What problem does it solve? Building effective AI agent skills requires more than writing instructions — descriptions must trigger reliably, outputs must be verified, and improvements must be validated against baselines. This Skill provides a complete workflow for drafting skills, running test cases with and without the skill, grading results quantitatively, and iterating based on user feedback. ## Core Features & Use Cases - Skill Authoring: Interview the user, capture intent, and generate a SKILL.md template with scripts, references, and assets directories via the init_skill.py script. - Benchmark Evaluation: Run test prompts with-skill and baseline configurations in parallel, grade assertions, and aggregate pass rates, timing, and token statistics into benchmark.json reports. - Description Optimization: Generate trigger/no-trigger eval queries, review them in an HTML interface, and run an automated optimization loop that improves the skill's description for better triggering accuracy. - Use Case: A user wants to build a skill that converts meeting notes into action items. This Skill helps draft the instructions, creates realistic test prompts, runs comparative evaluations, presents results in a browser-based review viewer, and iterates until the outputs meet expectations. ## Quick Start Ask the assistant to create a new skill for your desired workflow and it will interview you, draft the skill, and set up test cases to validate it.
