Frontend Behavioral QA

0•

Performs automated browser-based behavioral QA and visual audits to catch UI regressions in running web applications.

Frontend Behavioral QA acts as Stage 7 in the eng-flow production track, focusing on exercising the running web application in a browser rather than reviewing code diffs. It drives browser sessions via Playwright automation (or degrades smoothly to a guided manual checklist) to catch layout anomalies, broken interactions, unhandled form inputs, and console errors. To keep conversation context clean, browser navigation and DOM snapshots are executed in isolated foreground subagents, surfacing structured repro steps, screenshots, and severity ratings directly into a centralized QA report without attempting inline code fixes.

Key Features

01Automated browser interaction via Playwright with guided manual fallbacks
02Standardized severity tagging (Critical, Required, Nit) and structured QA report generation
030 GitHub stars
04Comprehensive verification covering visuals, forms, empty/error states, and console logs
05Multiple test scopes: diff-aware branch testing, full app sweep, or quick smoke check
06Isolated subagent execution to prevent context bloat from snapshots and logs

Use Cases

01Conducting diff-aware behavioral smoke tests on feature branches before code freeze
02Generating audit-ready QA bug reports and severity-tagged issue logs for implementation handoff
03Validating frontend route modifications and UI interactions on local development or preview servers