HalluciScan is an AI-powered hallucination detection tool that analyzes AI-generated content and identifies factual errors, unsupported claims, and hallucinations in real-time. Paste any AI response and let HalluciScan verify it using Google Gemini and Groq's LLaMA models.
- 🔬 Real-time Hallucination Detection — Analyze any AI-generated text instantly
- 🌐 Multi-Model Analysis — Uses Google Gemini + Groq LLaMA for cross-verification
- 📄 PDF Upload Support — Scan documents for AI-generated misinformation
- 📊 Confidence Scoring — Get a hallucination confidence score per sentence
- 🔦 Sentence Highlighting — Visual breakdown of flagged vs. verified claims
- 📋 Evidence List — See the reasoning behind each flagged claim
- 📱 Browser Extension — Detect hallucinations directly in your browser
- 🌙 Dark Mode — Sleek dark-first design
👉 https://halluciscan.vercel.app
| Technology | Purpose |
|---|---|
| Next.js 16 | Full-stack React framework |
| TypeScript | Type safety |
| Tailwind CSS v4 | Styling |
| Framer Motion | Animations |
| Google Gemini API | Primary AI analysis model |
| Groq (LLaMA) | Secondary AI model for cross-verification |
git clone https://github.com/deepanshu83/HalluciScan.git
cd HalluciScannpm installCreate a .env.local file in the root directory:
# Google Gemini API Key (get it from https://aistudio.google.com/app/apikey)
GOOGLE_API_KEY=your_google_api_key_here
# Groq API Key (get it from https://console.groq.com/keys)
GROQ_API_KEY=your_groq_api_key_here
# Optional: Override Gemini model (default: gemini-2.0-flash)
# GEMINI_MODEL=gemini-2.0-flashnpm run devOpen http://localhost:3000 in your browser.
| Key | Where to Get |
|---|---|
GOOGLE_API_KEY |
Google AI Studio |
GROQ_API_KEY |
Groq Console |
halluciscan/
├── app/ # Next.js App Router
│ ├── api/analyze/ # API route for hallucination analysis
│ ├── analyzer/ # Analyzer page
│ ├── dashboard/ # Dashboard page
│ ├── blog/ # Blog page
│ └── page.tsx # Landing page
├── components/ # Reusable UI components
│ ├── landing/ # Landing page sections
│ ├── ResultPanel.tsx # Analysis result display
│ ├── TextAnalyzer.tsx # Text input analyzer
│ └── UploadPDF.tsx # PDF upload component
├── lib/ # Utility libraries
│ ├── geminiClient.ts # Google Gemini API client
│ └── llamaClient.ts # Groq LLaMA API client
├── extension/ # Browser extension source
└── public/ # Static assets
This project is deployed on Vercel.
npm install -g vercel
vercel --prodDon't forget to add your environment variables in the Vercel Dashboard under Settings → Environment Variables.
MIT License © 2025 Deepanshu