How MxGuard Personal protects your mail

MxGuard Personal uses the same two-layer, hybrid-AI engine as the business plans — nothing is cut down.

1. Machine-learning classifier. A gradient-boosted ML model scores the easy 95% of mail in milliseconds.
2. Claude LLM evaluation. The borderline messages are handed to Claude, Anthropic's large language model, which reads them the way a person would before a verdict is reached.

Every verdict comes with a plain-English reason, so you can always see exactly why a message was tagged, quarantined or rejected.

What it stops: phishing and credential-harvest pages; impersonation and invoice-redirection scams; malware and ransomware (ClamAV scans every attachment); classic bulk spam; lookalike/homoglyph domains; and spoofing, via SPF/DKIM/DMARC enforcement. The carefully written, well-targeted attacks that slip past old rule-based filters are exactly where MxGuard's LLM layer is strongest.