Skill UI

Browser automation expert for debugging flaky tests and running long-lived scrapers using Playwright/Puppeteer, focusing on robust selectors, auto-waits, isolation, and anti-detection practices.

Guides an agent through C++17/20 testing with CMake, CTest, and GoogleTest/Mock, covering new or flaky tests, CI gates, coverage builds, and sanitizer diagnostics to keep regression safety in place.

Guides writing Playwright-based E2E tests, configuring fixtures, debugging flaky browser flows, and integrating reporting/CI to keep automation stable and maintainable.

Tracks and runs regression suites, automating execution of critical tests and analyzing outcomes so Claude highlights failures, flaky checks, and stability impacts before deployments.

Diagnose and fix failing or flaky Playwright tests via a systematic reproduce-trace-categorize-fix workflow, validate stability by repeated runs, and recommend CI safeguards to prevent regressions.

This skill systematically reviews test code to identify and correct common anti-patterns. It detects issues such as flaky tests, overly mocked dependencies (over-mocking), brittle assertions, excessive test size (The Giant), and structural dependencies. It explains the flaw, provides best practices, and suggests refactored, reliable implementations, helping maintain high test quality and stability.