From blueprint-dev
Production A/B testing lifecycle for design variants. Covers hypothesis formation, feature flags, variant comparison, analytics tracking, statistical significance analysis, experiment setup, and cleanup.
How this skill is triggered — by the user, by Claude, or both
Slash command
/blueprint-dev:ab-testingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This skill provides the complete lifecycle for production A/B testing of design variants. Variants are real, production-quality code — not mockups.
This skill provides the complete lifecycle for production A/B testing of design variants. Variants are real, production-quality code — not mockups.
CREATE (/design) → DEPLOY (trunk + flags) → MEASURE (analytics) → DECIDE (/ab-decide) → CLEANUP (/ab-cleanup)
/blueprint-dev:bp:design uses the design-variant-generator to create 2-3 real component variants, the design-critic to evaluate them, and the ab-test-engineer to wire up flags and tracking.
Variants ship to trunk behind feature flags. Compatible with trunk-based development — no long-lived branches needed.
Analytics tracking fires at key interaction points. Users monitor their analytics dashboard for results.
/blueprint-dev:bp:ab-decide uses the design-decision-analyst to interpret results and recommend a winner based on statistical significance.
/blueprint-dev:bp:ab-cleanup follows the decision document's cleanup plan to remove the losing variant, promote the winner, and clean up flags/tracking.
references/tracking-plan-template.md — Template for tracking plansreferences/code-templates.md — Stack-specific code templates for wrappers, flags, and trackingnpx claudepluginhub dlabs/claude-marketplace --plugin blueprint-devGuides planning, designing, and implementing A/B tests, split tests, multivariate experiments. Covers hypotheses, sample sizes, test types, statistical principles.
Guides A/B test planning with hypothesis frameworks, statistical principles, single-variable testing, and metrics for CRO experiments.
Plans, designs, and implements A/B tests with statistical rigor, hypothesis frameworks, and sample size calculations.