steve-krug-ux-audit

What This Skill Does

The steve-krug-ux-audit skill is a specialized diagnostic tool designed to evaluate web interfaces through the lens of Steve Krug's seminal usability work, Don't Make Me Think. By leveraging OpenClaw's browser automation, the skill conducts a systematic audit of your digital products. It specifically tests for cognitive friction, navigation clarity, and the 'Trunk Test'—a methodology that forces an objective assessment of whether a user can instantly orient themselves upon landing on any given page. Instead of vague design feedback, the agent provides actionable, granular critique identifying where users are forced to stop and think, highlighting critical usability blockers that lead to bounce rates.

Installation

To integrate this skill into your environment, run the following command in your terminal: clawhub install openclaw/skills/skills/couriard/steve-krug-ux-audit Ensure your agent has the necessary browser-automation permissions enabled to allow the tool to capture page snapshots and render the DOM for analysis.

Use Cases

• Pre-Launch Reviews: Validate that your landing pages or registration flows are intuitive before going live.
• Competitor Benchmarking: Analyze how industry leaders handle navigation and mental models compared to your own implementation.
• Usability Debt Reduction: Identify and prioritize 'question marks' on existing pages that slow down user task completion.
• Prototype Testing: Audit wireframes or high-fidelity prototypes early in the design cycle to prevent costly rework.

Example Prompts

• "@steve-krug-ux-audit please audit the landing page at https://example-saas.com and perform the Trunk Test on the main navigation."
• "Run a Krug audit on this dev environment path: /local/auth-flow/signup-process. Focus specifically on unnecessary mental chatter in the form labels."
• "Audit our competitor's checkout page at [URL] and list the top 3 usability issues that make users think too hard."

Tips & Limitations

• Capture Multiple Pages: The audit is most effective when you feed the skill a complete user task flow rather than just a single page.
• Human-in-the-Loop: While the skill identifies potential friction points based on heuristics, it is best used as a supplement to live user testing, not a complete replacement.
• Visual Context: Always ensure the agent has access to screenshots/visual snapshots, as Krug's principles are heavily dependent on visual hierarchy and layout conventions.