CLI Tracing Setup

Automatic tracing code generation for your AI agents. The CLI analyzes your codebase and generates the new simplified tracing approach automatically.

Prerequisites: Node.js and a Handit.ai account. The CLI handles all SDK installation and configuration automatically.

Quick Setup

Step 1: Install the CLI

terminal


npm install -g @handit.ai/cli

Step 2: Run Setup

terminal


handit-cli setup

The CLI will:

Connect your Handit.ai account
Analyze your codebase to find agent functions
Generate tracing code using the new simplified approach
Configure your API key and environment

What the CLI Generates

The CLI creates tracing code using the new simplified approach:

Configuration Files

Agent Wrapper Code

The CLI generates wrappers for your agent functions using the new simplified approach:

Key point: The CLI only adds tracing to your main agent functions (entry points), not to individual helper functions. Everything inside the traced function gets captured automatically.

Managing Your Setup

Update Configuration


# Re-run setup to update configuration
handit-cli setup

When to re-run setup:

After adding new agent functions
When changing your project structure
If tracing stops working
To update to latest approach

Verify Setup

✅ Check your dashboard: Go to dashboard.handit.ai - you should see:

Agent traces appearing when you run your AI
Complete execution flows captured automatically
Performance metrics and timing data

Integration Examples

The CLI adapts to different codebase structures:

Class-based Agents


# Your original code (unchanged)
class CustomerAgent:
    async def process_request(self, msg):
        return await self.handle_request(msg)
 
# CLI generates traced wrapper
class TracedCustomerAgent(CustomerAgent):
    @tracing(agent="customer-agent")
    async def process_request(self, msg):
        return await super().process_request(msg)

API Endpoints


# FastAPI integration (CLI-generated)
from handit_ai import tracing
 
@app.post("/chat")
@tracing(agent="chat-api")
async def chat_endpoint(message: str):
    return await process_chat(message)

Multiple Agents


# CLI generates separate traced versions
@tracing(agent="customer-service")
async def handle_customer_query(message):
    return await process_customer_request(message)
 
@tracing(agent="technical-support")  
async def handle_technical_query(message):
    return await process_technical_request(message)

Next Steps

View your traces in the Dashboard
Set up Evaluation to assess trace quality
Configure Autonomous Fixes for continuous improvement

Your AI is now fully observable! The CLI has generated everything needed for your autonomous engineer to monitor your AI and create improvements.

Troubleshooting

CLI setup issues: Ensure Node.js is installed and you have a valid Handit.ai account. Try running handit-cli setup again.

No traces appearing: Verify your API key is configured correctly and your traced functions are being called.

Performance concerns: The new simplified approach has minimal overhead. Contact support for high-volume optimization.

For more help, visit our Support page or join our Discord community .