Check guide · In the browser

Remote-control tells

Direct evidence that something else is holding the wheel.

What this check is

Where other checks infer, this one looks for the machinery itself: the globals and document keys automation drivers leave behind, the fingerprints of the Chrome DevTools Protocol, input events no hand produced, input arriving while the page is hidden, and the patches stealth tools apply to the automation flags this page reads elsewhere.

What Tells reads

Known automation globals on window and driver keys on document
A CDP serialization probe (a known side channel of the DevTools protocol)
Whether the webdriver getter is native, and whether the property was redefined to hide
Whether everyday native functions (Function.prototype.toString, permissions.query) have been replaced by script
Input event trust: scripted events, events while the document is hidden or unfocused
Impossible window geometry (a zero outer size)

How to read your result

Fail: Driver globals or document keys found, scripted input, input into a hidden page, a zero-size window, or evidence the webdriver flag was hidden by patching. These are the most specific findings in the whole report.
Warning: The CDP probe tripped, everyday functions were replaced by script, or trusted input arrived while the page lacked focus.
Pass: No machinery found and no patch seams showing.

Why it matters

Most bot detection is circumstantial; this check is the smoking gun category. Its findings name a mechanism (a driver, a protocol, a synthetic event) rather than a resemblance, which is why they carry the verdict's heaviest per-point weight after the network flags. It is also an arms race in miniature: every item on the list exists because a tool left it there, and every patch for it is itself on the list.

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): detecting automation software and extensions is the stated enforcement mission [Vendor]
X (H): scripting the website can bring permanent suspension [Vendor]
Reddit (M): headless and automation markers contribute to network blocks [Reported]
GitHub (M): UI automation is detectable in principle; enforcement centres on bulk inauthentic activity [Inference]

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 Runtime section. The same findings appear in the report JSON under verdict.reasons, and the agent guide explains how to read them programmatically.