OpenAI Agents SDK Cookbook

This guide shows you how to create and evaluate agents with Phoenix to improve performance.

We’ll go through the following steps:

• Create an agent using the OpenAI agents SDK

• Trace the agent activity

• Create a dataset to benchmark performance

• Run an experiment to evaluate agent performance using LLM as a judge

• Learn how to evaluate traces in production

Google Colab

colab.research.google.com

Notebook Walkthrough

We will go through key code snippets on this page. To follow the full tutorial, check out the full notebook.

Create your first agent with the OpenAI SDK

Here we’ve setup a basic agent that can solve math problems. We have a function tool that can solve math equations, and an agent that can use this tool.

We’ll use the Runner class to run the agent and get the final output.

from agents import Runner, function_tool


@function_tool
def solve_equation(equation: str) -> str:
    """Use python to evaluate the math equation, instead of thinking about it yourself.

    Args:
       equation: string which to pass into eval() in python
    """
    return str(eval(equation))

from agents import Agent

agent = Agent(
    name="Math Solver",
    instructions="You solve math problems by evaluating them with python and returning the result",
    tools=[solve_equation],
)

result = await Runner.run(agent, "what is 15 + 28?")

# Run Result object
print(result)

# Get the final output
print(result.final_output)

# Get the entire list of messages recorded to generate the final output
print(result.to_input_list())

Now we have a basic agent, let’s evaluate whether the agent responded correctly!

Evaluating our agent

Agents can go awry for a variety of reasons.

1. Tool call accuracy - did our agent choose the right tool with the right arguments?

2. Tool call results - did the tool respond with the right results?

3. Agent goal accuracy - did our agent accomplish the stated goal and get to the right outcome?

We’ll setup a simple evaluator that will check if the agent’s response is correct, you can read about different types of agent evals here.

Let’s setup our evaluation by defining our task function, our evaluator, and our dataset.

import asyncio


# This is our task function. It takes a question and returns the final output and the messages recorded to generate the final output.
async def solve_math_problem(dataset_row: dict):
    result = await Runner.run(agent, dataset_row.get("question"))
    return {
        "final_output": result.final_output,
        "messages": result.to_input_list(),
    }


dataset_row = {"question": "What is 15 + 28?"}

result = asyncio.run(solve_math_problem(dataset_row))
print(result)

Next, we create our evaluator.

from phoenix.evals import LLM, ClassificationEvaluator, bind_evaluator

# Template for evaluating math problem solutions
MATH_EVAL_TEMPLATE = """
You are evaluating whether a math problem was solved correctly.

[BEGIN DATA]
************
[Question]: {question}
************
[Response]: {response}
[END DATA]

Assess if the answer to the math problem is correct. First work out the correct answer yourself,
then compare with the provided response. Consider that there may be different ways to express the same answer
(e.g., "43" vs "The answer is 43" or "5.0" vs "5").

Your answer must be a single word, either "correct" or "incorrect"
"""

# Run the evaluation
llm = LLM(provider="openai", model="gpt-4.1")
correctness_evaluator = ClassificationEvaluator(
    name="math_correctness",
    prompt_template=MATH_EVAL_TEMPLATE,
    llm=llm,
    choices={"correct": 1.0, "incorrect": 0.0},
)

correctness_eval = bind_evaluator(
    correctness_evaluator,
    input_mapping={
        "question": "input.question",
        "response": "output.final_output",
    },
)

Create synthetic dataset of questions

Using the template below, we’re going to generate a dataframe of 25 questions we can use to test our math problem solving agent.

MATH_GEN_TEMPLATE = """
You are an assistant that generates diverse math problems for testing a math solver agent.
The problems should include:

Basic Operations: Simple addition, subtraction, multiplication, division problems.
Complex Arithmetic: Problems with multiple operations and parentheses following order of operations.
Exponents and Roots: Problems involving powers, square roots, and other nth roots.
Percentages: Problems involving calculating percentages of numbers or finding percentage changes.
Fractions: Problems with addition, subtraction, multiplication, or division of fractions.
Algebra: Simple algebraic expressions that can be evaluated with specific values.
Sequences: Finding sums, products, or averages of number sequences.
Word Problems: Converting word problems into mathematical equations.

Do not include any solutions in your generated problems.

Respond with a list, one math problem per line. Do not include any numbering at the beginning of each line.
Generate 25 diverse math problems. Ensure there are no duplicate problems.
"""

import nest_asyncio

from phoenix.evals import LLM

nest_asyncio.apply()
pd.set_option("display.max_colwidth", 500)

# Initialize the model
llm = LLM(provider="openai", model="gpt-4o")

# Generate math problems
resp = llm.generate_text(prompt=MATH_GEN_TEMPLATE)

# Create DataFrame
split_response = resp.strip().split("\n")
math_problems_df = pd.DataFrame(split_response, columns=["question"])
print(math_problems_df.head())

Experiment in Development

During development, experimentation helps iterate quickly by revealing agent failures during evaluation. You can test against datasets to refine prompts, logic, and tool usage before deploying.

In this section, we run our agent against the dataset defined above and evaluate for correctness using LLM as Judge.

Create an experiment

With our dataset of questions we generated above, we can use our experiment feature to track changes across models, prompts, parameters for our agent.

Let’s create this dataset and upload it into the platform.

import uuid

from phoenix.client import AsyncClient

unique_id = uuid.uuid4()

# Upload the dataset to Phoenix
# Use AsyncClient here because the agent task is asynchronous.
px_client = AsyncClient()
dataset = await px_client.datasets.create_dataset(
    dataframe=math_problems_df,
    input_keys=["question"],
    name=f"math-questions-{unique_id}",
)
print(dataset)

initial_experiment = await px_client.experiments.run_experiment(
    dataset=dataset,
    task=solve_math_problem,
    evaluators=[correctness_eval],
    experiment_description="Solve Math Problems",
    experiment_name=f"solve-math-questions-{str(uuid.uuid4())[:5]}",
)

View Traces in Phoenix

Results

Evaluating in Production

In production, evaluation provides real-time insights into how agents perform on user data.

This section simulates a live production setting, showing how you can collect traces, model outputs, and evaluation results in real time.

Another option is to pull traces from completed production runs and batch process evaluations on them. You can then log the results of those evaluations in Phoenix.

!pip install openinference-instrumentation

from opentelemetry.trace import StatusCode, format_span_id

After importing the necessary libraries, we set up a tracer object to enable span creation for tracing our task function.

tracer = tracer_provider.get_tracer(__name__)

Next, we define our correctness evaluator. In the task function below, we call evaluator.evaluate() directly to get both a label and an explanation, enabling metadata to be captured during tracing.

We also revise the task function to include with clauses that generate structured spans in Phoenix. These spans capture key details such as input values, output values, and the results of the evaluation.

from phoenix.evals import LLM, ClassificationEvaluator

# Template for evaluating math problem solutions
MATH_EVAL_TEMPLATE = """
You are evaluating whether a math problem was solved correctly.

[BEGIN DATA]
************
[Question]: {question}
************
[Response]: {response}
[END DATA]

Assess if the answer to the math problem is correct. First work out the correct answer yourself,
then compare with the provided response. Consider that there may be different ways to express the same answer
(e.g., "43" vs "The answer is 43" or "5.0" vs "5").

Your answer must be a single word, either "correct" or "incorrect"
"""

# Run the evaluation
llm = LLM(provider="openai", model="gpt-4.1")
correctness_evaluator = ClassificationEvaluator(
    name="math_correctness",
    prompt_template=MATH_EVAL_TEMPLATE,
    llm=llm,
    choices={"correct": 1.0, "incorrect": 0.0},
)

# This is our modified task function.
async def solve_math_problem(dataset_row: dict):
    with tracer.start_as_current_span(name="agent", openinference_span_kind="agent") as agent_span:
        question = dataset_row.get("question")
        agent_span.set_input(question)
        agent_span.set_status(StatusCode.OK)

        result = await Runner.run(agent, question)
        agent_span.set_output(result.final_output)

        task_result = {
            "final_output": result.final_output,
            "messages": result.to_input_list(),
        }

        # Evaluation span for correctness — use direct single-record eval
        with tracer.start_as_current_span(
            "correctness-evaluator",
            openinference_span_kind="evaluator",
        ) as eval_span:
            evaluation_result = correctness_evaluator.evaluate(
                eval_input={"question": question, "response": result.final_output}
            )
            eval_span.set_attribute("eval.label", evaluation_result[0].label)
            if evaluation_result[0].explanation:
                eval_span.set_attribute("eval.explanation", evaluation_result[0].explanation)

        # Logging our evaluation
        span_id = format_span_id(eval_span.get_span_context().span_id)
        eval_data = {
            "span_id": span_id,
            "label": evaluation_result[0].label,
            "score": evaluation_result[0].score,
        }
        if evaluation_result[0].explanation:
            eval_data["explanation"] = evaluation_result[0].explanation
        df = pd.DataFrame([eval_data])

        from phoenix.client import AsyncClient

        px_client = AsyncClient()
        await px_client.spans.log_span_annotations_dataframe(
            dataframe=df,
            annotation_name="correctness",
            annotator_kind="LLM",
        )

    return task_result


dataset_row = {"question": "What is 15 + 28?"}

result = asyncio.run(solve_math_problem(dataset_row))
print(result)

Finally, we run an experiment to simulate traces in production.

initial_experiment = await px_client.experiments.run_experiment(
    dataset=dataset,
    task=solve_math_problem,
    experiment_description="Solve Math Problems",
    experiment_name=f"solve-math-questions-{str(uuid.uuid4())[:5]}",
)

View Traces and Evaluator Results in Phoenix as Traces Populate