🛡️ Claime AI - AI-Powered Fact-Checking System

Multi-Agent Verification System with AI-Powered Analysis

Features • Architecture • Setup • Usage • Workflows

---

🚩 Problem

Misinformation spreads 6x faster than truth. Manual fact-checking is slow, filters are easily gamed, and AI verdicts lack accountability.

---

💡 Solution

Claime AI is a security checkpoint for information:

• Multi-Agent AI: Three specialized agents
• Probability Verdicts: e.g. "78% likely FALSE" with reasoning
• PDF Reports: Comprehensive evidence documentation

---

✨ Features

• 🔍 Fact Checker: Async web search
• 🕵️ Forensic Expert: Linguistic & AI detection
• ⚖️ The Judge: Trust-weighted consensus
• ⚡ Parallel execution
• 📄 PDF reports
• 🔐 BYOK (Bring Your Own Key): Secure client-side API key management
• ☁️ Shelby decentralized storage

---

🏗️ Architecture

moveh/
├── backend/         # FastAPI, agents, PDF generation
├── frontend/        # Next.js UI
├── workflows/       # Mermaid diagrams
├── reports/         # PDF reports
└── logs/            # Execution logs

---

🛠️ Tech Stack

Component	Technology	Purpose
AI Orchestration	LangGraph	Multi-agent state machines
LLM	Gemini 2.0 Flash	Fast, cost-effective inference
Web Search	Tavily API	Real-time fact verification
PDF Generation	ReportLab	Professional reports
Decentralized Storage	Shelby Protocol	Evidence preservation

---

📦 Setup

Prerequisites

• Python 3.12+
• Node.js 18+ (for frontend)
• uv or pip
• Shelby CLI (optional for decentralized storage)

Quick Start

1. Clone the repository:

   git clone https://github.com/Sravan1011/Clamify
   

2. Backend Setup:

   cd backend
   
   # Create virtual environment
   python3 -m venv .venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   
   # Install dependencies
   pip install -r requirements.txt
   
   # Run the server
   uvicorn api:app --reload

3. Frontend Setup:

   cd frontend
   
   # Install dependencies
   pnpm install
   
   # Create environment file
   echo "NEXT_PUBLIC_API_BASE_URL=http://localhost:8000" > .env.local
   
   # Run development server
   pnpm dev

🔐 API Configuration (BYOK)

Claime AI uses a Bring Your Own Key architecture. You do not need to configure API keys in the backend for general use.

1. Open the application in your browser (http://localhost:3000).
2. You will be prompted to enter your Google Gemini API Key.
3. (Optional) Enter your Tavily API Key for web search capabilities.
4. Keys are stored securely in your browser's local storage and sent with each request.

Environment Variables (Optional Fallback)

If you are hosting a private instance or developing, you can still set default API keys in the backend .env file. These will be used if the user does not provide keys in the frontend.

Create backend/.env:

# Optional: Fallback keys if not provided by frontend
GOOGLE_API_KEY=your_gemini_api_key
TAVILY_API_KEY=your_tavily_api_key

Where to get API keys:

• GOOGLE_API_KEY: Google AI Studio
• TAVILY_API_KEY: Tavily API

---

🚀 Usage

• Frontend: Visit http://localhost:3000
  • Click the Settings (⚙️) icon in the navbar to manage your API keys.
• API: Visit http://localhost:8000/docs

---

🤝 Contributing

1. Fork and clone
2. Create a feature branch
3. Commit and push
4. Open a PR

---

📄 License

MIT License

---

🛡️ Claime AI - "Trust, but Verify"