VOOZH about

URL: https://docs.datadoghq.com/llm_observability/monitoring/agent_monitoring/

⇱ Agent Monitoring

For AI agents: A markdown version of this page is available at https://docs.datadoghq.com/llm_observability/monitoring/agent_monitoring.md. A documentation index is available at /llms.txt.

Agent Monitoring

This product is not supported for your selected Datadog site. ().

Overview

Agent Observability allows you to monitor, troubleshoot, and improve your agentic applications. With the Agent Observability SDK for Python, you can monitor the health and quality of your single- or multi-agentic systems built on OpenAI Agents SDK, LangGraph, or CrewAI.

For your agentic applications, Agent Observability allows you to:

  • Monitor error rate, latency build up, or cost
  • Visualize agent decisions, such as tools used or agent a task was handed off to
  • Trace and troubleshoot end-to-end requests of agent executions

Further reading