OpenAI API: Agents

AI agents generate plans and perform complex multi-step tasks based on the user input. The OpenAI API provides comprehensive support for building AI agents that use models, tools, knowledge and memory, audio and speech, guardrails, orchestration, and MCP that can be customized to fit specific needs and scenarios. This course explores how to develop advanced agents with the OpenAI API and the Agents Python SDK. Through videos, articles, and hands-on practice in GitHub Codespaces, you’ll define custom agents, set up agent runners, configure handoffs so agents can work together, add guardrails for content moderation, and extend the agents with off-the-shelf OpenAI features, custom functions, and MCP (Model Context Protocol) servers.

Learning objectives
Define what the OpenAI API Agents Python SDK is and how it works.
Describe the agent lifecycle including agents, runners, handoffs, and moderation.
Extend agents with custom tools through function calling.
Build OpenAI agents that use MCP servers.