What problem does it solve? When expected values cannot be written by hand for every case, this Skill provides a way to test stateful systems by defining an abstract model (such as a state machine) as the test oracle, auto-generating operation sequences, and checking the real implementation against the model step by step. ## Core Features & Use Cases - Abstract Model Definition: Represent system behavior with a deliberately naive model (e.g., an LRU cache as an insertion-ordered Map) that serves as a trustworthy oracle. - Stateful Property-Based Testing: Use fast-check commands with check/run contracts to generate operation sequences and run model and implementation in parallel, verifying return values and invariants after every step. - State and Transition Coverage: Confirm generated sequences reach all model states and transitions, using fc.statistics or TLA+/TLC transition sets as a coverage ledger. - Use Case: Testing an LRU cache where you generate random put/get sequences, advance both a naive Map-based model and the real cache, and catch divergence on any step. ## Quick Start Ask the AI to write a model-based test for your stateful component using fast-check commands, with a naive abstract model as the oracle and per-step state comparison.
