What is in place
- Every page is server-rendered, so content is available without JavaScript
- Semantic headings, landmarks, and lists throughout, with one h1 per page
- Visible keyboard focus indicators on every interactive element
- Text alternatives for meaningful images; decorative images are hidden from assistive technology
- Form fields have real labels, not placeholder text standing in for them
- FAQ answers use native disclosure elements that work without scripting
- Animation respects the prefers-reduced-motion setting
