AI Browser Automation That Doesn't Suck

Automate browser testing with Playwright CLI and AI agents like Claude Code, from exploration to bug reports
What you'll learn

---

Install and set up Playwright CLI, the AI-agent browser automation tool, and understand how it differs from classic Playwright and Playwright MCP
Direct an AI coding agent like Claude Code or Codex to navigate, inspect, and interact with real web pages using plain-language prompts
Save and reuse login sessions so an AI agent can test authenticated flows without logging in from scratch every time
Record video, capture , and turn an AI agent's findings into clear bug reports someone will actually act on
Turn scattered exploratory testing sessions into one reusable AI testing suite that can be rerun and expanded over time
Apply this entire workflow to any website and any AI coding agent, not just the ones shown in the course
Requirements

---

No prior experience with AI agents or Playwright CLI needed, just the testing judgment you already bring to manual or exploratory work
Node.js and npm installed, for installing the CLI itself
Claude Code, Codex, or another AI coding agent already set up
A website to practice on (the course uses a demo site throughout, but any site works)
Description

---

Playwright CLI is a new tool from Microsoft built for AI coding agents like Claude Code and Codex to drive a real browser. This course is not about replacing testers with AI. It's about giving you a faster way to get browser-driven testing done through an agent you're already directing, while you stay the one deciding what actually matters.
You'll learn by watching real, unscripted sessions: an AI agent installed and configured from scratch, navigating and inspecting a real web app, logging in and saving a session so it never has to log in again, recording video and turning what it found into an actual bug report, and finally building a reusable exploratory testing suite that can be picked up and rerun without starting over.
What you'll cover
What Playwright CLI actually is, and how it's different from the classic Playwright test runner and from Playwright MCP
Navigating, inspecting, and driving a real page through natural-language prompts, no CSS selectors required
Saving and reusing login sessions so your agent skips the login form every time
Capturing video, , and turning findings into a clear, actionable bug report
Wiring this into a repeatable skill your agent already knows, across Claude Code, Codex, or whichever agent you use
A full end-to-end capstone session that ties everything together on one real task
No prior experience with AI agents is required, just the testing judgment you already bring to manual or exploratory work. From the instructor behind the "That Doesn't Suck" testing series.
Who this course is for

---

QA testers who want a faster, AI-agent-driven way to do browser-based testing without giving up their own judgment
Manual or exploratory testers curious about AI-assisted workflows but new to coding agents
Testers already using Playwright who want to add an AI-agent-driven layer on top
Anyone who wants a practical, hands-on look at Playwright CLI specifically, not another generic AI overview
Homepage

---