Testing vs Design · both proven

skill-creator vs vercel-composition-patterns

Both skills were run through Claude twice: once on a fixed brief without the skill, once with it. The briefs differ because each is written for the job the skill claims to do, so compare each skill against its own baseline, then decide which difference matters for your work. They are in different categories, so this is a choice of job, not a head-to-head on the same job.

skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Category: Testing
Source: anthropics/skills
License: Not stated
Proof model: claude-opus-5-5
Brief: “Write a complete SKILL.md for a skill called release-notes that turns a list of merged pull request titles into customer-facing release notes grouped by feature, fix and breaking change, in the tone of a small developer-tools company. Present the finished SKILL.md as a single HTML page, with the frontmatter and body shown in a code block and a short note under it on how you decided what to include.”
Skill commit: dbd4588

Without the skillWith skill-creator

⇆

Without, runWith skill-creator, run3 independent runs per side, none selected by hand.

Download the with-skill page

vercel-composition-patterns

React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.

Category: Design
Source: vercel-labs/agent-skills
License: Not stated
Proof model: claude-opus-5-5
Brief: “Refactor this React Modal component. It currently takes fourteen props, including isConfirm, isDanger, showFooter, footerAlign, hideClose, title, subtitle, icon, primaryLabel, secondaryLabel, onPrimary, onSecondary, size and children, and a dozen screens pass different combinations of them. Present the refactor as a single HTML page with the new API, the implementation, three call sites rewritten before and after, and a short note on why the new shape scales.”
Skill commit: 063bee9

Without the skillWith vercel-composition-patterns

⇆

Without, runWith vercel-composition-patterns, run3 independent runs per side, none selected by hand.

Download the with-skill page

We don’t pick a winner for you: the pages are the evidence. Both proofs can be repeated with the brief, model and commit shown above. See alternatives to skill-creator or alternatives to vercel-composition-patterns.