Uncertain (68% match) because the evidence is mixed and external checks stayed inconclusive, so the system kept an uncertainty buffer instead of a harder automated verdict.
Uncertain
Uncertain
68%
External checks stayed inconclusive, so the model avoided a harder verdict.
Heuristic: The evidence is materially mixed or partially corroborated, so the model is avoiding a harder verdict.
Why: The evidence is mixed and external checks stayed inconclusive, so the system kept an uncertainty buffer instead of a harder automated verdict.
C
Caution
62%
Signals are mixed or weak, so this domain should be treated carefully.
Heuristic: The signal mix leans cautious and the site should be treated carefully until more evidence arrives.
Why: Review carefully: low trust score, elevated spam indicators. Showing the stored external intelligence snapshot. New search and review intelligence only runs after a visitor uses Refresh data manually.
T
Trusted
59%
Strong trust, stability, and low-risk signals were detected.
Heuristic: The site shows a reasonably trustworthy and stable profile.
Why: Low spam indicators plus strong trust, quality, registry, and stability signals were detected. Showing the stored external intelligence snapshot. New search and review intelligence only runs after a visitor uses Refresh data manually.
Strongest tag heuristic
Uncertain · 68%
The highest raw tag match from the heuristic engine before visibility thresholds and manual overrides.
Trust score
30/100
Confidence derived from authority, crawl quality, stability, and risk signals.
Authority score
14/100
A higher authority score usually means broader reputation and backlink confidence.
Spam risk
53/100
Lower is better. This blends spam indicators with false-positive protections for legitimate sites.
Quality score
67/100
Based on crawl quality, content completeness, and technical evidence.
Safety score
100/100
Higher is safer. Direct fraud signals can heavily cap this even when SEO signals look strong.
Fraud score
0/100
Lower is better. This reflects phishing, drainer, fake-support, fake-shop, and malware signals.
Infrastructure risk
0/100
Lower is better. This reflects the IP, registrar, and nameserver neighbourhood seen by the crawler.
Evidence confidence
53/100
Higher means the crawler had enough pages and registry evidence to make a stronger call.
External evidence
Low
Quality of the structured search and review evidence cluster.
External decision
Inconclusive
How third-party evidence compares with the current candidate tag.
NSFW score
0/100
Context-aware adult-content detection with medical / educational false-positive reduction.
Registry stability
100/100
Based on age, expiry runway, and the amount of ownership / registrar churn.
Search & review intelligence
Showing the stored external intelligence snapshot. New search and review intelligence only runs after a visitor uses Refresh data manually.
External intelligence is manual-only. Use Refresh data to trigger a fresh search and review snapshot for this domain.
Decision state
Inconclusive
Public search did not yield usable third-party matches, so the system fell back to first-party inspection only.
Evidence quality
Low
Source trust, entity match, freshness, and diversity.
Support / contradiction
0/0
How much external evidence supports the current tag versus contradicts it.
External confidence
24/100
1 fetched documents · 0 facts
Discovery layer
Queries12
Search intelligence discovers candidate evidence sources without blocking the page.
Exact matches0/100
How much of the discovery layer appears to reference this exact domain rather than a similar name.
Results found1
Raw search results considered before verification and extraction.
Entity coverage100/100
How strongly the discovered pages seem to refer to this exact domain.
Verification layer
Documents fetched1
Background verification fetches actual pages instead of relying only on snippets.
Source diversity0/100
Risk tags only promote when source trust or diversity thresholds are met.
Fetch success100/100
Shows how much of the shortlisted evidence set the server could actually fetch and verify.
Freshness18/100
Uses per-source decay instead of a flat one-size-fits-all expiry.
- External evidence remained inconclusive because diversity, exact matching, or confidence was still too weak.