How to identify it
Mozilla/5.0 (compatible; A11ySignalBot/1.0; +https://a11ysignal.com/bot)
What it does
readshtmlIt renders pages and reads the DOM. It never submits a form, follows a logout link, or sends anything but GET requests.GET only
obeysrobots.txtDisallowed paths are not fetched, and Crawl-delay is respected up to five seconds.always
limitsconcurrencyTwo pages at a time. A hundred-page weekly scan is less traffic than a single visitor browsing a catalogue.2 pages
skipsassetsMedia and fonts are blocked at the network layer — they cannot change an accessibility verdict and they would waste your bandwidth.media, fonts
storessnippetsFor each failing element we keep the CSS selector and a truncated snippet of its markup, so the customer can find the line. Nothing else is retained.1 KB max
Blocking it
Add this to your robots.txt and it will stop on the next scan:
User-agent: A11ySignalBot Disallow: /
Worth checking first whether a colleague set the scan up — blocking it will not remove the accessibility problems, only the report about them.
Something looks wrong
If the crawler is behaving badly — too fast, hitting something it should not, ignoring robots.txt — write to [email protected] with the timestamps and paths, and we will look the same day.