1.1The Problem with Traditional Browser Automation

For over two decades, browser test automation has followed the same pattern: a human writes a deterministic script that issues precise commands to a browser. Click this button. Type into that field. When it breaks -- because a CSS class changed or a button moved -- it breaks completely, and a human must fix it.

The industry name for this is brittle automation. Studies consistently show that QA teams spend 40-60% of their time maintaining existing tests rather than writing new ones. The root cause is architectural: traditional tools execute steps with no understanding of intent. They do not know that click("#submit-btn") means "submit the form." When the ID changes to submit-button, the test fails even though the form still works.

> KEY CONCEPT: The fundamental limitation of traditional browser automation is the gap between intent (what the test verifies) and implementation (the exact selectors and steps used to verify it). Agent skills bridge this gap with a reasoning layer between test definitions and browser interactions.