What problem does it solve? Building a reliable skill requires more than writing a prompt — it needs structured drafting, realistic test cases, quantitative benchmarking against baselines, and iterative refinement based on user feedback, which is hard to do systematically by hand. ## Core Features & Use Cases - Guided Skill Authoring: Captures intent through interview questions, searches community skills for proven structural patterns, and plans directory layouts with scripts, references, and assets before writing. - Evaluation & Benchmarking: Runs with-skill and baseline test cases in parallel, grades assertions, aggregates pass rates, timing, and token usage with variance analysis, and renders results in an interactive review viewer. - Description Optimization: Generates trigger/no-trigger eval queries and iteratively rewrites the skill description to improve triggering accuracy, with train/test split reporting. - Use Case: A user wants a skill that generates changelogs from git history. This skill interviews them, drafts the SKILL.md, creates test prompts, runs benchmarked evaluations, and iterates until the outputs pass review. ## Quick Start Help me create a new skill that converts meeting transcripts into structured action-item lists, then test and refine it.
