+ P Evaluate for OpenAI Agents SDK agents
Capture every span the OpenAI Agents SDK already wraps its own run in — agent, generation, function, guardrail, and handoff spans — as structured trace data, not a second tracing layer bolted on top.
Prefactor evaluates your OpenAI Agents SDK agents — score outcome quality against the captured spans, track drift by comparing custom spans across versions and environments, and classify every run's risk on two axes: data sensitivity and action consequence.
How to add evaluate to OpenAI Agents SDK
Install the SDK
Add prefactor-core to your environment.
Register & create spans
Register the agent instance and record spans around your OpenAI Agents SDK run.
Spans flow to Prefactor
Structured trace data lands in Prefactor as your agent runs.
# pip install prefactor-core
import os
from prefactor_core import PrefactorCoreClient, PrefactorCoreConfig
from prefactor_http import HttpClientConfig
config = PrefactorCoreConfig(
http_config=HttpClientConfig(
api_url="https://app.prefactorai.com",
api_token=os.environ["PREFACTOR_API_TOKEN"],
)
)
client = PrefactorCoreClient(config)
await client.initialize()
# then instrument your OpenAI Agents SDK run with spans — see docs.prefactor.aiShown with the Prefactor SDK — a first-class, working integration today.
How the OpenAI Agents SDK integration actually works
- The SDK already wraps every part of a run in its own span type — agent_span(), generation_span(), function_span(), guardrail_span(), handoff_span() — Prefactor's spans map directly onto these, so nothing needs re-deriving from raw call data.
- Guardrails are the SDK's own real-time enforcement point: they run input/output validation in parallel with the agent and fail fast when a check doesn't pass — the same mechanism a Prefactor runtime policy runs through to hold or block a run before it produces a final response.
- Beyond auto-captured spans, use withSpan to record any custom step you define — an API call, a quality check, a business action.
What Prefactor captures for OpenAI Agents SDK agents
Outcome quality
Score agent outputs against your own criteria, task by task, using the continuous run history as the record to measure against.
Drift detection
Compare custom spans for the same operation across versions and environments — a model update or prompt change shows up as a difference between spans, before it shows up as a quality drop.
Risk classification
Every run scored on data sensitivity and action consequence, weighted into a Low / Medium / High / Critical classification.
Data tagging
Tag PII and other sensitive fields once in the schema; every run carrying that tag becomes searchable — the record enforcement acts on.
Trace anything — not just LLM calls
The SDK captures LLM, tool, and agent spans automatically. With withSpan you wrap any operation in your own span type — an API call, a database query, a quality check, a business action — each with its own payload and schema. It all flows into the same Observe, Evaluate, and cost views.
import { withSpan } from "@prefactor/core";
// Wrap ANY operation in a span you define — an API call, a quality
// check, a business action — with its own spanType, inputs and schema
await withSpan(
{
name: "research competitor",
spanType: "research_competitor",
inputs: { competitor },
},
async () => {
const results = await search(competitor); // your tool / API calls
return summarize(results); // captured as one span
},
);Nest spans to capture business-level actions, and start with permissive schemas you tighten over time. Instrumentation strategy →
An example run, span by span
Illustrative — a single OpenAI Agents SDK run as nested spans.
Illustrative example.
Manual logging vs DIY vs Prefactor
| Capability | Manual | DIY OpenTelemetry | Prefactor |
|---|---|---|---|
| LLM / tool / agent spans | Hand-rolled | ✓ build it | ✓ via the SDK |
| Token usage captured per call | — | Build it | ✓ |
| Configurable capture & sampling | — | Partial | ✓ |
| Hosted Admin UI (agents, instances) | — | — | ✓ |
| Risk profiles & audit trail | — | — | ✓ |
OpenAI Agents SDK evaluate — FAQ
Does evaluation reuse the same OpenAI Agents SDK integration?+
Yes — evaluation runs on the spans the SDK already records for OpenAI Agents SDK, so there is one integration, not a separate one for evaluation.
What is quality drift and how is it detected for OpenAI Agents SDK agents?+
Drift is a change in behaviour over time — after a model update, a prompt change, or a data shift. Because a custom span records the same OpenAI Agents SDK operation every time it runs, this week's spans can be compared directly against last week's, or against a different version or environment.
How is risk scored for OpenAI Agents SDK runs?+
On two axes — data sensitivity (what kind of data the run touched) and action consequence (what the agent did with it) — combined via configurable weights into a Low/Medium/High/Critical classification per run.
Can I tag and find PII across OpenAI Agents SDK runs?+
Yes — tag the fields that matter once in your agent's data schema, and every run carrying that tag becomes searchable across your whole OpenAI Agents SDK deployment, not just the run it was tagged in.
Do I need a dedicated package for OpenAI Agents SDK?+
You can instrument OpenAI Agents SDK today with the framework-agnostic prefactor-core SDK; a dedicated package can be added on request.
What does Prefactor capture from OpenAI Agents SDK?+
Prefactor records agent runs, handoffs between agents, tool calls and LLM calls as structured, timestamped spans — so every OpenAI Agents SDK run is captured as trace data you can reconstruct, search and export end to end.
Does Prefactor add latency or change how OpenAI Agents SDK runs?+
No. Observability capture is designed to stay off your agent's critical path, so it doesn't alter your OpenAI Agents SDK logic or your users' responses. The only part that acts inline is the optional runtime guardrails you enable per agent — by design, so a high-risk or low-confidence action can be held for human approval before it executes.
Can I evaluate agents built with OpenAI Agents SDK and catch regressions?+
Yes. Once runs are captured, eval suites score quality and groundedness on real traffic, drift detection flags behaviour changes after deployment, and versioned eval history catches regressions before they ship — the observe → evaluate → improve loop applied to your OpenAI Agents SDK agents.
Keep going
Evaluate for other frameworks
LangChain →CrewAI →LangGraph →Claude Agent SDK →Microsoft AutoGen →LlamaIndex →See it on your OpenAI Agents SDK agents
Book a 15-minute setup and our team gets you evaluate in production.