1. Contrast and Alt Text
Detects readability and image-description gaps that block perception.
Accessibility UX as measurable operations, not guesswork.
Live-site accessibility checks with report-first remediation. Audit what users actually experience, score risk clearly, and fix in priority order.
Teams often believe a site is usable because it "looks fine" or passes a few static checks. Real users still hit friction: keyboard traps, unclear forms, weak semantics, inaccessible navigation.
B1C3-UX-EVAL closes that gap. It runs a repeatable Tier 0 accessibility UX baseline on live pages, then translates findings into a practical remediation path.
The baseline model covers six high-impact layers aligned to WCAG 2.1 AA intent:
Detects readability and image-description gaps that block perception.
Checks keyboard focus visibility/order and interaction continuity.
Inspects structural meaning so assistive tools can parse content correctly.
Finds missing or conflicting accessibility metadata in interactive components.
Evaluates labels, instructions, errors, and completion flow quality.
Reviews orientation and movement clarity across the site experience.
Run automated detection on real internet pages, then manually verify key findings.
Issue impact is weighted (critical/high/medium/low) and rolled into interpretable scoring.
Generate findings in report form: per-check breakdown, examples, and location-specific patterns.
Deliver prioritized fixes so teams resolve blockers first and improve user outcomes faster.
This is the UX angle: not only finding defects, but translating technical issues into human impact and decision-ready remediation.
These two repositories are complementary and intentional:
Assessment capability. Tells you what is wrong, where risk is concentrated, and what to fix first.
Implementation capability. Utilities and patterns that help teams build/fix accessible UX in code.
Read it as a loop: evaluate with B1C3-UX-EVAL, implement with ux-tools, re-evaluate for measurable improvement.
This is a Tier 0 baseline, not a complete UX maturity model. Automated checks catch common high-impact issues, but nuanced edge cases still require manual review and contextual judgment.
That is a feature, not a flaw: B1C3-UX-EVAL is designed for reliable first-pass clarity and fast remediation planning on live properties.