🔄 A2A Protocol Demo Suite

The ultimate interactive demonstration of Agent-to-Agent communication protocols

A comprehensive, educational platform showcasing Google's A2A protocol with real-time visualizations, interactive tutorials, and the groundbreaking Hybrid A2A + MCP Demo showing how agent coordination and tool access work together.

✨ What Makes This Special

🎼 First-ever A2A + MCP integration demo - See horizontal agent coordination meet vertical tool access
🔍 Real protocol traffic - Watch actual JSON-RPC 2.0 messages and SSE streams
🎓 Interactive tutorials - Step-by-step learning with floating controls
🌐 Production ready - Deploy anywhere, works with or without backend
📱 Mobile optimized - Responsive design for all devices

🚀 Quick Start (60 seconds)

# Clone and start
git clone https://github.com/dp-pcs/A2A.git
cd A2A
python -m venv venv && source venv/bin/activate
pip install -r requirements-a2a.txt
python start_complete_demo.py

# Open browser to http://localhost:3000

That's it! 🎉

🎯 Demo Suite Overview

🧠 Smart A2A Demo

Perfect for business audiences - See the value of agent coordination

• AI-powered decision making
• Dynamic workflow orchestration
• Customer-specific solutions
• Business impact demonstration

🔧 A2A Technical Demo

Perfect for developers - Deep dive into protocol mechanics

• Real-time HTTP traffic visualization
• JSON-RPC 2.0 message exchange
• Task Objects and Agent Cards
• SSE streaming infrastructure
• Interactive tutorial mode

🌐 A2A + MCP Hybrid Demo ⭐

Perfect for AI architects - See the future of AI integration

• Horizontal agent coordination (A2A)
• Vertical tool access (MCP)
• Live knowledge integration
• Dual protocol visualization
• Crisis resolution + Research orchestration scenarios

📖 Documentation

Document	Purpose	Audience
Quick Start Guide	Get running in minutes	Everyone
Deployment Guide	Host publicly	DevOps
System Architecture	Technical deep dive	Developers
API Reference	Integration guide	Integrators

🏗️ Repository Structure

A2A/
├── 📁 frontend/              # Demo web applications
│   ├── index.html           # Main demo selector
│   ├── smart-demo.html      # Business value demo
│   ├── technical-demo.html  # Protocol technical demo
│   ├── hybrid-demo.html     # A2A + MCP integration
│   └── js/                  # JavaScript modules
├── 📁 backend/              # A2A protocol implementation
│   ├── agents/              # Individual agent services
│   ├── orchestrator/        # Central coordination
│   ├── agent_registry/      # Service discovery
│   └── shared/              # Common utilities
├── 📁 scripts/              # Start/utility scripts
├── 📁 docs/                 # Comprehensive documentation
├── 📁 examples/             # Integration examples
└── 📁 deploy/               # Deployment configurations

🎮 Usage Modes

🖥️ Local Development (Full Experience)

python start_complete_demo.py

• Real A2A protocol traffic
• Live agent coordination
• Backend services running
• Perfect for learning protocol internals

🌐 Demo Mode (Frontend Only)

cd frontend && python -m http.server 3000

• Simulated protocol responses
• All demos fully functional
• No backend required
• Perfect for presentations

🚀 Production Deployment

See DEPLOYMENT.md for hosting options:

• Netlify (recommended)
• Vercel
• AWS Amplify
• GitHub Pages

🔧 Configuration

Environment Variables (Optional)

# For enhanced AI responses (optional)
export ANTHROPIC_API_KEY="your-key"
export OPENAI_API_KEY="your-key"

# For MCP integration (optional)
export MCP_SERVER_URL="your-mcp-server"

Customization

• Scenarios: Edit demo_scenarios.json
• Agents: Modify files in backend/agents/
• UI: Customize frontend/ files
• Deployment: Update netlify.toml or create amplify.yml

🎓 Learning Path

1. Start with Smart Demo - Understand business value
2. Explore Technical Demo - Learn protocol mechanics
3. Try Hybrid Demo - See integration possibilities
4. Read Documentation - Deep dive into architecture
5. Experiment with Code - Modify and extend

🤝 Contributing

We welcome contributions! See CONTRIBUTING.md for:

• Development setup
• Code standards
• Pull request process
• Feature requests

Quick Contribution Guide

# Fork repo, then:
git clone https://github.com/your-username/A2A.git
cd A2A
python -m venv venv && source venv/bin/activate
pip install -r requirements-a2a.txt
# Make changes, test, submit PR

📋 What You'll Learn

• A2A Protocol Fundamentals - Agent Cards, Task Objects, JSON-RPC 2.0
• Multi-Agent Orchestration - Coordinated workflows and decision making
• Protocol Integration - Combining A2A with MCP for enhanced capabilities
• Real-time Communication - SSE streaming and event-driven patterns
• Production Patterns - Scalable agent coordination architectures

🔧 Troubleshooting

Common Issues

Port already in use?

# Kill existing processes
pkill -f "python.*start_"
# Or use different ports in scripts

Demos not loading?

# Check if frontend server is running
curl http://localhost:3000
# Restart if needed
cd frontend && python -m http.server 3000

Backend not responding?

# Check services
curl http://localhost:8000/.well-known/agents
# Restart complete demo
python start_complete_demo.py

Need help? Open an issue 🆘

📊 Features

✅ Multiple Demo Types - Business, technical, and integration focused
✅ Real Protocol Implementation - Authentic A2A and MCP protocols
✅ Interactive Tutorials - Step-by-step learning with controls
✅ Production Ready - Deploy anywhere with graceful fallbacks
✅ Mobile Optimized - Works on all devices
✅ Comprehensive Docs - Guides for every use case
✅ Open Source - MIT licensed, fork and customize

🙏 Acknowledgments

• Google - A2A Protocol specification
• Anthropic - Model Context Protocol (MCP)
• Trilogy AI Center of Excellence - Knowledge integration examples
• Open Source Community - Inspiration and best practices

📜 License

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

⭐ Star This Repo

If this helped you understand agent coordination protocols, please star this repository! It helps others discover this educational resource.

---

Ready to explore the future of agent communication? 🚀

git clone https://github.com/dp-pcs/A2A.git && cd A2A && python start_complete_demo.py

Open http://localhost:3000 and start your journey!