What problem does it solve? Testing code where expected outputs are hard to write by hand—such as serializers, parsers, or multi-parameter configurations—often leaves gaps because example-based tests only cover a few fixed inputs. This Skill provides a method catalog for binding correctness through properties that hold for all inputs (property-based testing) and through small covering arrays that exercise parameter interactions (combinatorial testing). ## Core Features & Use Cases - Property-Based Testing (PBT): Enumerate invariants, round-trip relations, metamorphic relations, and known-oracle matches, then implement them with fast-check generators, fixed seeds, and shrink-based counterexample capture. - Combinatorial Testing: Define parameter axes and value domains, encode invalid-combination constraints, and generate 2-way or 3-way covering arrays with tools like pairwise libraries, ACTS, or PICT. - Use Case: When testing a JSON codec, declare the round-trip property decode(encode(x)) === x with fc.jsonValue(), fix numRuns and seed for reproducibility, and pin any shrunk minimal counterexample as a permanent regression test. ## Quick Start Ask the AI to apply the generative-property catalog to design property-based tests with fast-check for a codec or a pairwise covering array for a build matrix.