visual-qa

name	visual-qa
description	Visually verify the rendered dashboard application by running it, checking logs, and reviewing screenshots to evaluate visual appeal, UX quality, and layout correctness.
metadata	{"author":"Antigravity","version":"1.0","license":"MIT"}

Visual QA & UX Evaluation

Visually verify and evaluate Shiny dashboards against professional modern design standards. Dashboards should center on user empathy, ruthlessly prioritize metrics, and structure clear visual hierarchies so users can absorb critical data within five seconds. An effective dashboard is a functional, interactive story rather than a cluttered dumping ground for charts.

Progressive Loading

Read this file to understand the verification workflow, then load only the references needed for your specific evaluation task:

Evaluation Target	Reference Document
User Empathy, Ruthless Metric Prioritization, and the 5-Second Rule	references/principles.md
Accessibility, Color Contrast, Legible Typography, and Graceful State Handling	references/accessibility_and_ux.md

Hands-on Visual Self-Evaluation Workflow

To evaluate visual quality, layout correctness, chart visibility, text readability, colors, and styling, you must run the app and take screenshots.

Python Verification Pipeline

nohup python -m shiny run app.py --port 8000 > /tmp/app.log 2>&1 &
tail -n 80 /tmp/app.log
python /home/user/project/.tools/screenshot_helper.py
tail -n 80 /tmp/app.log
pkill -f "shiny run" || true

R Verification Pipeline

nohup Rscript -e "shiny::runApp('app.R', port=8000, launch.browser=FALSE)" > /tmp/app.log 2>&1 &
tail -n 80 /tmp/app.log
python3 /home/user/project/.tools/screenshot_helper.py
tail -n 80 /tmp/app.log
pkill -f "Rscript" || true