a11y-agent
Foundation — loads automatically
Deep accessibility knowledge for every code review and
implementation. Loaded into your agent's context whenever you
build UI.
10 reference files:
semantic-html
keyboard
aria
forms
focus
testing
gotchas
screen-readers
routing
widgets
/a11y-audit
Command
Structured accessibility assessment. Walks through keyboard →
tools → contrast → zoom → screen readers. Stores
results for tracking progress.
keyboard
headings
axe scan
contrast
zoom
screen readers
/a11y-fix
Command
Fix accessibility issues from audit, scan, or review findings.
Explains rationale, verifies each fix, flags what to test
manually.
div→button
labels
focus
mega menu
routing
modals
/a11y-scan
Command
Run automated axe-core scans via Playwright. Captures WCAG
violations, compares with previous runs, tracks remediation
progress.
axe-core
playwright
WCAG violations
trend tracking
/a11y-test
Command
TDD for accessibility. Write tests that fail when a11y breaks,
then fix. Jest + Testing Library, Cypress + cypress-axe, Puppeteer
snapshots.
jest
testing-library
cypress-axe
CI
/a11y-component
Command
Scaffold accessible components with correct ARIA, keyboard
handling, and focus management baked in from the start.
modal
menu
tabs
accordion
/a11y-review
Command
Lightweight code review for accessibility. Checks markup, ARIA,
keyboard handling, and test coverage in changed files.
markup
ARIA
keyboard
styles
tests
/a11y-advocate
Command
Build accessibility culture. Generate statements, prioritize
issues, write tickets, create team checklists, produce risk
assessments.
statement
prioritize
ticket
checklist
risk