EARS (Easy Approach to Requirements Syntax) is a technique for writing natural-language requirements using a small set of fixed sentence patterns. It was developed by Alistair Mavin and colleagues at Rolls-Royce to improve requirements for aircraft engine control systems and was published in 2009.
EARS classifies requirements into five types: ubiquitous, event-driven (WHEN), state-driven (WHILE), unwanted behaviour (IF-THEN), and optional feature (WHERE), each written with its keyword and the word SHALL. The aim is to reduce ambiguity and omissions and produce testable requirements. It is also used in spec-driven development as a format for passing requirements to AI coding agents.