Set Up

Learn how to set up Sentry AI Agent Monitoring

Sentry AI Agent Monitoring helps you track and debug AI agent applications using our supported SDKs and integrations. Monitor your complete agent workflows from user interaction to final response, including tool calls, model interactions, and custom logic.

To start sending AI agent data to Sentry, make sure you've created a Sentry project for your AI-enabled repository and follow one of the guides below:

The Sentry JavaScript SDK supports AI agent monitoring through the Vercel AI integration, which works with Node.js and Bun runtimes. This integration automatically captures spans for your AI agent workflows using the AI SDK's built-in telemetry.

Copied
import { Sentry } from "@sentry/node";

// Sentry init needs to be above everything else
Sentry.init({
  tracesSampleRate: 1.0,
  integrations: [
    Sentry.vercelAIIntegration({
      recordInputs: true,
      recordOutputs: true,
    }),
  ],
});

import { generateText } from "ai";
import { openai } from "@ai-sdk/openai";

// Your AI agent function
async function aiAgent(userQuery) {
  const result = await generateText({
    model: openai("gpt-4o"),
    prompt: userQuery,
    experimental_telemetry: {
      isEnabled: true,
      functionId: "ai-agent-main",
    },
  });

  return result.text;
}

The Sentry Python SDK supports OpenAI Agents SDK.

Copied
import sentry_sdk
from sentry_sdk.integrations.openai_agents import OpenAIAgentsIntegration
import agents
from pydantic import BaseModel

sentry_sdk.init(
    dsn="YOUR_DSN",
    traces_sample_rate=1.0,
    send_default_pii=True,  # Include LLM inputs/outputs
    integrations=[
        OpenAIAgentsIntegration(),
    ],
)

# Create your AI agent
my_agent = agents.Agent(
    name="My Agent",
    instructions="You are a helpful assistant.",
    model="gpt-4o-mini",
)

# Your AI agent function
result = await agents.Runner.run(
    my_agent,
    input=user_query,
)
Previous
AI Agents
Next
AI Agents Dashboard
Was this helpful?
Help improve this content
Our documentation is open source and available on GitHub. Your contributions are welcome, whether fixing a typo (drat!) or suggesting an update ("yeah, this would be better").