows/
β”‚   β”‚   β”œβ”€β”€ pages.yml            # GitHub Pages auto-deploy
β”‚   β”‚   └── ci.yml               # Build verification
β”‚   └── ISSUE_TEMPLATE/          # Bug report & feature request
└── [Documentation files]
    β”œβ”€β”€ README.md                ← You are here
    β”œβ”€β”€ ARCHITECTURE.md          # System architecture
    β”œβ”€β”€ STYLE_ENGINE.md          # Tone/style system design
    β”œβ”€β”€ API_USAGE.md             # Provider API details
    β”œβ”€β”€ SECURITY.md              # Security policy
    β”œβ”€β”€ CONTRIBUTING.md          # How to contribute
    β”œβ”€β”€ ROADMAP.md               # Future plans
    β”œβ”€β”€ CODE_OF_CONDUCT.md       # Community guidelines
    └── LICENSE                  # MIT License

πŸ—ΊοΈ Roadmap

  • [x] v1: Core humanization engine
  • [x] v2: 13 providers, 13 tones, multi-pass, ninja mode
  • [x] GitHub Pages deployment
  • [x] SEO optimization & structured data
  • [x] GitHub Actions CI/CD
  • [ ] v3: Browser extension (Chrome/Firefox)
  • [ ] v3: Real-time streaming humanization
  • [ ] v3: Batch processing (multiple texts)
  • [ ] v3: Custom model fine-tuning guide

See ROADMAP.md for full details.


🀝 Contributing

Contributions are welcome! Please read CONTRIBUTING.md first.

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

πŸ“„ License

This project is licensed under the MIT License β€” use it however you want.

See LICENSE for details.


**Built with ❀️ by [Rudra Sarker](https://github.com/rudra496)** [![Twitter](https://img.shields.io/badge/Twitter-@Rudra496-1DA1F2?style=flat-square&logo=twitter)](https://twitter.com/Rudra496) [![Website](https://img.shields.io/badge/Website-rudra496.github.io-22c55e?style=flat-square&logo=github)](https://rudra496.github.io/site) [![GitHub](https://img.shields.io/badge/GitHub-rudra496-181717?style=flat-square&logo=github)](https://github.com/rudra496)

Shahjalal University of Science and Technology, Bangladesh

**⭐ If you find this useful, please give it a star! It helps more than you know.**
Share this project:

Updates