A/B Test Setup guides an agent through designing statistically valid, actionable experiments and building a continuous growth experimentation program. It establishes core principles such as formulating data-backed hypotheses, isolating single variables, maintaining statistical rigor, and measuring primary, secondary, and guardrail metrics. The skill covers assessment of test context, sample size reference calculations, variant design best practices, traffic allocation strategies, client-side versus server-side implementation considerations, and pre-launch checklists. It also provides a structured analysis workflow for evaluating statistical significance and interpreting results. Furthermore, it supports scaling experimentation as an ongoing growth engine through hypothesis generation from multiple data sources, ICE prioritization scoring, experiment velocity tracking, and building a reusable experiment playbook. Reach for this skill when you need to plan, design, or implement an A/B test, split test, or multivariate experiment, or when establishing a systematic product growth experimentation practice.