Title II Snapshot
How the Title II Snapshot works
What the free Snapshot checks, how it fetches your pages, what it keeps, and what it can't tell you.
Your deadline
The deadline comes from the type of public body and the population band you choose, under 28 CFR 35.104 and 35.200 with the compliance dates as extended by the interim final rule published April 20, 2026. Public entities with a total population of 50,000 or more: April 26, 2027. Under 50,000, and all special district governments: April 26, 2028. If you choose “Not sure”, both dates are shown. See laws and guidelines for details, and confirm your date with your counsel.
Which pages we check
We check the page you enter, then up to 9 more pages on the same site (with or without www), taken from your sitemap.xml first, then your main navigation, then other links on the page. We never follow links to other sites, never log in and never submit forms.
Every check, and its known limits
Each line in the results is a basic signal read from the page HTML, without running scripts. There is no score and no grade.
| Check | What we look for | What it can miss or get wrong |
|---|---|---|
| Page language | Whether the <html> element has a lang attribute. | We don't check that the language code is correct, or language changes inside the page. |
| Page title | Whether the page has a non-empty <title>. | We don't judge whether the title describes the page. |
| Images with no alt attribute | Counts <img> elements without an alt attribute, skipping images hidden from assistive technology. | An empty alt="" is fine for decoration and is not counted, but we can't tell whether an image that has alt text needs a better description. Images added by scripts or as CSS backgrounds aren't seen. |
| Form fields with no label | Counts visible inputs, selects and text areas with no <label>, aria-label, aria-labelledby or title. | Fields built by scripts aren't seen. A label can exist but be unclear. |
| Links and buttons with no text | Counts links and buttons with no text, no image alt text and no accessible name. | Names added by scripts aren't seen, and vague text such as “click here” isn't flagged. |
| Zoom disabled | Whether the viewport tag sets user-scalable=no or a maximum scale below 2. | Some browsers ignore these settings, but the tag still signals a problem. |
| Frames with no title | Counts visible <iframe> elements with no title or aria-label. | Frames added by scripts aren't seen. |
| Autoplaying media | Counts <video> and <audio> with autoplay, and embedded players with autoplay=1 in their address. | We can't tell whether the media has sound, lasts more than 3 seconds, or has a pause control. |
| Accessibility statement | Whether any checked page links to something with “accessibility” in the link text or address. | We don't read or judge the statement itself. |
| PDF signals | Counts PDFs linked from the checked pages and reads up to 5 on your site (10 MB each). For each we look for tagged structure (StructTreeRoot with MarkInfo marked true), a document title, a document language and selectable text. | Tags can exist and still be wrong, and text can be selectable but in the wrong reading order. Encrypted, damaged or very large files show “couldn't read”, which is never counted as a problem. |
How we fetch your site
- We read robots.txt first and follow it. Our user agent token is
ValleyFillSnapshot; rules for*apply if there are none for it. If robots.txt returns a server error or can't be reached, we treat the whole site as off limits and check nothing. - We identify ourselves:
ValleyFill Snapshot/1.0, with a link to this page and hello@valleyfill.com. - At most 10 pages and 5 PDFs per snapshot, at most one request per second to your site, 8 seconds per request and 45 seconds in total.
- Each site can be checked once an hour, whoever asks.
- Only public web addresses: we refuse IP addresses, private and internal networks, non-standard ports and addresses with passwords, and we re-check every redirect.
To opt out, add User-agent: ValleyFillSnapshot and Disallow: / to your robots.txt, or email hello@valleyfill.com.
What we keep
Nothing about the site you checked or what we found. Results go straight to your browser and are not saved, published or indexed. To enforce the daily limits we keep, for one day, a counter keyed by a salted one-way hash of your IP address, one total count for the day, and for one hour a salted hash of each checked site name for the once-an-hour rule. Our web host keeps standard server logs, as described in our privacy notice. There are no cookies and no third-party scripts on these pages; ValleyFill only counts visits anonymously, as described in the privacy notice.
What this can't tell you
This is an automated, partial check of a few public pages. Automated tools catch roughly a third of accessibility issues. It is not an audit, not an Accessibility Conformance Report (ACR), and not legal advice. A clean result does not mean your site conforms to WCAG 2.1 AA or meets the Title II rule. It reads page HTML only, without running scripts, so it can miss problems and can flag things that are fine.
Accessibility overlays and widgets don't make a site conform, and the FTC has acted against overlay claims. For a fuller picture, our automated audit covers up to 50 pages and 10 PDFs. Need a full audit? Contact us.