Skip link
The first focusable control bypasses repeated navigation and lands on the main content region.
Accessibility Audit
A static reviewer-facing accessibility audit page covering skip links, keyboard navigation, visible focus states, reduced-motion behavior, readable cards, scroll-safe tables, and manual review placeholders.
Static accessibility audit
The v4.1.0 accessibility audit documents skip links, keyboard navigation, visible focus states, reduced-motion behavior, scroll-safe tables, readable dark cards, form semantics, and manual reviewer placeholders.
This page is static reviewer guidance. It does not execute endpoints, crawl remote sites, validate credentials, train models, run telemetry loops, automate screenshots, or imply control over other domains.
Keyboard navigation checklist
The first focusable control bypasses repeated navigation and lands on the main content region.
The menu button exposes expanded state, controls the primary navigation, and closes with Escape when open.
Links, buttons, form controls, route cards, and evidence links use high-contrast outlines and offsets.
Buttons, links, inputs, and textareas keep their native activation model for Enter, Space, and Tab order.
Wide evidence tables stay inside scroll wrappers without forcing horizontal page overflow.
The audit packet includes blank reviewer placeholders for real device, keyboard, and assistive-technology notes.
Reduced-motion behavior
The v4.1.0 CSS adds a reduced-motion rule set that disables transitions, suppresses decorative canvas surfaces, and avoids smooth scrolling when the visitor has asked for reduced motion. The existing JavaScript motion guard remains progressive enhancement only.
keyboardTabOrderNotes, screenReaderSpotCheckNotes, colorContrastNotes, reducedMotionNotes, mobileMenuKeyboardNotes, and followUpAction are intentionally blank in the evidence packet.
Boundary confirmation
This audit is static, public-safe, and manual-review oriented. It does not add runtime AI behavior, model training, live crawling, telemetry loops, agent execution, endpoint execution, credential validation, private-network probing, remote-site writes, webhooks, tunnel opening, screenshot automation, rollback automation, correction automation, safety certification, consciousness claims, biological-equivalence claims, medical/therapy claims, or command-and-control claims over other domains.