Policy decides whether a knowledge source matters.
Documentation may be normal for a senior engineering exercise and unacceptable for a closed-book screen. Running on the slow loop, the detector records the activity from browser context so your team applies the correct policy. It does not read the page.
InterviewWatch does not block sites or capture what was searched. It notes that a knowledge source was active during the round, with timing, so a reviewer can decide whether it fits the interview's rules.
Search engines
Detects general search activity visible in browser titles during the interview.
Developer resources
Flags common technical references and coding-practice sites.
Correlated timing
Most useful when a visit lines up with a question, an answer pause, a clipboard event, or an AI signal.
Open-book interviews
Use the signal as context. Review whether the candidate used allowed resources in a way that matches the interview policy.
Closed-book interviews
Knowledge-source activity during a closed-book technical round can become a higher-risk integrity finding.
Knowledge-source questions, answered.
Does this block candidates from using the web?
Which knowledge sources does it recognise?
Is it different for open-book and closed-book interviews?
Does it capture what the candidate searched?
Match the evidence to your interview policy.
Give reviewers open-book or closed-book context on external knowledge use, without a web filter.
Try nowRead the coding-test guide