evaluate

Diagnoses UX quality through heuristic evaluations, cognitive walkthroughs, and anti-pattern detection with prioritized findings.

Updated Jul 21, 2026
One-click install
npx skills add https://github.com/rakedzior/rootine --skill evaluate-rakedzior
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: evaluate
Source: https://github.com/rakedzior/rootine/tree/main/agent/skills/evaluate
Command: npx skills add https://github.com/rakedzior/rootine --skill evaluate-rakedzior

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Design reviews often produce vague feedback like "the navigation feels off" without actionable direction. This Skill runs structured UX evaluations that score interfaces against established heuristics, walk through tasks step by step, detect manipulative dark patterns, and measure task success — then routes every finding to the specialist skill that owns the fix. ## Core Features & Use Cases - Heuristic Evaluation: Applies Nielsen's 10 usability heuristics with a 0-4 severity scale, documenting specific violations with evidence for each. - Cognitive Walkthrough: Examines every step of key task flows against four questions (motivation, visibility, understanding, feedback) to locate exact failure points. - Anti-Pattern Detection: Scans for deceptive patterns, prechecked consent, fake urgency, addictive design, and accessibility weaponization, with severity classification and regulatory risk flags. - Assessment-to-Action Routing: Maps every issue to the Intent skill that owns the remediation (e.g., /organize for navigation, /fortify for edge cases, /include for accessibility), producing a prioritized roadmap instead of a laundry list. - Use Case: Before launching a new signup flow, run an evaluation to get a UX health score, a P0-P3 prioritized issue list, cognitive walkthrough results per step, and explicit routing like "Engage /fortify for missing loading states and /articulate for confirmation copy." ## Quick Start Ask the AI to evaluate the UX of your signup flow and produce a scored, prioritized report with recommended remediation skills.

Frequently Asked Questions about evaluate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a heuristic evaluation of a user interface?

Apply Nielsen's 10 usability heuristics systematically, scoring each on a 0-4 scale from no issues to catastrophic. Document specific violations with evidence, such as missing loading indicators under visibility of system status or jargon in error messages under match with the real world.

What is a cognitive walkthrough in UX evaluation?

A cognitive walkthrough examines each step of a task flow against four questions: will the user know what to do, notice the action, understand what it does, and see feedback afterward. Steps with two or more failures are rated as likely abandonment points.

How do I detect dark patterns in a design?

Scan the experience against categories like confirmshaming, prechecked consent, fake urgency, roach motels, and asymmetric opt-outs. Classify each finding by severity from critical regulatory risk to minor, since dark patterns can violate GDPR, ADA, or FTC guidelines.

Does this evaluation skill fix the UX issues it finds?

No, it only diagnoses and routes. Each finding is mapped to a specialist skill such as /organize for navigation, /fortify for edge cases, /include for accessibility, or /articulate for copy, which own the actual remediation work.

When should I use a structured UX audit versus user research?

Use a structured evaluation when you need an expert assessment of an existing design against established heuristics without recruiting participants. When the evaluation surfaces questions that require talking to users, route to a research skill like /investigate instead.