ui-test

Analyze UI changes and run browser tests with deterministic checks.

Updated Aug 27, 2026

Search Tags:#accessibility #qa #ui-testing #browser #adversarial #parallel-testing #diff-driven

Ready to run

npx skills add https://github.com/andreasasprou/polaris --skill ui-test-andreasasprou

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: ui-test
Source: https://github.com/andreasasprou/polaris/tree/main/.agents/skills/ui-test
Command: npx skills add https://github.com/andreasasprou/polaris --skill ui-test-andreasasprou

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

UI testing can be tedious and unreliable when relying on manual exploration; this Skill automates adversarial UI testing in real browsers to catch issues that traditional scripts miss.

Core Features & Use Cases

Diff-driven testing analyzes git diffs to test only what changed.
Exploratory testing navigates the app to discover edge cases developers might miss.
Parallel testing runs independent Browserbase sessions to speed up QA across localhost or deployed sites.
Deterministic checks include axe-core audits, console health capture, and responsive viewport validation.

Quick Start

Run a local UI QA session with the browse CLI and observe before/after interactions to verify changes.

Frequently Asked Questions about ui-test

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

FAQPage Schema

How do I catch UI defects that automated testing scripts miss?▼

AI-powered adversarial UI testing runs in a real browser to reveal defects automated scripts miss. It applies deterministic checks like axe-core audits, console health capture, and responsive viewport validation across localhost and deployed sites.

Can I run UI testing on only the git diff changes in my pull request?▼

Yes, diff-driven UI testing analyzes git diffs to test only what changed in your PR. It runs in a real browser to validate modified code paths using axe-core audits and responsive viewport checks.

What's the best way to automate exploratory testing for edge cases?▼

Exploratory UI testing navigates your app to discover edge cases developers might miss. It runs in real browsers to reveal defects through adversarial interactions that traditional scripts cannot predict.

Does parallel UI testing work across localhost and deployed sites?▼

Yes, parallel UI testing runs independent browser sessions to speed up QA across both localhost and deployed sites. It requires cookie-sync for remote sessions to maintain authenticated state during testing.

Do I need the browse CLI to run accessibility audits on my UI?▼

Yes, the browse CLI is required to execute accessibility audits. It drives a real browser to run axe-core audits, capture console errors, and validate responsive viewports during UI testing.

Why use adversarial UI testing instead of traditional automated scripts?▼

Adversarial UI testing reveals defects that traditional automated scripts miss by analyzing real browser interactions. It combines exploratory navigation with deterministic checks like axe-core to catch edge cases scripts cannot predict.