This is a portable lifecycle benchmark harness for comparing AI coding agents on real QA engineering tasks. It rigorously evaluates agent competencies across test authoring, test repair, root-cause debugging, token efficiency, framework adaptation, and complex browser or mobile automation workflows.