OpenAI agent telemetry

What to record and check

Link traces to the final task result

Keep detailed steps in your tracing tool. Send one event with the final result so you can query it alongside cost, reliability, and product usage.

1
Agent run
Start with an approved run identifier, workflow, model, and prompt version.
2
Tools and retries
Retain detailed steps in tracing and categorize failures without copying payloads.
3
Terminal outcome
Emit success, failure, cancellation, or human handoff with cost and duration.
4
SQL decision
Compare accepted outcomes, failures, and unit economics by feature and release.

Before you start

The OpenAI Agents SDK and telemetry-sh packages initialized in a server runtime
A documented workflow name and final success, failure, and handoff outcomes
A decision about which prompt, tool-input, and output content must remain excluded

Delivery setup

Install and initialize server-side

Import telemetry-sh in server-only code and initialize it once with process.env.TELEMETRY_API_KEY. Keep ingestion credentials out of browser bundles, client-visible environment variables, source control, logs, and exception messages.

openai-agents-install

npm installation

bash

npm install telemetry-sh

1Create one reusable server-side client. Set its timeout and retry limit.
2Log an event when the operation succeeds, fails, retries, or times out.
3Send test events with known results and inspect the stored rows before enabling an alert.

Snippet

Start with one structured event

Add this shape where the workflow completes, fails, or retries. Then build the dashboard from real fields.

openai-agents

OpenAI agent telemetry event

javascript

await telemetry.log("agent_tool_called", {
  run_id: runId,
  workflow: "support_resolution",
  agent_name: "support_agent",
  model: "gpt-4.1",
  tool_name: "lookup_order",
  status: "success",
  latency_ms: 842,
  retry_count: 0,
  estimated_cost_usd: 0.018,
  prompt_version: "support-v3",
  release: process.env.APP_RELEASE,
});

Event schema

run_id, workflow, agent_name, model, and prompt_version

status, duration_ms, retry_count, total_tokens, and estimated_cost_usd

tool_name, tool_status, human_handoff, reviewer_outcome, and release

Check your setup

Checkpoint 1

Emit one compact product outcome when the complete run finishes; keep SDK traces for step-level debugging instead of copying every span into a second system.

Checkpoint 2

Use a shared run_id or approved trace_id to connect separate tool events without storing tool arguments or results.

Checkpoint 3

Exercise tool failure, guardrail rejection, handoff, cancellation, and exporter-flush behavior before relying on completion-rate dashboards.

Verification

Prove the event arrived

Run this after exercising known success and failure cases. Replace the fallback table name if your final event contract differs from the snippet.

openai-agents-verification

OpenAI agent telemetry verification query

sql

SELECT *
FROM agent_tool_called
ORDER BY timestamp_utc DESC
LIMIT 20;

Confirm one terminal row per logical outcome, with the expected status, identifiers, units, and UTC time.

Inspect the inferred schema and verify that retries do not change field types or generate a new logical event ID.

Search the stored fields for credentials, raw payloads, prompts, private content, and unbounded error messages.

Exercise a provider timeout, ingestion rejection, and process shutdown before treating the dashboard as complete.

Implementation references

Review the event contract, data-safety guidance, and upstream primary documentation before enabling a new production path.

AI agent quality evaluation Redacting sensitive data Nested tool-call SQL OpenAI Agents SDK tracing

Where to log

Keep the outcome event small and recoverable

This pattern provides

Record the outcome as an event you can query with SQL.
Stable fields for dashboards, alerts, and cross-event correlation.
Test events for checking success, failure, retries, and timeouts.

This pattern does not provide

An OTLP exporter, automatic collection pipeline, or replacement for detailed traces and diagnostic logs.
Exactly-once delivery merely because the payload contains an event ID.
Permission to collect raw provider payloads, user content, credentials, or regulated data.

Delivery and idempotency Batching and shutdown Troubleshoot ingestion

Example event schemas

Check what each event records, when to send it, and which field types it needs. Review the example payload and privacy checklist before using it in production.

AI operations10 fields