Back to directory

Agent Skills

Vercel's official collection of skills for AI coding agents, covering Vercel project optimization, React and Next.js performance, web design compliance and Vercel's own writing style guide.

Agent Skills

The collection ships four skills following the Agent Skills format. vercel-optimize audits a deployed Vercel project for cost, performance, reliability, caching and function usage, first collecting Vercel's own metrics and then investigating only the routes and files those metrics point to. react-best-practices packages more than 40 React and Next.js performance rules across categories such as eliminating request waterfalls, bundle size, server-side performance and re-render optimization, each prioritized by impact.

web-design-guidelines reviews UI code against more than 100 rules covering accessibility, focus states, forms, animation, typography, images, performance, navigation, dark mode and internationalization. writing-guidelines reviews documentation and prose against more than 80 rules from Vercel's writing handbook, covering voice and tone, structure, content types, code samples, typography and disclosure of AI-assisted writing.

It is written in JavaScript.

More in Agent skills

See all 581
  • Provenance gate

    github.com

    A deterministic authorization gateway that sits between an AI agent and its tools, tracing where each parameter of a side-effect call came from, and blocking the call or asking a human when the source does not check out.

  • A markdown framework of system rules that restates Isaac Asimov's Four Laws of Robotics as a set of prioritized loss calculations for large-language-model agents to follow, published as a discovery file for agents and crawlers to read.

  • Vitruvius

    github.com

    An engineering research agent that runs a discover, read, synthesize, verify and review loop over technical questions, reading sources directly rather than inferring from titles, and publishing its own scored verifier benchmark, weaknesses included.