ThreatLens AI
Model A · URL classifier

Malicious URL Scanner

Classifies a URL as benign, phishing, malware or defacement from 30 lexical features and 512 character n-grams — no page is fetched, so scanning is instant and safe.

Try an example
--awaiting input

Paste a URL or pick an example. The scanner extracts 30 lexical features + 512 hashed character n-grams and runs them through the URL neural network.