OpenAI Agents SDK

How to integrate Maxim's observability and real-time evaluation capabilities with OpenAI Agents SDK.

The OpenAI Agents SDK enables you to build agentic AI apps in a lightweight, easy-to-use package with very few abstractions. It's a production-ready upgrade of our previous experimentation for agents, Swarm. The Agents SDK has a very small set of primitives:

Agents, which are LLMs equipped with instructions and tools
Handoffs, which allow agents to delegate to other agents for specific tasks
Guardrails, which enable the inputs to agents to be validated

Integrating with Maxim

Create a Maxim account and a Log repository. Follow the instructions in the quickstart section.

Install Maxim SDK

pip install maxim-py

Set the following environment variables to configure the Maxim SDK

Environment Variable	Description
`MAXIM_API_KEY`	Your Maxim API key
`MAXIM_LOG_REPO_ID`	ID of the log repository to use

And then Maxim SDK will automatically initialize using these env variables.

from maxim import Maxim,Config
from maxim.logger.openai.agents import MaximOpenAIAgentsTracingProcessor
# Creating a new logger instance
# It automatically initializes using MAXIM_API_KEY and MAXIM_LOG_REPO_ID from env variables.
logger = Maxim(Config()).logger()

OR you can manually initialize the SDK

from maxim import Maxim,Config
from maxim.logger import LoggerConfig
 
logger = Maxim(Config(api_key="your_api_key")).logger(LoggerConfig(id="your_log_repo_id"))

Add the MaximOpenAIAgentsTracingProcessor to your agent using add_trace_processor or set_trace_processor.

from agents import add_trace_processor
from maxim.logger.openai.agents import MaximOpenAIAgentsTracingProcessor
 
add_trace_processor(MaximOpenAIAgentsTracingProcessor(logger))
 
# Your agent code

Cookbooks

Here are a few cookbooks that you can use to get started with Maxim:

OpenAI Agents SDK

Integrating with Maxim

Cookbooks

On this page