Architecture of Grace · Accessibility
An honest accessibility statement
What is true by design, what is not done yet, and how to tell me when something gets in your child's way. No conformance badge here — a straight account instead.
Last reviewed: September 20, 2026 · Written by the founder · What was tested
I built this system as a classroom teacher, and many of the students I built it for learn differently — that shaped choices from the beginning. But I will not claim a standard I have not verified. The site has now been tested against WCAG 2.1 Level AA, by me, with automated tools — and that is not the same as conformance. Below is exactly what was tested, exactly what it found, and exactly what it cannot tell you. What I can promise is the full account and a commitment to fix what you find.
What is true by design
- No logins and no time limits. Nothing expires, nothing locks, and no activity races a clock. A student can take an hour on what another finishes in five minutes, and nothing on the page treats that as anything but normal.
- No autoplay and no flashing content. Nothing moves, sounds, or flashes without the student asking it to.
- Hints are part of the lesson. Hint buttons sit in the open, framed exactly this way to students: asking for a hint is not cheating. A student who needs the scaffold takes it without penalty or flag.
- Print alternatives for most activities. Most activities have printable versions, so a student who works better on paper — or with a reader, a scribe, or larger print made on the school copier — has a first-class path, not a workaround.
- Layouts tolerate browser zoom and large text. Pages are built to stay usable when the browser's zoom or text-size settings are turned well up.
- Works offline. Once loaded, the core tools keep working without a connection, so a spotty network doesn't strand a student mid-activity.
Known gaps, stated honestly
- No built-in text-to-speech yet
- The site does not yet read content aloud on its own. Until it does, the device's built-in reader works well with these pages: on Chromebooks, turn on Select-to-Speak (Settings › Accessibility) and students can hear any selected text; on iPad, turn on Spoken Content (Settings › Accessibility) for the same. If your school's devices are managed, IT can enable these for a student in the admin console.
- Screen-reader coverage not yet audited
- The pages are plain HTML and much of it will read sensibly, but I have not systematically tested with VoiceOver, ChromeVox, or NVDA, and I will not guess at the result. If your child uses a screen reader, I would genuinely value hearing what works and what doesn't — that report will drive the audit.
- No independent audit
- Everything on this page was measured by me, on my own work. An outside evaluator has not looked at it. If you are an accessibility specialist willing to review any part of this site, the materials and the test method are yours for the asking, and what you find will be published here whichever way it points.
What was tested, and what it found
Standard: WCAG 2.1 Level A and AA. Tested: 20 September 2026, with
axe-core in Chromium at 1280 px, plus a static pass over every page in the site.
Tested by: me. This is a self-assessment, not an independent audit.
Scope of the browser pass: 18 pages chosen to cover every kind of page here
— the home page, the standards crosswalk, a room lesson page, a student
worksheet, a subject hub, a printable handout, the Calming Corner, the family
letters, a daily-drops door, the error page, and all six privacy, terms, data and
agreement pages.
Result: zero automated violations on all eighteen, after the repairs below.
What the testing actually fixed
- 86 color-contrast failures across 8 pages. Every one of them was on the
privacy, terms, data-agreement and school-facing pages, which carry their own copy
of the palette. The worst was the small gold uppercase label at the top of each
page: 2.47:1 where 4.5:1 is the bar. The most common was a near miss —
gold on the cream callout at 4.41:1, across 65 places. Both are corrected
and re-measured.
- 95 pages had no heading element at all — not a missing top-level
heading, none at all, so anyone navigating by headings got a flat wall. Every one
now has its own visible title as a real heading.
- 171 worksheet fields had no name a screen reader could announce. A student
would have heard "edit text, blank" with no way to tell which question they were
answering. Every one is now wired to the visible question text, so the name follows
the problem and the language.
- 24 Calming Corner tools could not be reached by keyboard — breathing,
grounding, safe-place, animal yoga. The tools most directly built for students with
disabilities were the ones a keyboard could not open. Fixed.
- Two images had no alt text in the whole site — generated QR codes on
the printable link sheet. They now name where the code goes.
Across all 444 pages
- Every page declares its language, and every page has a title.
- No page blocks pinch-zoom. Not one sets user-scalable=no.
Where a mobile keyboard causes zoom trouble, it is solved by making the field bigger
— never by taking zoom away from the students who need it.
- Form fields resolve to a real visible label rather than a bolted-on attribute, so
they stay clickable and cannot drift out of sync with the question.
What this does not establish
I would rather undersell this than have a district discover the gap after adopting.
- Automated testing catches a minority of accessibility problems. A tool can
see that alt text exists; it cannot see whether the alt text is right. It can
see that a heading exists; it cannot tell you whether it describes the section. It
cannot tell you whether a lesson makes sense read aloud, in order, by a machine.
- No screen reader has been tested against this site. That has not changed and
I will not guess at the result.
- 426 of the 444 pages were not individually loaded in a browser. They share
templates and a stylesheet with the ones that were, and the static pass covered all
of them — but "covered by the same template" is an argument, not a measurement.
- This is my own work, checked by me. No independent auditor has reviewed it,
and there is no VPAT or third-party conformance report behind it.
- So: no conformance claim. Tested, repaired, and documented is what I can
honestly say. If you need a conformance statement from an independent evaluator for
your procurement file, tell me and I will say plainly that one does not exist yet.
For districts doing their own review. Your own obligations under
the Department of Justice's Title II web rule run on your own timetable —
April 2027 for entities serving 50,000 or more, April 2028 for smaller ones.
This page exists so your review can start from a straight account rather than a badge.
If your evaluation turns up something this page missed, that is exactly the report I want.
If something gets in the way
Tell me, and I will fix it. If your child — or your student, or you — hits a barrier anywhere on this site, write to me and describe what happened: which page, which device, and what stood in the way. Reports go to the founder directly, not a ticket queue, and barriers reported by families of students with disabilities go to the top of the list. This page exists because a parent went looking for accessibility information and found none; I would rather hear about the next gap from you than have you discover it in silence.