Guides
The exposure scanner
Nine checks, one score, and what each one costs you if it fails.
#How it scores
Give it any link page, yours or anyone's. It fetches the page the way a crawler does, renders it the way a phone does, and runs nine checks. Each check has a weight; the score is what is left out of 100 after the failures. The public widget at /scan shows three checks. The dashboard and the API show all nine.
#The nine checks
| Check | The question it answers |
|---|---|
destination | Can anyone read your paid link off this page? |
consistent | Does this page show a robot something different? |
noindex | Can this page turn up in a search engine? |
preview | Does sharing this link show your name and photo? |
words | Does the page use words a filter looks for? |
trackers | Is anything on the page reporting your visitors to someone else? |
hops | Can your outgoing links be copied straight off the page? |
transport | Is the connection missing basic security? |
referrer | Does the page tell the other site where people came from? |
destination is the heavy one. If the paid address is in the HTML, in a script, in the query string of the link itself, or in a redirect a crawler can follow, the page has already told the platform everything.
consistent catches cloaking: a page that serves a crawler an empty shell and a person the real thing. Platforms treat that as deception when they find it, and they do find it.
#Your scans
Scans run from the dashboard are kept against your account, so you can see how a page changed over time rather than scanning it again. The list stays on one screen with a Back button.
#Over the API
POST /api/v1/scan with {"url": "..."} returns the full
result. Twenty an hour per key. See the reference.