What problem does it solve?
This Skill helps you turn an idea for a Claude Code skill into a reliable, well-triggering automation by guiding you through drafting the skill, creating evaluation prompts, running trigger/evaluation tests, and iterating until performance is improved.
Core Features & Use Cases
- Skill drafting & improvement loop: Turn requirements into a usable SKILL.md, then iteratively refine it based on evaluation outcomes.
- Eval-driven refinement: Build a test set, run Claude with/without the skill, and use results (including variance/benchmarking) to identify where the skill should be improved.
- Trigger accuracy optimization: Improve the SKILL.md frontmatter description so the skill triggers at the right times (reducing under-triggering).
- User collaboration model: Progresses from intent capture and research to evaluation design, benchmark review, and repeat iterations.
Quick Start
Ask to create a new skill by telling what you want Claude to do, when it should trigger, and what output format you expect.
