The YRA-100k Metatesting Study

YRA Tech is running a public research study measuring how well automated web-accessibility testing engines agree with each other across a large, representative sample of the web. It replicates and extends prior published work (Pool et al., arXiv:2304.07591). If your site is in the sample, our crawler loaded a small number of your public pages, ran read-only accessibility checks against the returned HTML, and recorded aggregate findings. We do not republish your content — only aggregate, engine-comparison statistics.

Operated by YRA Tech. Questions or opt-out: [email protected].

Identifying the study crawler

The study crawler self-identifies with a stable User-Agent (unlike our product scanner, it does not emulate devices):

YRA-100k-Study/1.0 (+https://yratech.com/100k)

It also sends X-YRA-Scanner: YRA-100k-Study/1.0 and egresses from the same published ranges as the product scanner — /scanner-ips.json, forward-confirmable via scan-*.yratech.com reverse DNS.

What it does — and doesn’t

• Honors robots.txt. The study crawler is a discovery crawler and respects robots.txt. Disallow YRA-100k-Study to exclude your site.
• Reads a handful of public pages per site, at a polite rate. No login, no forms, no crawling behind authentication.
• Runs read-only checks. It does not modify, probe, or stress your site, and makes no attempt to evade bot detection.
• Publishes only aggregates. Results are engine-agreement statistics across the whole corpus; individual sites are not named in outputs.

Opting out

Any of these removes your site from the study:

• Disallow YRA-100k-Study in your robots.txt.
• Email [email protected] with your domain — we’ll exclude it and drop any already-recorded data for it.

Our product scanner (authorized, user-initiated scans) is documented separately at /scanner.