ux-compare

Compares UX patterns across multiple reference app pattern libraries into an opinionated markdown report.

1|Updated Aug 18, 2026
One-click install
npx skills add https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent --skill ux-compare-scsm-unrestrict
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ux-compare
Source: https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent/tree/main/frontend-engineer/skills/ux-compare
Command: npx skills add https://github.com/scsm-unrestrict/dsh-frontend-engineer-agent --skill ux-compare-scsm-unrestrict

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? When designing a new feature, teams lack a structured way to learn from how leading apps solve the same UX problems. This Skill turns multiple UX pattern libraries into a single decision document that shows where apps converge, diverge, and what they all miss. ## Core Features & Use Cases - Cross-app pattern comparison: Reads 2+ pattern-library.md files produced by ux-extract and walks them category by category (empty states, keyboard shortcuts, modals, etc.). - Convergence/divergence analysis: Identifies safe defaults (convergent patterns), genuine design choices (divergent patterns), unique approaches, and notable absences across the set. - Opinionated recommendations: Produces a dated comparison document with actionable guidance for new or existing builds, including staleness warnings for old extracts. - Use Case: Before building a billing page, compare how claude.ai, Linear, Notion, and Superhuman handle destructive actions and empty states, then receive a recommendation on which convention to follow and which gaps to fill. ## Quick Start Compare the UX pattern libraries in docs/ux-extracts and write a comparison of how these apps handle empty states.

Frequently Asked Questions about ux-compare

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

FAQPage Schema
How do I compare UX patterns across multiple apps?

Run ux-extract on each reference app first to generate pattern-library.md files, then run the comparison to walk each category across libraries. The output identifies convergent defaults, divergent choices, unique approaches, and gaps, ending with opinionated recommendations.

What inputs does a UX pattern comparison need?

It needs at least two pattern-library.md files located in docs/ux-extracts/<app-name>/ or .jez/artifacts/ux-extracts/<app-name>/. With only one library the comparison stops and suggests extracting more apps first, since a single library is just a reference, not a comparison.

Does UX comparison require a browser or live app access?

No browser is needed. The comparison is pure markdown analysis of previously extracted pattern libraries, so it works entirely offline against the existing extract files and their documented screenshots and copy.

How many apps can I compare at once?

Three apps is comfortable and five is the practical upper bound for readable prose synthesis. Beyond five, bucket libraries into tiers or run several narrower comparisons focused on specific categories like empty states or keyboard shortcuts.

What happens when a pattern library is outdated?

Stale libraries are still usable but flagged with their extraction date in the comparison output and source health table. Recommendations based on old extracts carry a qualifier, and re-running ux-extract is suggested for libraries several months old.