Extra security with restricted reference domains

Security and reliability are central to vragen.ai. So we are introducing 'Restrict reference domains'. This setting lets you decide which external domains are allowed to supply links and images in the answers vragen.ai generates. That keeps unwanted or unsafe references out of the AI output.

What the feature does

Under Settings you'll find the option Restrict reference domains (beta). Once you turn it on:

• vragen.ai only accepts links and images from domains you have explicitly listed;

• all other references are automatically removed or marked as safe.

You enter the allowed domains as a list, one domain per line.

Why this matters

This feature helps raise the quality and safety of answers. By limiting which external sources can end up in generated output, you reduce the chance of incorrect, unwanted or potentially risky references.

Background: an improvement following the ethical hacker assessment

The introduction of Restrict reference domains follows directly from an independent ethical hacker assessment carried out under the Cyber Security Pen Test certification of the Dutch CCV. Read more about it in the article "How vragen.ai holds up under pressure from an ethical hacker".