Back to directory

Interface Skills

A collection of agent skills covering UI, typography, color, accessibility, layout and product writing, to help a coding agent build and review a good interface.

Interface Skills

This is a collection of agent skills covering topics the author writes about on their personal website and design engineering magazine, Interfaces, aimed at helping build a great interface. Skills include better-ui for concentric border radius, optical alignment, contextual icons, hit areas and animation; better-typography for type scale, spacing, variable fonts, OpenType features, wrapping and truncation; better-colors for generating palettes, using semantic tokens, converting formats and checking contrast; better-accessibility for standards compliance; better-layout for grouping, alignment and reading order; and better-writing for consistent product copy, with better-interface combining all of the "better-*" skills into a single review.

User-invoked skills include interface-review, which analyzes a project across UI, typography, layout, color, writing and accessibility and reports detailed findings; explain-interface, which figures out how an existing animation or piece of UI was built on the web; break, which renders a chosen component in every state and scenario on a temporary page to stress test it; and variant, which builds multiple variants of a component being worked on.

The skills install as a package with a skills CLI, or as a Claude Code plugin through a marketplace command.

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.