Skill Eval

Overview

This skill owns the repository's skill-eval test operation through skills/skill-eval/scripts/eval_runner.py. The runner keeps executor and grader separate, records artifacts and metrics, and aggregates with_skill versus without_skill results. Never hand-run prompts, hand-record results, let one agent execute and grade the same answer, or estimate metrics.

Critical CLI Contract

For any command-drafting response, reproduce these exact shapes before adding explanation: