What problem does it solve? Teams routinely claim a product idea is "validated" based on weak evidence: testing the comfortable assumption instead of the riskiest one, trusting "would you use this?" survey answers, setting success criteria after results arrive, or generalizing from a handful of users. This Skill reviews any validation plan, experiment, discovery approach, or "we validated it" claim and produces a severity-rated issue list with concrete fixes. ## Core Features & Use Cases - Seven-principle rigor framework: Checks that the riskiest assumption was tested, hypotheses are falsifiable, method fidelity matches decision risk, behavior is trusted over stated intent, success criteria were pre-set, evidence is read without goalpost-moving, and disconfirming results actually change the plan. - Severity-rated output: Classifies each finding as Critical, Major, Minor, or Cosmetic based on decision damage, and always ends with a prioritized top-3 fix list. - Scope discipline and deferral: Reviews only validation rigor when invoked directly, deferring metric validity to pm-okr-metric-validity-audit and spec risk surfacing to pm-spec-quality-audit. - Use Case: A PM pastes a discovery plan that surveyed 10 users with "would you pay $20?" and declared the idea validated. The audit flags the untested leap-of-faith assumption, the hypothetical-intent evidence, and the missing pre-set success threshold, then recommends a painted-door test with a defined conversion bar. ## Quick Start Ask the assistant to run an assumption rigor audit on your validation plan or experiment results and list the severity-rated issues with fixes.
