Help the user write task verifiers with Reward Kit. Reward Kit is a lightweight Python package that turns a directory of criteria files into a reward score. Each criterion is a Python function call or a TOML judge file; folders become separate rewards. Setup in a Harbor task Put criteria alongside in the task's directory: : This runs all criteria in against the workspace at and writes . Defaults match Harbor's conventions — no extra config needed. If judge criteria need API keys, pass them through : Ask whether Reward Kit should run in the agent's shared environment or in a separate verifier…