Conduct WCAG 2.2 accessibility audits with automated testing, manual verification, and remediation guidance. Use when auditing websites for accessibility, fixing WCAG violations, or implementing accessible design patterns.
Install with Tessl CLI
npx tessl i github:wshobson/agents --skill wcag-audit-patternsOverall
score
81%
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillEvaluation — 96%
↑ 1.05xAgent success when using this skill
Validation for skill structure
Form accessibility remediation
Visible labels
100%
100%
Required field indication
100%
100%
aria-invalid on error fields
100%
100%
aria-describedby links error
100%
100%
role='alert' on errors
100%
100%
Error describes problem
63%
75%
Status uses aria-live polite
100%
100%
No color-only error state
100%
100%
Focus outline present
75%
100%
Text contrast ratio
100%
100%
Format hints provided
56%
89%
Without context: $0.3137 · 1m 26s · 11 turns · 12 in / 5,227 out tokens
With context: $0.6152 · 2m 18s · 22 turns · 317 in / 8,640 out tokens
Custom widget ARIA and keyboard patterns
Correct ARIA role
100%
100%
aria-expanded state
100%
100%
aria-labelledby or aria-label
100%
100%
tabindex on custom element
100%
100%
Enter and Space activate
100%
100%
Escape closes widget
100%
100%
Arrow keys navigate options
100%
100%
preventDefault on arrows
100%
100%
prefers-reduced-motion
0%
80%
Focus outline not removed
88%
100%
ARIA state updates on interaction
100%
100%
Native HTML preferred
75%
100%
Without context: $0.4121 · 1m 43s · 15 turns · 64 in / 7,135 out tokens
With context: $0.7029 · 2m 42s · 24 turns · 23 in / 10,325 out tokens
Page structure audit and skip navigation
Skip link present
88%
63%
Skip link class
100%
100%
Main landmark id
83%
67%
HTML lang attribute
100%
100%
Heading hierarchy fixed
100%
100%
ARIA landmarks used
100%
100%
No generic link text
100%
100%
Descriptive link text
100%
100%
axe-core WCAG tags
70%
70%
axe-core package used
100%
100%
Audit report produced
100%
100%
Manual testing noted
100%
100%
No zoom disabled
100%
100%
Without context: $0.7744 · 3m 43s · 25 turns · 26 in / 12,363 out tokens
With context: $1.0570 · 3m 50s · 32 turns · 78 in / 13,815 out tokens
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.