Gridex Inc. · Accessibility Conformance Report

Accessibility exception request accessibility report

WCAG 2.1 Levels A and AA · version 1.0 · October 4, 2026

Self-evaluation organized according to the criterion-level reporting structure of the VPAT 2.5 WCAG edition. This is an ACR-style report, not an ITI certification or an independently issued audit. The WCAG 2.2 additions in the broader VPAT 2.5 template are outside this requested WCAG 2.1 scope.

Open the sample

Product and evaluation scope

Four-step HTML/JavaScript demonstration, including validation, review, saved browser drafts, confirmation dialogs, restore, cancellation, submission confirmation and browser-storage failure.

Supplier: Gridex Inc. · Contact: ryan@gridex.dev. Content is synthetic. Power Apps, SharePoint tenant integration and real client performance are outside scope.

Results and limitations

34 Supports; 2 Partially Supports; 14 Not Applicable. No unqualified claim of full WCAG 2.1 AA conformance is made while spoken-output verification remains incomplete.

All completed axe scans and reproducible web keyboard tests pass after remediation. Chromium and WebKit accessibility snapshots and native Safari semantics were inspected. VoiceOver was started on macOS 26.5, but its spoken output and caption transcript could not be reliably observed through the available native tooling (VoiceOver app inspection timed out). No NVDA test was performed. An accessibility tree is not equivalent to a screen reader speech test.

Production verification: the published sample passed the complete keyboard journey (where applicable) and axe scans in Chromium and WebKit. All eight suite/sample/report HTML pages and internal links returned successful responses. See the live axe evidence below.

Methods and tools

axe-core 4.13.0 through @axe-core/playwright 4.13.0; WCAG 2 A, AA and WCAG 2.1 A, AA tags.
Playwright 1.63.0, headless Chromium 153.0.8010.12 and WebKit 26.6. Synthetic keyboard events only for form interaction; no mouse or programmatic filling in the keyboard journey.
320 CSS pixel viewport; 200% CSS zoom at 1280px; text-spacing overrides; screenshots and computed overflow checks.
Color-pair contrast calculation using WCAG relative luminance, plus axe contrast rules.
DOM, accessible name, landmark, heading and table review; native Safari 26.5 semantic inspection.

Conformance terms

Supports: Completed tests and inspection support the criterion within the documented evaluation scope.
Partially Supports: The criterion has an identified limitation or incomplete assistive-technology verification, stated in the remarks. This report distinguishes unverified behavior from known defects.
Not Applicable: The work does not contain the feature addressed by the criterion.

WCAG 2.1 Level A

Level A criterion results
Criterion	Conformance	Remarks and test basis
1.1.1 Non-text Content	Supports	No informative images; controls have text labels. Native checkbox meaning is identified by its visible label.
1.2.1 Audio-only and Video-only (Prerecorded)	Not Applicable	No audio-only or video-only media.
1.2.2 Captions (Prerecorded)	Not Applicable	No prerecorded synchronized media.
1.2.3 Audio Description or Media Alternative (Prerecorded)	Not Applicable	No prerecorded synchronized media.
1.3.1 Info and Relationships	Supports	HTML headings, native controls, explicit labels, definition lists, list semantics and scoped table headers were reviewed in the DOM and accessibility tree.
1.3.2 Meaningful Sequence	Supports	DOM order matches the visual reading order. No CSS reordering. PDF structure-tree traversal and extracted text were inspected where provided.
1.3.3 Sensory Characteristics	Supports	Instructions use names, dates and text, rather than position, shape or color alone.
1.4.1 Use of Color	Supports	Statuses, errors and risk levels are written as text. Links are underlined.
1.4.2 Audio Control	Not Applicable	No audio playback.
2.1.1 Keyboard	Supports	Tab, Shift+Tab, Enter and Space operate the web controls. Headless Playwright keyboard tests provide reproducible interaction records; document reading and downloads use the reader application.
2.1.2 No Keyboard Trap	Supports	Native modal dialogs with explicit first/last-button Tab wrapping permit Escape to exit. Static documents introduce no traps.
2.1.4 Character Key Shortcuts	Not Applicable	No single-character application shortcuts.
2.2.1 Timing Adjustable	Not Applicable	No session time limit or timed interaction. An exception expiration date is business content, not an interaction time limit.
2.2.2 Pause, Stop, Hide	Not Applicable	No moving, blinking, scrolling, or auto-updating content.
2.3.1 Three Flashes or Below Threshold	Not Applicable	No flashing content.
2.4.1 Bypass Blocks	Supports	A first-in-order skip link targets the main landmark. HTML heading navigation, Word heading styles and PDF heading tags support bypassing sections.
2.4.2 Page Titled	Supports	Each page has a specific title. DOCX core properties and PDF metadata identify the document.
2.4.3 Focus Order	Supports	Tab order follows DOM order. Focus is moved to error summaries, new step headings and submission confirmation; native dialog closure restores the trigger.
2.4.4 Link Purpose (In Context)	Supports	Navigation, reports and downloads have descriptive link text.
2.5.1 Pointer Gestures	Not Applicable	No multipoint or path-based gestures.
2.5.2 Pointer Cancellation	Supports	Actions activate on standard click/release, not pointer down. Draft deletion has a confirmation step.
2.5.3 Label in Name	Supports	Accessible names contain their visible label text; verified by axe and accessibility snapshots.
2.5.4 Motion Actuation	Not Applicable	No motion-actuated functions.
3.1.1 Language of Page	Supports	HTML uses lang="en". Word default language and PDF catalog language are en-US.
3.2.1 On Focus	Supports	Focusing a field or link does not navigate or submit.
3.2.2 On Input	Supports	Typing and selecting values do not change steps. Explicit Continue or Submit activates the change.
3.3.1 Error Identification	Supports	Empty fields, missing DAC consultation, invalid email, invalid calendar dates and date-order errors produce an error summary linked to inline errors; invalid fields use aria-invalid.
3.3.2 Labels or Instructions	Supports	Required fields are labeled in text and programmatically required. Field help is associated with aria-describedby. Dates use explicit YYYY-MM-DD instructions.
4.1.1 Parsing	Supports	HTML DOM was checked for duplicate IDs, invalid label targets and broken describedby references. axe checks pass. Documents use well-formed XML and tagged structure.
4.1.2 Name, Role, Value	Partially Supports	Native HTML controls and links expose names and roles in Chromium and Safari accessibility trees. Actual VoiceOver spoken output could not be reliably captured by the available native tooling, so this conservative report leaves assistive-technology interoperability partially verified.

WCAG 2.1 Level AA

Level AA criterion results
Criterion	Conformance	Remarks and test basis
1.2.4 Captions (Live)	Not Applicable	No live synchronized media.
1.2.5 Audio Description (Prerecorded)	Not Applicable	No prerecorded video.
1.3.4 Orientation	Supports	No orientation lock; 320px narrow viewport and landscape rendering were checked.
1.3.5 Identify Input Purpose	Supports	Requester name and email inputs have autocomplete="name" and autocomplete="email".
1.4.3 Contrast (Minimum)	Supports	axe color-contrast checks and calculated color pairs pass. See contrast.json for ratios.
1.4.4 Resize text	Supports	200% CSS zoom layout checks in Chromium and WebKit pass without horizontal overflow. Native Safari 200% was inspected on the form. Document view uses the reader application zoom.
1.4.5 Images of Text	Supports	All essential prose is real text. The guide includes a configuration relationship diagram with an equivalent textual description; no essential text is provided solely as an image.
1.4.10 Reflow	Supports	All HTML views tested at 320 CSS pixels without document-wide horizontal scrolling. Tagged PDFs have fixed page layouts; equivalent HTML is the recommended reflow reading path.
1.4.11 Non-text Contrast	Supports	Borders, control boundaries and focus outlines use contrasting colors. Contrast ratios are calculated in contrast.json.
1.4.12 Text Spacing	Supports	Text spacing overrides (1.5 line height, .12em letter spacing, .16em word spacing and 2em paragraph spacing) were tested without clipped content or horizontal overflow in HTML.
1.4.13 Content on Hover or Focus	Not Applicable	No custom hover or focus-triggered supplemental content.
2.4.5 Multiple Ways	Supports	Form steps are sequential process steps and exempt from multiple-ways requirements. The suite index and common navigation locate the entry page.
2.4.6 Headings and Labels	Supports	Headings identify request, risk, approvals, renewals and administrative tasks. Labels identify field purpose and required state.
2.4.7 Focus Visible	Supports	Keyboard focus has a 3px contrasting outline with a 4px offset. Keyboard navigation and screenshots were inspected.
3.1.2 Language of Parts	Not Applicable	Content is entirely English; technical names are proper names or terms.
3.2.3 Consistent Navigation	Supports	The same sample navigation appears in the same order on pages and reports.
3.2.4 Consistent Identification	Supports	Download links, navigation labels and headings use consistent identification.
3.3.3 Error Suggestion	Supports	Errors give specific corrective instructions and calendar examples. Error links focus the affected field.
3.3.4 Error Prevention (Legal, Financial, Data)	Supports	Final review and confirmation precede demonstration submission. Saving, restoring and deleting a browser draft require confirmation; Keep editing and Escape cancel the action. No legal or financial transaction occurs.
4.1.3 Status Messages	Partially Supports	Draft status uses role="status", aria-live="polite" and aria-atomic="true"; state is present in the DOM and axe reports no violation. Actual VoiceOver announcement delivery remains unverified.

Reproducible evidence

Remediation and remaining work

Read the remediation log. Complete an actual screen reader review of the documented scenarios before replacing the conservative Partially Supports conclusions with Supports.