- Home
- Skills
- Pbakaus
- Impeccable
- Critique
critique_skill
- JavaScript
10.4k
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 pbakaus/impeccable --skill critique- SKILL.md5.2 KB
Overview
This skill evaluates design effectiveness from a UX perspective and produces a director-level critique with actionable fixes. It inspects visual hierarchy, information architecture, emotional resonance, microcopy, states, and detects common AI-generated anti-patterns. Use it to turn subjective impressions into prioritized design work you can act on.
How this skill works
Follow the Context Gathering Protocol from the frontend-design skill first; if no context exists, run teach-impeccable before proceeding. The skill then runs a 10-dimension critique (AI Slop Detection, Visual Hierarchy, IA, Emotional Resonance, Discoverability, Composition, Typography, Color, States, Microcopy) and outputs an Anti-Patterns Verdict, Overall Impression, What’s Working, Priority Issues with fixes and commands, Minor Observations, and provocative Questions to Consider.
When to use it
- Before a design review to surface high-impact UX problems
- When preparing a redesign or sprint backlog, to prioritize fixes
- After an initial prototype to catch AI-generated template sameness
- To audit product pages, dashboards, or onboarding flows
- When handing off to engineers to clarify visual and behavioral intent
Best practices
- Always run the Context Gathering Protocol first; gather goals and target user
- Start critiques by checking AI Slop Detection—eliminate fingerprint patterns early
- Prioritize 3 highest-impact fixes, not a laundry list
- Give concrete, implementable fixes: size, color, copy, state behavior
- Reference specific components (e.g., primary CTA, card grid, nav) when recommending changes
Example use cases
- Audit a SaaS dashboard that feels generic and has poor task focus
- Evaluate onboarding screens to improve completion and clarity
- Review marketing pages to ensure hierarchy and conversion paths
- Critique a mobile app layout for discoverability and affordance
- Assess a component library for consistent typography and color purpose
FAQ
Yes. Run the Context Gathering Protocol from frontend-design; if no context exists, run teach-impeccable first so the critique aligns with product goals.
What if the design is clearly AI-generated?
The Anti-Patterns Verdict will call it out and list specific tells. Start by removing those fingerprint patterns (colors, glassmorphism, identical card grids) and re-establish a distinct visual system tied to user needs.