Observe
Indicator identity, DNS, HTTP and public scan state.
Check a domain, exact URL or public IP against the confirmed CRDF blocklist and the independent CRDF AI V14 intelligence report from one clear public page.
Reading the public intelligence index.
The confirmed blocklist could not be checked.
This result is independent from the CRDF AI V14 assessment.
Reading a known report is unrestricted. Starting a scan, refreshing a report or sending feedback requires this anti-abuse check.
First confirm whether an indicator is actively listed by CRDF, then inspect the separate CRDF AI V14 assessment and its public context.
Indicator identity, DNS, HTTP and public scan state.
Retained domain and IP history across earlier observations.
WHOIS, graph relations, shared infrastructure and peers.
A V14 verdict with evidence strength, scope and freshness.
No indicator loaded.
No report has been loaded.
Public-safe taxonomy and model output.
Retained domain and IP observations.
Exact URL, host and registrable domain evidence remain distinct.
Grouped explanations without private thresholds or raw intelligence.
When the retained score, memory and report state changed.
Public, non-personal registration intelligence.
Public relations are context, never automatic guilt.
A structured map of retained infrastructure, WHOIS and peer relations.
Swipe horizontally to explore the map. Every relation remains available in the detail list.
Retained decisions for this indicator.
Publicly retained campaign or registrable peers.
Feedback is queued for review. It never changes a production verdict automatically.