R2E - Report to Earn: Decentralized Misinformation Detection Platform

A comprehensive Web3 platform that incentivizes users to report and verify misinformation through AI-powered content analysis, blockchain-based rewards, and community-driven verification.

Table of Contents

• Overview
• Architecture
• Tech Stack
• Project Structure
• Features
• Getting Started
• Deployment
• API Documentation
• Smart Contracts
• Frontend
• AI Verification System
• Contributing
• License

Overview

R2E (Report to Earn) is a decentralized platform that combats misinformation by rewarding users for accurate reporting and verification. The platform combines advanced AI verification systems, blockchain technology, and community-driven fact-checking to create a trustworthy information ecosystem.

Key Components

• AI Verification System: Multi-agent AI system using 5 different models for content verification
• Multi-Model AI Verification: Advanced ensemble of 5 specialized AI models working in parallel for comprehensive content analysis
• Blockchain Integration: Algorand-based smart contracts for transparent reward distribution
• Web Frontend: Modern Next.js interface for user interaction
• Content Scraping: Automated content extraction from multiple social media platforms
• Dynamic Rewards: Algorithm-based reward calculation based on content popularity and verification accuracy
• Transparent Reporting: Public verification reports with detailed reasoning and evidence
• Multi-language Support: Global accessibility with support for multiple languages and regions

Architecture

Tech Stack

Frontend

• Framework: Next.js 14.2.16
• Language: TypeScript
• Styling: Tailwind CSS
• Animations: Framer Motion
• Blockchain: Algorand SDK (@perawallet/connect)
• State Management: React Context

Backend

• Framework: FastAPI
• Language: Python 3.8+
• AI/ML: LangGraph, LangChain
• Image Processing: Pillow, Groq Vision
• Web Scraping: Selenium, BeautifulSoup4
• HTTP Client: httpx, requests

Blockchain

• Network: Algorand Testnet/Mainnet
• Smart Contracts: TEAL (Transaction Execution Approval Language)
• SDK: Algorand SDK (algosdk)
• Wallet Integration: Pera Wallet

AI Models

• Text Analysis: 5 Free OpenRouter Models (128K-164K context limits):
  • NVIDIA Nemotron Nano 9B v2 (128K context)
  • Kimi Dev 72B (164K context - highest)
  • Z-AI GLM 4.5 Air (131K context)
  • Mistral Small 3.2 24B Instruct (131K context)
  • DeepSeek R1 Qwen3 8B (131K context)
• Vision Processing: Groq Vision Models (Llama 4)
• Consensus: Custom multi-agent decision system

Infrastructure

• Containerization: Docker
• Deployment: Vercel (Frontend), Railway/Render (Backend)
• Environment: Node.js, Python

Project Structure

r2e/
├── ai-verify/                    # AI Verification Backend
│   ├── main.py                   # FastAPI server entry point
│   ├── ai_verification_system.py # Multi-agent AI system
│   ├── content_scraper.py        # Social media content scraper
│   ├── image_processor.py        # Image analysis and processing
│   ├── requirements.txt          # Python dependencies
│   ├── Dockerfile               # Container configuration
│   └── nixpacks.toml           # Deployment configuration
├── frontend/                     # Next.js Web Application
│   ├── app/                     # App Router pages
│   │   ├── page.tsx            # Landing page
│   │   ├── verify/             # Verification interface
│   │   └── community/          # Community dashboard
│   ├── components/             # React components
│   │   ├── ui/                 # Reusable UI components
│   │   ├── SmartContractVerificationFlow.tsx
│   │   ├── Leaderboard.tsx
│   │   └── RandomLiveFeed.tsx
│   ├── contexts/               # React contexts
│   ├── hooks/                  # Custom React hooks
│   ├── lib/                    # Utility functions
│   └── config/                 # Configuration files
├── smart-contracts/            # Algorand Smart Contracts
│   ├── approval.teal          # Main contract logic
│   ├── clear.teal             # Contract cleanup
│   ├── deploy-contract.js     # Deployment script
│   └── deploy-creator.js      # Creator account setup
└── README.md                   # This file

Features

Core Features

• Multi-Platform Content Verification: Support for Twitter, Instagram, Reddit, YouTube, and generic websites
• AI-Powered Analysis: 5 different AI models working in consensus
• Dynamic Reward System: Rewards based on content popularity and verification accuracy
• Blockchain Integration: Transparent reward distribution via Algorand smart contracts
• Real-time Processing: Fast content analysis and verification
• Community Dashboard: Leaderboards and community statistics

Advanced Features

• Image Analysis: Advanced image processing with manipulation detection
• Content Scraping: Automated extraction from social media platforms
• Wallet Integration: Seamless Algorand wallet connection
• Responsive Design: Mobile-first, modern UI/UX
• Real-time Updates: Live feed of verification activities

Getting Started

Prerequisites

• Node.js 18+ and npm/pnpm
• Python 3.8+
• Algorand wallet (Pera Wallet recommended)
• API keys for OpenRouter and Groq

Installation

1. Clone the repository

   git clone https://github.com/your-username/r2e.git
   cd r2e

2. Install Frontend Dependencies

   cd frontend
   npm install
   # or
   pnpm install

3. Install Backend Dependencies

   cd ../ai-verify
   pip install -r requirements.txt

4. Install Smart Contract Dependencies

   cd ../smart-contracts
   npm install

Environment Setup

1. Backend Environment Variables Create ai-verify/.env:

   OPENROUTER_API_KEY=your_openrouter_api_key
   GROQ_API_KEY=your_groq_api_key

2. Frontend Environment Variables Create frontend/.env.local:

   NEXT_PUBLIC_ALGOD_TOKEN=your_algod_token
   NEXT_PUBLIC_ALGOD_SERVER=https://testnet-api.algonode.cloud
   NEXT_PUBLIC_ALGOD_PORT=443
   NEXT_PUBLIC_REWARD_CONTRACT_APP_ID=your_deployed_app_id

3. Smart Contract Configuration Create smart-contracts/.env:

   ALGOD_TOKEN=your_algod_token
   ALGOD_SERVER=https://testnet-api.algonode.cloud
   ALGOD_PORT=443

Running the Application

1. Start the AI Verification Backend

   cd ai-verify
   python main.py

   Server runs on http://localhost:8000

2. Start the Frontend Development Server

   cd frontend
   npm run dev
   # or
   pnpm dev

   Application runs on http://localhost:3000

3. Deploy Smart Contracts (Optional for development)

   cd smart-contracts
   node deploy-contract.js

Deployment

Frontend Deployment (Vercel)

1. Connect to Vercel

   cd frontend
   vercel login
   vercel --prod

2. Set Environment Variables

   • NEXT_PUBLIC_ALGOD_TOKEN
   • NEXT_PUBLIC_ALGOD_SERVER
   • NEXT_PUBLIC_ALGOD_PORT
   • NEXT_PUBLIC_REWARD_CONTRACT_APP_ID

Backend Deployment (Railway/Render)

1. Using Railway

   cd ai-verify
   railway login
   railway init
   railway up

2. Using Render

   • Connect GitHub repository
   • Set build command: pip install -r requirements.txt
   • Set start command: python main.py
   • Add environment variables

Smart Contract Deployment

1. Deploy to Algorand Testnet

   cd smart-contracts
   node deploy-contract.js

2. Deploy to Algorand Mainnet

   • Update network configuration
   • Ensure sufficient ALGO balance
   • Run deployment script

API Documentation

Base URL

• Development: http://localhost:8000
• Production: https://your-backend-url.com

Endpoints

Health Check

GET /health

Verify Content

POST /verify
Content-Type: application/json

{
  "content_url": "https://example.com/content",
  "content_text": "Content to verify...",
  "content_images": ["image1.jpg", "image2.jpg"]
}

Scrape and Verify

POST /scrape-and-verify
Content-Type: application/json

{
  "url": "https://twitter.com/username/status/1234567890"
}

Get Available Models

GET /agents

Response:

{
  "models": [
    {
      "name": "NVIDIA Nemotron Nano 9B v2",
      "model_id": "nvidia/nemotron-nano-9b-v2:free",
      "context_limit": "128,000 tokens",
      "created": "September 5, 2025",
      "pricing": "$0/M input tokens, $0/M output tokens",
      "specialty": "Efficient reasoning and fact-checking",
      "strengths": "Fast processing, good at logical analysis"
    },
    {
      "name": "Kimi Dev 72B",
      "model_id": "moonshotai/kimi-dev-72b:free",
      "context_limit": "163,840 tokens",
      "created": "January 20, 2025",
      "pricing": "$0/M input tokens, $0/M output tokens",
      "specialty": "Latest reasoning model",
      "strengths": "State-of-the-art analysis, cutting-edge capabilities"
    },
    {
      "name": "Z-AI GLM 4.5 Air",
      "model_id": "z-ai/glm-4.5-air:free",
      "context_limit": "131,072 tokens",
      "created": "July 25, 2025",
      "pricing": "$0/M input tokens, $0/M output tokens",
      "specialty": "Chinese-developed model with unique perspective",
      "strengths": "Different cultural context, alternative viewpoints"
    },
    {
      "name": "Mistral Small 3.2 24B Instruct",
      "model_id": "mistralai/mistral-small-3.2-24b-instruct:free",
      "context_limit": "131,072 tokens",
      "created": "June 20, 2025",
      "pricing": "$0/M input tokens, $0/M output tokens",
      "specialty": "High-quality instruction following",
      "strengths": "Detailed analysis, comprehensive reasoning"
    },
    {
      "name": "Dolphin Mistral 24B Venice Edition",
      "model_id": "cognitivecomputations/dolphin-mistral-24b-venice-edition:free",
      "context_limit": "131,072 tokens",
      "created": "May 29, 2025",
      "pricing": "$0/M input tokens, $0/M output tokens",
      "specialty": "Advanced reasoning capabilities",
      "strengths": "Complex logical analysis, pattern recognition"
    }
  ],
  "workflow": "Parallel analysis with consensus-based decision-making",
  "total_models": 5,
  "consensus_method": "Majority voting with confidence weighting",
  "total_context_range": "128K-164K tokens"
}

Response Format

{
  "success": true,
  "result": {
    "final_decision": "authentic|fake|uncertain",
    "confidence": 0.85,
    "consensus_score": 0.75,
    "popularity_score": 0.75,
    "dynamic_reward": 0.1875,
    "individual_decisions": [...],
    "group_reasoning": "Detailed explanation..."
  },
  "scraped_content": {
    "platform": "twitter",
    "url": "https://twitter.com/username/status/1234567890",
    "content_text": "Scraped text content...",
    "content_images": ["image1.jpg", "image2.jpg"],
    "author": {"username": "username", "full_name": "Full Name"},
    "engagement": {"likes": 100, "comments": 50, "shares": 25},
    "metadata": {...}
  }
}

Smart Contracts

Contract Overview

The R2E smart contract manages reward distribution and user opt-in functionality on the Algorand blockchain.

Key Features

• User Opt-in: Users must opt into the contract to claim rewards
• One-time Claims: Each user can claim rewards only once
• Automatic Distribution: Contract automatically sends ALGO to users
• Creator Control: Only contract creator can update/delete contract

Contract Functions

• opt_in(): User opts into the contract
• claim_reward(): User claims their reward
• update_contract(): Creator updates contract (admin only)
• delete_contract(): Creator deletes contract (admin only)

Deployment Process

1. Compile TEAL programs
2. Create application transaction
3. Sign and submit transaction
4. Wait for confirmation
5. Extract application ID

Frontend

Key Components

SmartContractVerificationFlow

Main verification interface handling the complete verification process.

Leaderboard

Community dashboard showing top contributors and statistics.

RandomLiveFeed

Real-time feed of verification activities and community updates.

Wallet Integration

• Pera Wallet: Primary wallet integration
• Algorand SDK: Blockchain interaction
• Transaction Management: Automatic transaction handling

State Management

• React Context: Global state management
• Wallet Context: Wallet connection state
• Theme Provider: Dark/light mode support

AI Verification System

Multi-Agent Architecture

The system uses 5 different AI models working in parallel:

1. NVIDIA Nemotron Nano 9B v2 (nvidia/nemotron-nano-9b-v2:free)

   • Context Limit: 128,000 tokens
   • Created: September 5, 2025
   • Pricing: $0/M input tokens, $0/M output tokens
   • Specialty: Efficient reasoning and fact-checking
   • Strengths: Fast processing, good at logical analysis

2. Kimi Dev 72B (moonshotai/kimi-dev-72b:free)

   • Context Limit: 163,840 tokens (highest context)
   • Created: January 20, 2025
   • Pricing: $0/M input tokens, $0/M output tokens
   • Specialty: Latest reasoning model
   • Strengths: State-of-the-art analysis, cutting-edge capabilities

3. Z-AI GLM 4.5 Air (z-ai/glm-4.5-air:free)

   • Context Limit: 131,072 tokens
   • Created: July 25, 2025
   • Pricing: $0/M input tokens, $0/M output tokens
   • Specialty: Chinese-developed model with unique perspective
   • Strengths: Different cultural context, alternative viewpoints

4. Mistral Small 3.2 24B Instruct (mistralai/mistral-small-3.2-24b-instruct:free)

   • Context Limit: 131,072 tokens
   • Created: June 20, 2025
   • Pricing: $0/M input tokens, $0/M output tokens
   • Specialty: High-quality instruction following
   • Strengths: Detailed analysis, comprehensive reasoning

5. Dolphin Mistral 24B Venice Edition (cognitivecomputations/dolphin-mistral-24b-venice-edition:free)

   • Context Limit: 131,072 tokens
   • Created: May 29, 2025
   • Pricing: $0/M input tokens, $0/M output tokens
   • Specialty: Advanced reasoning capabilities
   • Strengths: Complex logical analysis, pattern recognition

Consensus Mechanism

• Individual Analysis: Each model analyzes content independently
• Confidence Scoring: Models provide confidence scores
• Majority Voting: Final decision based on majority consensus
• Weighted Scoring: Confidence scores influence final decision

Dynamic Reward System

• Base Reward: 0.05 ALGO for detecting fake content
• Popularity Multiplier: 1.0x to 5.0x based on content popularity
• Maximum Reward: 1.0 ALGO (capped)
• Formula: Dynamic Reward = Base Fee × Popularity Multiplier

Reward Calculation Algorithm

The dynamic reward system uses a sophisticated algorithm to calculate rewards based on multiple factors:

Base Components

• Detection Reward: 0.05 ALGO for successfully identifying fake content
• Verification Bonus: Additional 0.02 ALGO for high-confidence verification
• Community Consensus: 0.01 ALGO bonus when multiple users report the same content

Popularity Metrics

The popularity multiplier is calculated using:

• Social Media Engagement: Likes, shares, comments, views
• Content Virality Score: Rate of spread across platforms
• Temporal Relevance: How recent the content is
• Platform Weight: Different platforms have different impact factors

Multiplier Calculation

Image Processing

• Vision Models: Llama 4 Scout and Maverick
• OCR: Text extraction from images
• Manipulation Detection: AI-generated content detection
• Multi-image Support: Process up to 5 images per request

Contributing

Development Setup

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests if applicable
5. Submit a pull request

License

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

Support

For support and questions:

• GitHub Issues: Create an issue
• Discord: Join our community
• Email: support@r2e.com

Roadmap

Phase 1 (Current)

• Basic AI verification system
• Algorand smart contract integration
• Web frontend
• Multi-platform content scraping

Phase 2 (Next)

• Mobile application
• Advanced analytics dashboard
• Community governance
• Cross-chain support

Phase 3 (Future)

• Decentralized storage integration
• Advanced AI model training
• Global expansion
• Enterprise solutions

---

Built with ❤️ by the R2E Team

Fighting misinformation, one verification at a time.