critique-usability

Solid

Reviews HTML or markdown UI specs, wireframe write-ups, and page mockups against Nielsen's 10 usability heuristics: system status, user control and exits, consistency, error prevention and recovery, recognition over recall, and minimalist design. Judges the interface's flow, controls, and states, not the wording of error or empty-state message text (critique-microcopy covers that), and not conformance against accessibility standards such as contrast or screen-reader access (critique-accessibility covers that). Use when the user asks for a usability review, design feedback, a second opinion, a red-line pass, a heuristic evaluation, or a quality check on a screen, a flow, or an interface spec before it goes to build. Covers static specs and mockups, not live running applications.

AI & Automation 1 stars 0 forks Updated 4 days ago Apache-2.0

Install

View on GitHub

Quality Score: 80/100

Stars 20%
10
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# critique-usability Critiques an interface design against Nielsen's 10 usability heuristics, operationalized into the 20 `NNG-H*` criteria in `references/NNG-HEURISTICS.md`. It reports where a design as specified would obstruct the user: a state with no way out, an action that commits without confirmation, a control nothing on screen names, an error that reports failure without a next step. ## Artifact claim **This skill critiques HTML or markdown UI specs, wireframe write-ups, and page mockups. It does not critique live running applications.** That narrow claim is deliberate ([S-05 (skills slate)](../../docs/internal/release-plans/plan_v0.1.0/S-05_skills-slate/spec.md) AC-8): every criterion here is decided against what the artifact commits to, never against observed runtime behavior. Timing, responsiveness, actual latency, real input handling, and anything that requires operating the interface are out of scope, because a static artifact cannot evidence them and a critique that pretends otherwise is speculation with a criterion ID attached. What that means in practice: - A flow whose next state is undefined is a finding about the artifact's silence, reported as such. - A criterion that would ask how fast something responds instead asks whether the artifact specifies any status indication at all, and whether what it specifies is unambiguous. - A request to evaluate a deployed URL, a running build, or a recorded session is out of claim. Say so, and critique the spec ...

Details

Author
product-on-purpose
Repository
product-on-purpose/critique-skills
Created
3 weeks ago
Last Updated
4 days ago
Language
Python
License
Apache-2.0

Similar Skills

Semantically similar based on skill content — not just same category

AI & Automation Solid

critique-accessibility

Reviews HTML pages and fragments (markdown where mappable) against WCAG 2.2 AA: contrast, alt text, heading hierarchy for screen readers, link text, and keyboard and screen-reader access. Judges conformance against WCAG, not an interface's general usability, flow, or controls (critique-usability covers that). Use when the user asks for an accessibility review, feedback, a second opinion, a red-line pass, an a11y audit, or a pre-launch quality check on a page or component.

1 Updated 4 days ago
product-on-purpose
AI & Automation Solid

critique-microcopy

Reviews error messages, empty states, and other short microcopy strings, including screens annotated with placement, container, timing, and behavior context, against NN/g's error-message guidelines: plain language, specificity, constructive next steps, neutral tone, and recovery grace. Judges the message text itself, not the surrounding screen's flow, controls, or confirmation behavior (critique-usability covers that). Use when the user asks for a review, feedback, a second opinion, a red-line pass, or a quality check on error copy, empty-state copy, form validation messages, or other short UI text before it ships.

1 Updated 4 days ago
product-on-purpose
Web & Frontend Listed

design-review

Run a structured UX critique on a screenshot, URL, or HTML snippet - a 0-4 score, 3-6 prioritized issues each with Before/After/Why, and one citation per claim. Given two or more variants, compares them on shared dimensions and names a winner. Given a screen plus a decision question, answers the question with the screen as evidence instead of returning a critique. Trigger phrases - "review this design", "critique this screen", "what's wrong with this UI", "score this design", "which of these two is better", "did the redesign improve this", "should I use a modal or a drawer". Do not use for WCAG-only accessibility audits (use accessibility-audit) or turning a mockup into a dev spec (use design-handoff).

2 Updated yesterday
humbleteam