Skip to content

Usability Tester

Verified

Design usability test plans with task scenarios, success metrics, and observer guides. Analyzes session recordings or notes to identify friction points and severity-ranked usability issues.

By Anthropic 1,500 v1.0.0 Updated 2026-03-05

Install

Claude Code

Copy the SKILL.md file to .claude/skills/usability-tester.md

About This Skill

Usability Tester creates the research infrastructure for finding and prioritizing real usability problems before they reach production users.

Test Plan

Produces a complete test plan: research objectives, participant criteria (5-8 participants for qualitative usability testing), task scenarios written as goal-based prompts (not step-by-step instructions that bias behavior), success metrics (task completion rate, time-on-task, error rate, SUS score), and a test environment setup guide.

Task Scenario Design

Writes task scenarios in goal-based language: "You want to add a new team member and give them editor access" rather than "Click the Settings menu, then click Team Members". Avoids priming participants with UI terminology that biases their mental model.

Observer and Note-Taking Guide

Generates a structured observation form for note-takers: behavioral observations (what did the participant do), verbatim quotes (what did they say), and confusion/frustration/delight moments timestamped for video review.

Issue Analysis

Analyzes provided session notes to identify: recurring failure patterns across participants, severity ratings (critical = task failure, high = severe difficulty, medium = confusion with recovery, low = minor friction), root cause analysis, and recommended design changes with rationale.

Use Cases

  • Writing usability test tasks for a new checkout flow before launch
  • Designing an unmoderated remote usability test with think-aloud protocol
  • Analyzing session observation notes from 8 usability test participants
  • Producing a severity-ranked issue list for the design team to action

Pros & Cons

Pros

  • + Writes goal-based task scenarios that avoid biasing participant behavior
  • + Generates structured observer forms for consistent note-taking across sessions
  • + Analyzes session notes to identify recurring failure patterns and root causes
  • + Severity-ranks issues using a standard critical/high/medium/low framework

Cons

  • - Requires session notes or observation data as input; cannot observe sessions directly
  • - 5-participant sampling may miss edge cases in large diverse user populations

Related AI Tools

Related Skills

Stay Updated on Agent Skills

Get weekly curated skills + safety alerts

每周精选 Skills + 安全预警