Plan and execute statistically valid A/B tests with sample size calculations.
npx skills add https://github.com/m0at/claudemd --skill ab-test-setup-m0at
Please help me install this Agent Skill. Skill: ab-test-setup Source: https://github.com/m0at/claudemd/tree/main/skills/ab-test-setup Command: npx skills add https://github.com/m0at/claudemd --skill ab-test-setup-m0at
Planning and running A/B tests can be complex and error-prone without a structured framework. This Skill helps teams design statistically valid experiments, articulate hypotheses, and ensure proper measurement and guardrails.
Define your hypothesis, choose a primary metric, and calculate the required sample size to start your first A/B test.
High-intent search queries and answers about installing and using this skill.
A/B test sample size calculation requires defining your primary metric, expected effect size, and statistical significance threshold to determine the exact number of participants needed for reliable outcomes.
A hypothesis-driven A/B test plan articulates a clear, testable hypothesis linking a product change to a business outcome, paired with primary metrics, variants, and decision criteria for structured measurement.
Interpreting A/B test results involves analyzing statistical significance alongside practical significance, evaluating variants against guardrail metrics, and extracting learnings to guide next steps and product decisions.
A/B test duration is determined by dividing the calculated sample size by your expected daily traffic, ensuring the experiment runs long enough to reach statistical significance and capture normal user behavior.
Guardrails in experimental design are secondary metrics monitored during an A/B test to ensure that improvements in the primary metric do not negatively impact critical business outcomes or user experience.
A/B testing works for web and product features where you can define a clear hypothesis, measure a primary metric, calculate required sample sizes, and establish guardrails to ensure statistical validity.
WanLanglinGenerate three to five testable hypotheses with predictions and experimental plans.
spandiosAnalyze A/B test results for statistical significance and actionable recommendations.
mk-162Design and document A/B testing plans for website experiments.
nahisahoAutomates adaptive experimental planning and analysis for multi-armed bandit, SPRT, Bayesian dose-finding, and response-adaptive randomized trials.
