Sample Accessibility Audit: Own-Site Demonstration

A dated own-site accessibility audit demonstration using actual axe-core evidence from JubilantWeb.com, with WCAG references, impact, fixes, status, and limits.

This is an OWN-SITE DEMONSTRATION based on JubilantWeb's September 22, 2026 programmatic audit evidence. It is not a customer audit, current full-site result, certification, legal opinion, or claim that the site has zero issues. The historical findings below remain IDENTIFIED because no dated retest evidence is attached here.

Prepared 2026-09-22 · Evidence and scope dates are identified below

Evidence boundary

The source is the completed September 22, 2026 own-site revised audit: its axe-results.json is approximately 190 MB. Its summary.json records 5,188 WCAG 2.1 A/AA axe occurrences across five rule families and 2,267 failed scripted assertions. Repetition across routes, devices and states means these totals are not independent defect counts. The same summary records 179,864 incomplete WCAG 2.1 occurrences; incomplete means human review is unresolved, not that the content passed. The original partial run remains a separate historical record, not a later September 28 retest.

The target was a Replit preview URL. This page shows four exact finding examples and preserves their historical status. Changes made after capture do not resolve these records without a dated retest.

Finding 1 — unnamed icon submit button

WCAG reference: 4.1.2 Name, Role, Value
Priority: Critical axe impact; high-confidence component fix
Component and scope: Attorney Toolkit send control; /attorney-toolkit, desktop, state open:4:Ask AI about The Midnight Test
User impact: A screen-reader user may encounter a button without a name that explains its purpose.
Before evidence: Captured 2026-09-22T19:51:02.404Z. Rule button-name; selector .h-9; element test ID button-ai-send-midnight-test. axe reported no discernible text or naming attribute.
Recommended fix: Add a stable accessible name to the shared icon-button implementation, then rerun the rule in every relevant state and confirm the announced name through scoped human testing.
Snapshot status: IDENTIFIED — NOT RETESTED

Finding 2 — keyboard access to horizontal overflow

WCAG reference: 2.1.1 Keyboard and 2.1.3 Keyboard (No Exception), as tagged by axe
Priority: Serious axe impact; repeated shared pattern
Component and scope: Capture Layer comparison region; /capture-layer, mobile, default state
User impact: A keyboard user may be unable to reach and operate a horizontally scrollable content region.
Before evidence: Captured 2026-09-22T20:07:21.370Z. Rule scrollable-region-focusable; selector .overflow-x-auto; source metadata referenced capture-layer.tsx:1544.
Recommended fix: Confirm real overflow, then provide keyboard access with an appropriate label and visible focus, or restructure the content so it reflows without a separate scroller.
Snapshot status: IDENTIFIED — NOT RETESTED

Finding 3 — link distinguished only weakly from text

WCAG reference: 1.4.1 Use of Color
Priority: Serious axe impact; isolated static-page finding
Component and scope: External HBR citation; /intake-reality-check.html, desktop, default state
User impact: A reader with low vision or color-perception differences may not recognize the inline citation as a link.
Before evidence: Captured 2026-09-22T20:06:00.226Z. Rule link-in-text-block; selector a[target="_blank"]. axe reported 1.04:1 contrast with surrounding text and no non-color distinction.
Recommended fix: Add a persistent non-color cue such as an underline and verify sufficient contrast against surrounding text.
Snapshot status: IDENTIFIED — NOT RETESTED

Finding 4 — CTA text contrast

WCAG reference: 1.4.3 Contrast (Minimum)
Priority: Serious axe impact
Component and scope: Cloud Infrastructure booking CTA; /cloud-infrastructure, desktop, default state
User impact: Low-vision users may have difficulty reading the action label.
Before evidence: Captured 2026-09-22T19:50:24.844Z. Rule color-contrast; selector .bg-[#4285f4]. axe measured white text on #4285f4 at 3.56:1 where 4.5:1 was expected.
Recommended fix: Adjust the foreground, background, size, or weight, then measure the rendered state again.
Snapshot status: IDENTIFIED — NOT RETESTED

Limitations and pending work

This is a four-record excerpt, not the complete approximately 190 MB artifact.
axe incomplete contrast checks remain unresolved review items and are not reported as passes.
Scripted Playwright assertions are not human keyboard observations.
Hands-on NVDA and VoiceOver testing was not performed.
PDFs, authenticated experiences, and third-party transaction flows require separate scope and a confirmed tester.
No finding is marked resolved, and this sample is not a certification or statement of current full-site conformance.

Read how the testing stages and status rules work.

Sources

Request a website accessibility scan

We review the submitted site and return a plain-English summary. This is a request for review, not an instant public scanner or legal advice.

Request a scan

Frequently Asked Questions

Is this a customer audit?

No. It is an own-site demonstration using historical programmatic evidence captured from a JubilantWeb preview environment on September 22, 2026.

Are these findings resolved now?

This sample does not make that claim. Every shown record preserves the historical status IDENTIFIED — NOT RETESTED. A resolved label requires dated retest evidence.

Does this sample prove the whole site fails or passes WCAG?

No. It shows selected detected patterns from finite routes, states, viewports, and automated or scripted methods. It is neither a full human audit nor proof of conformance.