ahmedhamadto/software-forge/ux-usability-review
- community
Checking that the SKILL.md file is still in the repository.
Use when reviewing any interface for usability — walks through Krug's principles from Don't Make Me Think covering cognitive load, scanning, navigation, homepage clarity, mobile usability, accessibility, and the goodwill reservoir.
Install
npx vskill@latest install ahmedhamadto/software-forge/ux-usability-reviewSKILL.md
UX Usability Review
Overview
A structured usability audit methodology based on Don't Make Me Think by Steve Krug. Guides you through 8 sequential phases to systematically evaluate an interface for cognitive load, scanability, navigation clarity, content quality, mobile usability, accessibility, and the overall user experience. The core principle: if a user has to stop and think about how to use your interface, something needs to be redesigned.
When to Use
- Before launching a product, feature, or redesign
- When users report confusion, abandonment, or support requests about basic tasks
- During design reviews or code reviews that affect user flows
- When evaluating a competitor's interface for strengths and weaknesses
- When onboarding a new team member to UX standards
- As a complement to the ui-polish-review skill (that skill covers visual quality; this one covers usability)
When NOT to Use
- For purely visual/aesthetic evaluation (use the ui-polish-review skill for that)
- For performance or technical audits
- As a substitute for actual usability testing with real users (this is a heuristic review, not a user test)
- For content strategy or copywriting projects (though content clarity is covered)
Process
Work through each phase sequentially. At each phase:
- Ask targeted questions about the specific interface being reviewed
- Read the relevant chapter summary from
front-end-design/dont-make-me-think/for detailed guidance - Suggest and execute concrete tests (grep patterns in source code, manual inspection tasks, quick user tests)
- Flag findings with severity:
CRITICAL/HIGH/MEDIUM/LOW/INFO - Summarize findings before moving to the next phase
If the application does not have a relevant surface for a phase (e.g., no homepage for an internal tool), acknowledge and skip with rationale.
Phase 1: The "Don't Make Me Think" Test
Reference: front-end-design/dont-make-me-think/ch01-dont-make-me-think.md, ch02
Goal: Evaluate whether each screen is self-evident -- a user should immediately understand what it is and what to do without expending mental effort.
Questions to ask:
- For each screen: would a first-time user immediately understand what this is and what to do next?
- Are there clever, creative, or branded labels where conventional ones would be clearer? ("Job-o-Rama" vs "Jobs", "Quick Procurement Module" vs "Shopping Cart")
- Is it obvious what is clickable and what is not? (Links look like links, buttons look like buttons, non-interactive elements do not look clickable.)
- Are users expected to read and comprehend, or can they scan and act?
- The 5-second test: if someone saw this page for only 5 seconds, could they tell you what the site does and what the primary action is?
Tests to run:
The 5-Second Test (manual):
- Open the page. Look at it for exactly 5 seconds. Look away.
- Can you answer: What is this site/app? What is the main thing I can do? Where would I start?
- If any of these are unclear, the page fails this test.
Visual inspection checklist:
- [ ] Is every interactive element visually distinguishable from non-interactive elements?
- [ ] Are labels conventional and immediately understandable (no jargon, no cleverness)?
- [ ] Can you identify the primary action on each screen within 2 seconds?
- [ ] Is the page self-evident (no instructions needed) or at minimum self-explanatory (brief moment to understand)?
- [ ] Would removing any "How to" text or instructions break the experience? (If yes, the UI needs redesign, not instructions.)
Finding template:
Phase 2: Visual Hierarchy & Scanning
Reference: front-end-design/dont-make-me-think/ch03-billboard-design-101.md
Goal: Verify the page supports scanning behavior -- users do not read web pages, they scan them. The design must make the most important content immediately findable.
Questions to ask:
- Is there a clear visual hierarchy where the most important thing is the most prominent?
- Are pages broken into clearly defined, visually distinct areas?
- Is text formatted for scanning? (Meaningful headings, short paragraphs, bold key terms, bullet lists.)
- Is there excessive visual noise? (Too many competing elements, background patterns, unnecessary decoration, animation overload.)
- Are related items visually grouped (proximity) and unrelated items visually separated?
Tests to run:
Visual inspection checklist:
- [ ] Can you identify the page's purpose and primary content within 3 seconds of looking at it?
- [ ] Are there clearly defined visual areas (header, main content, sidebar, footer) with obvious boundaries?
- [ ] Is body text broken into short paragraphs (3-4 sentences max) with descriptive headings?
- [ ] Are key terms bolded or otherwise highlighted for scanners?
- [ ] Is the page free of excessive visual noise (competing elements, unnecessary decoration, distracting animations)?
- [ ] Is whitespace used effectively to create breathing room and visual grouping?
Finding template:
Phase 3: Navigation
Reference: front-end-design/dont-make-me-think/ch06-street-signs-and-breadcrumbs.md
Goal: Verify that navigation is clear, persistent, and enables users to always answer three fundamental questions: Where am I? Where can I go? How do I get back?
Questions to ask:
- Is there persistent global navigation visible on every page?
- Can the user always answer: Where am I? Where can I go? How do I get back?
- Is there a clear "you are here" indicator in the navigation (active state, highlight, bold)?
- Does every page have a name/title that matches what was clicked to reach it?
- Are breadcrumbs present for deep hierarchies (3+ levels)?
- Is there a persistent, accessible search function?
- Is the logo linked to the homepage?
Tests to run:
The Trunk Test (standalone quick-check): Land on any interior page of the site blindly (not the homepage). Can you immediately answer:
- What site is this? (Site identity/logo visible)
- What page am I on? (Page name/title visible)
- What are the major sections? (Global navigation visible)
- What are my options at this level? (Local navigation visible)
- Where am I in the scheme of things? (Breadcrumbs or "you are here" indicator)
- How can I search? (Search box visible or easily findable)
If any of these questions cannot be answered within 5 seconds, flag it.
Visual inspection checklist:
- [ ] Is global navigation persistent and visible on every page (not hidden behind a hamburger on desktop)?
- [ ] Is there a "you are here" indicator showing the current page/section?
- [ ] Do page titles match the navigation labels that led to them?
- [ ] Are breadcrumbs present for content deeper than 2 levels?
- [ ] Is the logo clickable and linked to the homepage?
- [ ] Can the user always get back to where they came from (back button works, breadcrumbs, or explicit back link)?
Finding template:
Phase 4: Words & Content
Reference: front-end-design/dont-make-me-think/ch04, ch05-omit-needless-words.md
This is the start of the file. Read the full SKILL.md on GitHub.
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/ux-usability-review/SKILL.md
Popularity
- Repository stars
- 4
- Repository forks
- 1
- 7-day trend score
- 8
Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 27 Mar 2026, 7 months ago.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/ahmedhamadto/software-forge/ux-usability-review)