Guide

Monitor your Pydantic AI agents in production.

Pydantic AI gives you typed, structured agents. AgentPing instruments them so each run reports cost, status and the quality of the structured output, without touching your agent definitions.

Track your first AI workflow Jump to setup

Two lines, then every run reports in.

Install the SDK and point it at your Pydantic AI 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="extractor")
agentping.instrument_pydantic_ai()

# Run the agent as usual; every run is captured.
result = await agent.run("…")

instrument_pydantic_ai() captures each agent run, including tool calls and retries, as one record. Full reference is in the docs.

What AgentPing tracks for Pydantic AI.

  • Cost per agent run, attributed to the customer and feature you tag.
  • A missed-run alert when a scheduled agent stops firing.
  • Output quality, including whether the structured result matched your schema and your rubric.

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

Does it work with structured output?
Yes. The run records the structured result, and you can add deterministic checks (required fields, schema) plus a rubric so a successful-but-wrong output is caught.
Do I rewrite my agents?
No. Call instrument_pydantic_ai() once after init; your agent definitions and run calls stay the same.
Will it slow runs down?
No. Telemetry is off the hot path with a hard timeout and a bounded queue, so your agent runs unchanged if AgentPing is unreachable.

Monitor your Pydantic AI 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