Check guide ยท Behaviour
Optional human check
A small drill that adds interaction evidence. Nothing real is asked.
What this check is
The drill at the top of the tool page is a voluntary form exercise: an invented call sign, a phrase to type, a few choices, a confirm step, and a table of small values to match. It exists to generate the kind of evidence passive reading cannot: how a form gets filled in. Typing rhythm, pasting versus typing, corrections, pauses, and the order things happen in.
What Tells reads
- Whether and when the form was submitted, and the confirm answer
- Interaction timing and counts only. The answers themselves are never read, sent or stored, which is why the form asks for invented ones
How to read your result
- Pass: The drill was completed: its evidence joins the behaviour reading.
- Not available: The drill was not taken. That is a neutral result; the verdict is computed without it.
Why it matters
Interactive challenges exist because passive signals sometimes cannot separate a careful script from a person, and asking the visitor to do something small and observable can. This one is deliberately gentle (no identity, no real data, nothing stored) and deliberately optional: it demonstrates the category without the hostility of a hard gate. What it measures is the manner of the work, never the content of the answers.
Where this matters most
Across the platforms Blowhorn works with, rated in the platform guide (H = matters a lot, M = contributes, L = no public evidence; tags mark vendor-stated, reported, or inferred evidence):
- LinkedIn (H): checkpoint and identity verification is the standard consequence of a restriction [Vendor]
- X (H): Arkose challenges gate account unlocks and sensitive actions [Reported]
- Reddit (M): flagged accounts face human-verification challenges [Reported]
- GitHub (M): CAPTCHA and verification appear at signup and on flagged logins [Reported]
Everywhere else in the set this one rates Low.
In your report
Find it in the checks table on the tool page, with the detail in
the the drill panel section. The same findings appear in the
report JSON under verdict.reasons, and the agent guide explains
how to read them programmatically.