What problem does it solve? Example-based tests only cover the inputs you thought to write, so parsers, serializers, and math code ship with bugs hiding in null bytes, huge integers, and empty keys. This Skill guides you through writing generative property tests that explore thousands of inputs against a stated invariant. ## Core Features & Use Cases - Property catalog: Choose from five proven patterns — round-trip, oracle, invariant, idempotence/algebra, and metamorphic — instead of improvising weak assertions. - Generator design and reach checks: Build strategies that actually produce boundary, empty, and huge values, verified with library statistics rather than assumed. - Shrinking and pinning: Turn minimized counterexamples into permanent regression tests, with seed and failure-database handling so CI failures reproduce locally. - Use Case: A JSON config parser passes 30 example tests but crashes in production on a null byte; one round-trip property with a nested-structure generator finds four bugs in the first 200 generated cases. ## Quick Start Write a property-based test for my serializer using hypothesis that checks parse(render(x)) == x and pins any counterexample it finds.
