Create an Agent with OpenAI SDK

Introduction

Intro to AI Agents

CLI Agent Demo

Understanding Evals

Create an Agent Tool Call

Evals Telemetry with Laminar

Adding Laminar Tracing

Evaluators

Agent Loop Overview

Running Evaluations

Single-Turn Eval Executor

Running an Agent Loop

Analyze Eval Results

Coding an Agent Loop

Coding a System Prompt

Multi-Turn Evals

Coding the Eval

Coding a User Message

Read & Write

Use Cases for File System Tools

Testing

List & Delete

Web Search for Agents

Creating a Web Search Tool

Strategies for Managing Context

Model Token Limits

Context Window Compaction

Shell & Code Execution

Adding Context Window Management

Code Execution Tool

Human in the Loop

Sandboxed Execution

Approval Flow Architectures

Adding Approvals

Wrapping Up