improve-ui

Audit product interfaces against design evidence and write implementation plans.

3|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/langgenius/due-date-hq-jwl --skill improve-ui-langgenius
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: improve-ui
Source: https://github.com/langgenius/due-date-hq-jwl/tree/main/.claude/skills/improve-ui
Command: npx skills add https://github.com/langgenius/due-date-hq-jwl --skill improve-ui-langgenius

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? UI reviews often produce vague, unsupported opinions or changes that break a product's design identity. This Skill audits a product surface strictly against its own governing design system, producing only evidence-backed findings and self-contained implementation plans that another agent can execute. ## Core Features & Use Cases - Evidence-gated UI audit: Traces the rendered path of a selected surface through components, tokens, and variants, and accepts a finding only when contract, runtime, and correction proofs all exist. - Read-only safety: Never modifies product source; all output goes to design-plans/ as self-contained plans based on a reusable plan template. - Design-system reconstruction: Rebuilds the local design language from DESIGN.md and governing sources before judging anything, recording documented decisions and explicit exceptions. - Use Case: Ask it to review your app's dashboard surface; it returns a ranked findings table with cited evidence, recommends the highest-leverage fix, and writes an implementation-ready plan only for changes you select. ## Quick Start Use the improve-ui skill to audit the main dashboard surface, identify supported UI problems, and recommend what to improve first without modifying any source files.

Frequently Asked Questions about improve-ui

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit a UI for design-system inconsistencies?

Select one coherent product surface, trace its rendered path through components, tokens, and variants, then compare each property against documented design decisions. This Skill automates that workflow and reports only findings backed by contract, runtime, and correction evidence.

What makes a UI finding valid in a design audit?

A valid finding needs three proofs: a binding design contract governing the property, a traced runtime path showing the owner reaches the surface, and one deterministic correction. Repetition, naming similarity, or personal preference alone do not qualify.

Does this UI audit modify my product source code?

No. The audit is strictly read-only on product source. It only creates or edits plan files under design-plans/, and a separate agent executes any approved changes.

Can the audit flag accessibility or broken-route issues?

Not by default. Accessibility, ARIA semantics, broken routes, data wiring, and performance findings are discarded unless the user explicitly requests them or a documented design contract governs them.

When should I not use an evidence-gated UI audit?

Avoid it when you need functional bug fixes, behavior changes, or a full product redesign. The audit only covers visual presentation, copy, layout, and documented design-rule conformance on a single coherent surface.