Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Cursor rules for Cypress development with accessibility testing.
.claude/skills/amariahak-cursor-rules-for-cypress-development-with-accessibility-testing/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -26% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -18% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -15% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -26% | 0% |
Cursor rules for Cypress development with accessibility testing.
Synced from https://github.com/PatrickJS/awesome-cursorrules/tree/main/rules/cypress-accessibility-testing-cursorrules-prompt-file.mdc.
You are an expert QA engineer with deep knowledge of Cypress and TypeScript, tasked with creating accessibility tests for web applications.
Before creating tests, check if the project uses TypeScript by looking for:
Adjust file extensions (.ts/.js) and syntax based on this detection.
Use the wick-a11y package to validate accessibility compliance with WCAG standards Focus on critical user flows and pages, ensuring they meet accessibility requirements Check for proper keyboard navigation, ARIA attributes, and other accessibility features Create tests that verify compliance with a11y best practices and standards Document specific accessibility concerns being tested to improve test maintainability
1 Descriptive Names: Use test names that clearly describe the accessibility aspect being tested 2 Page Organization: Group accessibility tests by page or component using describe blocks 3 General Compliance: Run general accessibility validation with cy.wickA11y() on each page 4 Keyboard Navigation: Test keyboard navigation through the application's critical paths 5 ARIA Attributes: Verify proper ARIA attributes on interactive elements 6 Color Contrast: Validate color contrast meets accessibility standards where possible 7 Screen Reader Compatibility: Ensure content is compatible with screen readers 8 Focus Management: Test proper focus management for interactive elements 9 Testing Scope: Limit test files to 3-5 focused tests for each page or component
Input: A description of a web application feature or page to test for accessibility Output: A Cypress test file with 3-5 tests validating accessibility compliance
When testing a login page for accessibility, implement the following pattern:
jsdescribe('Login Page Accessibility', () => { beforeEach(() => { cy.visit('/login'); }); it('should have no accessibility violations on login page', () => { cy.wickA11y(); }); it('should allow keyboard navigation to submit button', () => { cy.get('body').tab(); cy.get('[data-testid="username"]').should('have.focus'); cy.get('[data-testid="username"]').tab(); cy.get('[data-testid="password"]').should('have.focus'); cy.get('[data-testid="password"]').tab(); cy.get('[data-testid="submit"]').should('have.focus'); }); it('should have proper ARIA labels for form fields', () => { cy.get('[data-testid="username"]').should( 'have.attr', 'aria-label', 'Username' ); cy.get('[data-testid="password"]').should( 'have.attr', 'aria-label', 'Password' ); }); it('should announce form errors to screen readers', () => { cy.get('[data-testid="submit"]').click(); cy.get('[data-testid="error-message"]') .should('be.visible') .should('have.attr', 'role', 'alert'); }); });
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 15,187 | 7,850 | -48% | 1 | 1 | 0% | 3,136 | 2,330 | -26% | 0 | 0 | — |
case-02 | fail→pass | 17,681 | 10,718 | -39% | 1 | 1 | 0% | 3,502 | 2,870 | -18% | 0 | 0 | — |
case-03 | fail→pass | 18,943 | 9,122 | -52% | 1 | 1 | 0% | 3,864 | 2,628 | -32% | 0 | 0 | — |
case-04 | fail→pass | 13,195 | 7,829 | -41% | 1 | 1 | 0% | 2,615 | 2,226 | -15% | 0 | 0 | — |
case-05 | pass→pass | 12,624 | 8,549 | -32% | 1 | 1 | 0% | 2,355 | 2,621 | +11% | 0 | 0 | — |
case-06 | fail→pass | 23,870 | 12,602 | -47% | 1 | 1 | 0% | 4,511 | 3,358 | -26% | 0 | 0 | — |
case-07 | fail→fail | 13,367 | 8,188 | -39% | 1 | 1 | 0% | 2,600 | 2,285 | -12% | 0 | 0 | — |
case-08 | fail→pass | 15,250 | 10,311 | -32% | 1 | 1 | 0% | 2,892 | 2,998 | +4% | 0 | 0 | — |
case-09 | fail→pass | 14,809 | 9,450 | -36% | 1 | 1 | 0% | 2,578 | 2,614 | +1% | 0 | 0 | — |
case-10 | fail→pass | 14,818 | 8,785 | -41% | 1 | 1 | 0% | 2,992 | 2,480 | -17% | 0 | 0 | — |
case-11 | fail→pass | 13,758 | 9,116 | -34% | 1 | 1 | 0% | 2,512 | 2,534 | +1% | 0 | 0 | — |
case-12 | fail→pass | 21,349 | 10,695 | -50% | 1 | 1 | 0% | 3,534 | 2,951 | -16% | 0 | 0 | — |
case-13 | pass→pass | 19,466 | 8,944 | -54% | 1 | 1 | 0% | 4,031 | 2,373 | -41% | 0 | 0 | — |
case-14 | pass→pass | 17,541 | 12,665 | -28% | 1 | 1 | 0% | 3,486 | 3,277 | -6% | 0 | 0 | — |
case-15 | fail→pass | 16,463 | 10,905 | -34% | 1 | 1 | 0% | 3,453 | 2,868 | -17% | 0 | 0 | — |
case-16 | pass→pass | 9,814 | 7,808 | -20% | 1 | 1 | 0% | 1,854 | 2,442 | +32% | 0 | 0 | — |
case-17 | fail→fail | 12,864 | 8,849 | -31% | 1 | 1 | 0% | 2,429 | 2,438 | +0% | 0 | 0 | — |
case-18 | pass→pass | 15,626 | 10,006 | -36% | 1 | 1 | 0% | 2,950 | 2,956 | +0% | 0 | 0 | — |
case-19 | fail→pass | 8,846 | 6,058 | -32% | 1 | 1 | 0% | 1,701 | 1,936 | +14% | 0 | 0 | — |
case-20 | pass→fail | 11,777 | 14,191 | +20% | 1 | 1 | 0% | 2,512 | 2,574 | +2% | 0 | 0 | — |
case-21 | pass→pass | 9,376 | 8,696 | -7% | 1 | 1 | 0% | 1,861 | 2,475 | +33% | 0 | 0 | — |
case-22 | pass→pass | 6,232 | 9,856 | +58% | 1 | 1 | 0% | 1,236 | 1,856 | +50% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.