What problem does it solve? Competitive programming problem setters need proof that their model solution is correct and that their test data actually catches wrong or slow submissions. This Skill automates that validation: it stress-tests the intended solution against a brute-force oracle and attacks the test suite with a small zoo of plausible WA/TLE solutions. ## Core Features & Use Cases - Stress Testing: Runs the model solution against a brute-force oracle for many rounds, mixing tiny N with the largest N the brute can handle, and flags any mismatch. - Solution Zoo: Authors a small set of deliberately wrong (greedy WA, slower-complexity TLE) and alternative accepted solutions, each annotated with @tag, @expect, @algorithm, and @complexity metadata in the C++ header. - Verdict Matrix: Compiles and runs every solution on generated tests, comparing stdout against expected answers to expose wrong answers that exit codes alone would hide. - Use Case: After writing a model solution and test generators for a Polygon problem, use this Skill to confirm the brute and model agree, verify a plausible greedy fails on official tests, and decide whether the suite needs a killer test. ## Quick Start Ask the AI to stress-test the model solution against the brute force and check whether the current test suite catches a greedy wrong-answer solution.
