
from Vern Bot14
Persona-driven UX reviewer that evaluates features, journeys, and error states from an empathetic, user-first viewpoint to surface usability issues and fixes.
UX Vern is a persona-based UX reviewer skill that evaluates product interactions from the perspective of the end user. It maps user journeys, identifies friction and accessibility problems, and recommends concrete design/prioritisation changes to improve usability and onboarding.
Invoke this skill when you need an empathy-first critique of UI/UX: early design reviews, post-implementation usability checks, onboarding assessments, error-state improvements, or accessibility audits. Also useful for reviewing copy, progressive disclosure, and first-time experiences.
Designed for agents that produce user-facing copy and interaction suggestions (Claude, Codex-like agents, Cursor). Useful where model-based evaluation and human-friendly explanations are needed.
UX Vern is a persona-driven skill that instructs the agent to evaluate features and user journeys from an empathy-first, user-centric viewpoint. It has no scripts or references — it's purely a prompt-engineering persona with frontmatter, a workflow, and catchphrases. Instructions are clear enough for an agent to follow but lack specificity on output format, integration points, or how findings should be structured and delivered.
Clean skill with no security concerns. Architecture is thin — no scripts/, references/, or output contracts. Frontmatter has name and description but missing most optional fields. The skill is essentially a system prompt masquerading as a skill; it would benefit from structured output templates and optional reference materials on UX heuristics. Usefulness is moderate — UX review from an agent is handy but this adds little beyond what a well-crafted prompt could achieve.