An open-source RAG (Retrieval-Augmented Generation) chatbot framework built on Cloudflare's infrastructure. Deploy your own AI-powered document Q&A system in minutes.

Framework Note: This is an open-source framework designed for evaluation, demonstration, and rapid deployment. By default, it runs in permissive mode to facilitate testing and integration. See Security Configuration for production hardening options.

• 🚀 One-Command Deployment - Deploy to your Cloudflare account with a single script
• 🌍 Multi-Language Support - EN, DE, FR, IT with automatic language detection
• 📚 Smart Document Organization - Category and product-based document structure
• 🤖 Multiple AI Models - Workers AI, OpenAI, and Anthropic support
• 🎨 Embeddable Widget - Drop into any website with two lines of code
• 📁 R2 Storage Integration - Scalable document storage
• 🔍 AutoRAG Indexing - Automatic document indexing and retrieval
• 🎮 Interactive Playground - Test configurations before deployment

• Cloudflare account (free tier works)
• Node.js 18+ and npm (Node 20+ recommended for all features)
• Git

# 1. Clone and setup
git clone https://github.com/arnemoor/rag-chatbot.git
cd rag-chatbot

# 2. Follow the setup guide
# See docs/SETUP.md for detailed instructions

📖 Complete Setup Guide →

Manual steps required:

1. Create AutoRAG instance in Cloudflare Dashboard
2. Create API Token with proper permissions
3. Copy credentials to .env file

Automated by scripts:

• R2 bucket creation
• Worker deployment
• Widget deployment
• URL configuration

After deployment, you'll receive:

• Your Worker API URL: https://your-worker.workers.dev
• Your Widget URL: https://your-widget.pages.dev
• Integration code for your website
• Direct links to playground, demo, and tools

Add the chatbot to any website:

<script src="https://your-widget.pages.dev/autorag-widget.min.js"></script>
<autorag-widget></autorag-widget>

With custom configuration:

<autorag-widget 
  language="en"
  category="fiction"
  theme="dark"
  position="bottom-right">
</autorag-widget>

Your Cloudflare Account
├── Worker (API Backend)
│   ├── AutoRAG Integration
│   ├── Document Processing
│   └── Chat Endpoints
├── Pages (Widget Frontend)
│   ├── Embeddable Widget
│   ├── Standalone Chat
│   └── Admin Tools
└── R2 Storage
    └── Document Storage

rag-chatbot/
├── worker/              # Backend API (Cloudflare Worker)
├── widget/              # Embeddable widget (Cloudflare Pages)  
├── scripts/             # Deployment and utility scripts
├── docs/                # Documentation
├── examples/            # Configuration examples
├── sample-documents/    # Sample documents for testing
└── README.md           # Quick start guide

1. Account ID: Found in Cloudflare Dashboard → Right sidebar
2. API Token: Create at Profile → API Tokens with permissions:
   • AutoRAG: Read/Write
   • R2: Read/Write
   • Workers: Edit
   • Pages: Edit

Add to .env for external models:

OPENAI_API_KEY=your-key
ANTHROPIC_API_KEY=your-key

• Documentation Hub - All documentation
• Quick Start Guide - Get started in 5 minutes
• Document Management - Upload and sync documents
• API Token Setup - Configure permissions
• Security Configuration - Security options and production hardening
• Troubleshooting - Common issues

• Customer Support - AI-powered support chatbot
• Documentation Assistant - Interactive documentation helper
• Knowledge Base - Searchable knowledge base with AI
• Educational Platform - Course material Q&A system
• Internal Tools - Employee handbook assistant

# Install all dependencies
./scripts/install-dependencies.sh

# Start Worker locally
cd worker
npm run dev

# Build Widget
cd widget
npm run build -- --watch

# Run all tests
./scripts/test-all.sh

Upload documents to R2:

# Upload sample documents
./scripts/upload-library-documents.sh

# Or upload your own files
cd worker
npx wrangler r2 object put your-bucket/document.pdf --file=../path/to/document.pdf

Documents are automatically indexed by AutoRAG after upload.

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

MIT License - See LICENSE file

• Documentation: Documentation Hub
• Issues: Report issues in your fork
• Discussions: Use your preferred communication channel

Built with:

• Cloudflare Workers
• Cloudflare AutoRAG
• Cloudflare R2
• Cloudflare Pages

Note: This is an open source template. When you deploy it, all services run on YOUR Cloudflare infrastructure with YOUR URLs. No data or requests go to the original authors.