i-critique_skill

This skill helps evaluate interface design from a UX perspective, offering actionable critiques on hierarchy, IA, emotion, and usability.

Shell

31

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

3 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill steveclarke/dotfiles --skill i-critique

SKILL.md5.4 KB

Overview

This skill evaluates product interfaces from a UX and design-director perspective, producing a prioritized, actionable critique. It focuses on whether an interface actually works for real people—visual hierarchy, information architecture, emotional resonance, microcopy, states, and the deadly signs of generic AI output.

How this skill works

Follow the i-frontend-design Context Gathering Protocol before any review; if no context exists, run teach-impeccable first to collect goals and user context. The skill runs a ten-point inspection (including an AI Slop Detection check) and generates a structured critique with an Anti-Patterns verdict, overall impression, what’s working, top priority issues with concrete fixes and commands, minor observations, and provocative questions to unblock design decisions.

When to use it

Before a launch to catch high-impact UX risks
During design reviews to benchmark quality and consistency
When refactoring or redesigning a feature to prioritize work
To audit product pages for trust and conversion issues
When suspecting the design feels generic or AI-generated

Best practices

Always run the Context Gathering Protocol (i-frontend-design) first to anchor critiques in goals
Start with the AI Slop Detection check—if it fails, prioritize originality and clarity
Give 3–5 prioritized fixes, each with a why, a concrete fix, and a command to execute
Favor observable outcomes (faster task completion, clearer primary action) over vague aesthetics
Use short, direct language and call out exact elements (e.g., ‘primary CTA’, ‘hero metric panel’)

Example use cases

Reviewing a signup flow that has low conversion and unclear primary action
Auditing a dashboard suspected of overwhelming users with metrics and cards
Checking a marketing landing page for trust, emotional fit, and CTA prominence
Evaluating mobile app screens for discoverability and affordance issues
Spotting AI-generated visual patterns and recommending authentic alternatives

FAQ

Yes—supply prototype links, screenshots, or files and the Context Gathering answers so the critique is grounded in intent and constraints.

What is AI Slop Detection?

A focused check for visual fingerprints of generic AI-generated interfaces (color palettes, gradients, glassmorphism, identical grids). If users would assume ‘AI made this’, that’s the highest-priority problem.