Public sample · fictional data
This shows how AuditLume organises automatable signals to guide improvements. The domain is not a customer and has not been scanned.
Indicative score
Needs improvement
The score subtracts the assigned impact of each detected signal. It helps order work; it does not measure compliance or total risk.
01
Codes and values come from reproducible rules. The wording explains what each rule observed without extending the evidence.
The fictional HTML contains an analytics pattern and no recognised consent-management signal.
Recommendation: block non-essential analytics until a valid choice has been recorded.
The fictional html element has no lang attribute.
Recommendation: declare the primary language on the html element.
The fictional response has no Content-Security-Policy header.
Recommendation: first test a report-only policy and tune it before enforcement.
02
Review analytics loading first, then declare the language, and finally design and test a CSP. This order comes from impact points, not a human assessment.
03
04
Illustrative text; not AI-generated
A future explanation could connect these three signals and suggest an understandable order, but it could only explain the evidence above. It would not add findings, obligations or guarantees.
05
One URL, deterministic rules, an indicative score and short priorities.
It could extend pages, context and the action plan while preserving evidence and limitations. It is not for sale yet.
06
This sample uses invented data. AuditLume observes one public response and its initial HTML: it does not certify GDPR, WCAG or EAA, provide legal advice, perform a human or professional security audit, or demonstrate absence of risk.
The semantic structure and print styles form the basis of a later PDF template. This page does not generate or download a PDF.