Guide

Monitor your CrewAI crews in production.

A CrewAI crew runs multiple agents through a sequence of tasks, each calling the model. AgentPing instruments the crew so one record lands per kickoff with the cost and output behind it, across every agent and task.

Track your first AI workflow Jump to setup

Two lines, then every run reports in.

Install the SDK and point it at your CrewAI agent. One record lands per run with the cost, status, latency and output behind it.

shellinstall
pip install agentping
pythonmonitor your agent
import agentping

agentping.init(api_key="apk_…", agent="content-crew")
agentping.instrument_crewai()

# Run your crew exactly as before; every agent and task is captured.
result = crew.kickoff(inputs={"topic": "…"})

One call to instrument_crewai() captures every agent and task in the crew under a single run per kickoff. Full reference is in the docs.

What AgentPing tracks for CrewAI.

  • Cost per crew kickoff, rolled up across every agent and task.
  • A missed-run alert when a scheduled crew stops running.
  • Output quality on the crew's result, scored against your rubric.

See the pillars behind these: Spend, Pulse and Verify.

Does one call cover the whole crew?
Yes. instrument_crewai() captures every agent and task in the crew, so a kickoff that runs five agents is one run with the combined cost and the full trace.
Do I change my crew code?
No. Call instrument_crewai() once after init and run kickoff as you already do. Nothing about the crew definition changes.
Can I attribute cost per task?
The run aggregates cost across the crew. Tag the run with a customer or feature at kickoff and spend is attributed that way in the dashboard.

Monitor your CrewAI agents in minutes.

Add the SDK, send one run, and watch cost, status and quality land in the dashboard.

Track your first AI workflow All monitoring guides