LANGFUSE_SETUP.md

Langfuse Monitoring Setup

This guide helps you set up Langfuse monitoring for your LangGraph application to track LLM responses, costs, and performance.

🚀 Quick Start

1. Start Langfuse Services

# Start Langfuse database and server
docker-compose up -d langfuse-db langfuse-server

# Or use the setup script
python setup_langfuse.py

2. Configure Langfuse

1. Open http://localhost:3000 in your browser
2. Create an account or sign in
3. Create a new project (or use the default one)
4. Go to Settings → API Keys
5. Copy the Secret Key and Public Key

3. Update Environment Variables

Add to your .env file:

LANGFUSE_SECRET_KEY=sk-lf-your-secret-key-here
LANGFUSE_PUBLIC_KEY=pk-lf-your-public-key-here
LANGFUSE_HOST=http://localhost:3000

4. Start the Application

# Start all services including your app
docker-compose up --build

# Or start just your app if Langfuse is already running
docker-compose up langgraph-agenticai

📊 What Gets Monitored

• LLM Requests & Responses: All interactions with Groq models
• Token Usage: Input and output tokens for cost calculation
• Response Times: Latency metrics for performance monitoring
• User Sessions: Track user interactions across conversations
• Tool Usage: Monitor MCP and other tool executions
• Error Tracking: Capture and analyze failures

🎯 Features

In-App Monitoring

• ✅ Real-time monitoring status in sidebar
• ✅ Direct link to Langfuse dashboard
• ✅ Session tracking across conversations
• ✅ Automatic cost tracking

Langfuse Dashboard

• 📈 Traces: Detailed view of each conversation
• 💰 Costs: Token usage and cost breakdown
• ⚡ Performance: Response time analytics
• 🔍 Search: Find specific interactions
• 📊 Analytics: Usage patterns and trends

🔧 Configuration Options

Environment Variables

Variable	Description	Default
LANGFUSE_SECRET_KEY	Your Langfuse secret key	Required
LANGFUSE_PUBLIC_KEY	Your Langfuse public key	Required
LANGFUSE_HOST	Langfuse server URL	http://localhost:3000
TELEMETRY_ENABLED	Enable Langfuse telemetry	true

Docker Compose Services

• langfuse-db: PostgreSQL database for Langfuse
• langfuse-server: Langfuse web application
• langgraph-agenticai: Your main application

🛠️ Troubleshooting

Langfuse Not Starting

# Check service logs
docker-compose logs langfuse-server
docker-compose logs langfuse-db

# Restart services
docker-compose restart langfuse-server langfuse-db

Connection Issues

1. Verify environment variables are set correctly
2. Check that Langfuse is accessible at http://localhost:3000
3. Ensure API keys are valid and have proper permissions

Monitoring Not Working

1. Check the sidebar for monitoring status
2. Verify Langfuse credentials in .env file
3. Look for error messages in the Streamlit app

📚 Advanced Usage

Custom Traces

from src.langgraphagenticai.monitoring.langfuse_integration import langfuse_manager

# Create custom trace
trace = langfuse_manager.create_trace(
    name="custom_operation",
    user_id="user123",
    session_id="session456"
)

Cost Analysis

Access detailed cost breakdowns in the Langfuse dashboard:

1. Go to Analytics → Usage
2. Filter by model, user, or time period
3. Export data for further analysis

🔗 Useful Links

• Langfuse Documentation
• Langfuse GitHub
• LangChain Integration

🆘 Support

If you encounter issues:

1. Check the troubleshooting section above
2. Review Docker Compose logs
3. Consult Langfuse documentation
4. Open an issue in the project repository