OpenAI Agents SDK

OpenAI Agents SDK is OpenAI’s framework for building AI agents. SideSeat captures traces via Logfire’s OpenTelemetry instrumentation.

Prerequisites

• SideSeat running locally (sideseat)
• Python 3.9+
• OpenAI API credentials configured

Quick Start

1. Install dependencies

   pip install openai-agents "sideseat[openai-agents]"
   # or
   uv add openai-agents "sideseat[openai-agents]"

2. Add telemetry

   from sideseat import SideSeat, Frameworks
   SideSeat(framework=Frameworks.OpenAIAgents)
   
   from agents import Agent, Runner
   
   agent = Agent(
       name="Assistant",
       instructions="You are a helpful assistant."
   )
   
   result = Runner.run_sync(agent, "Hello!")
   print(result.final_output)

Without SideSeat SDK

Use Logfire’s instrumentation with a manual OTLP exporter:

import logfire
from opentelemetry import trace
from opentelemetry.sdk.trace.export import BatchSpanProcessor
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter

logfire.configure(
    service_name="my-agent",
    send_to_logfire=False,
    console=False
)

logfire.instrument_openai_agents()

provider = trace.get_tracer_provider()
if hasattr(provider, "add_span_processor"):
    provider.add_span_processor(
        BatchSpanProcessor(
            OTLPSpanExporter(endpoint="http://localhost:5388/otel/default/v1/traces")
        )
    )

Tools

Define tools using the @function_tool decorator:

from agents import Agent, Runner, function_tool

@function_tool
def get_weather(city: str) -> str:
    """Get weather for a city."""
    return f"Sunny in {city}"

agent = Agent(
    name="WeatherAssistant",
    instructions="Use get_weather to provide forecasts.",
    tools=[get_weather]
)

result = Runner.run_sync(agent, "What's the weather in Paris?")

SideSeat shows:

• Parent span for the agent run
• Child spans for each OpenAI call
• Child spans for each tool execution

Session Tracking

Add session context with manual spans:

from opentelemetry import trace

tracer = trace.get_tracer(__name__)

with tracer.start_as_current_span(
    "openai_agents.session",
    attributes={
        "session.id": "session-123",
        "user.id": "user-456"
    }
):
    result = Runner.run_sync(agent, "Hello!")

Extracted Attributes

Attribute	Description
gen_ai.system	openai
gen_ai.request.model	OpenAI model ID
gen_ai.usage.input_tokens	Input token count
gen_ai.usage.output_tokens	Output token count
logfire.span_type	Span type from Logfire

Async Usage

The SDK supports async execution:

import asyncio
from agents import Agent, Runner

agent = Agent(
    name="Assistant",
    instructions="You are a helpful assistant."
)

async def main():
    result = await Runner.run(agent, "Hello!")
    print(result.final_output)

asyncio.run(main())

Next Steps

• OpenAI — provider details
• Python SDK — SDK reference
• Core Concepts — understanding runs, steps, and messages