Spam dies before it reaches your inbox.

Multi-layer detection

Every email passes through multiple independent detection layers. A spam message has to evade all of them to reach your inbox:

• Bayesian classifier — Machine learning trained on millions of messages, continuously improved by user feedback
• Authentication checks — SPF, DKIM, DMARC verification flags forged senders
• URL reputation — Links in the email are checked against known phishing and malware databases
• Content analysis — Zero-font tricks, hidden text, phishing layouts, and other evasion techniques are detected
• Crowd-sourced reputation — Sender scores aggregated across all Cleanbox users

Per-alias thresholds and quarantine

Not every alias needs the same sensitivity. Your public-facing contact alias might need aggressive filtering (threshold 3), while your personal alias can be relaxed (threshold 8).

Enable the quarantine zone on any alias. Messages scoring between your quarantine and reject thresholds are held for review instead of being silently delivered or rejected. Accept false positives, reject actual spam, and train the filter with every decision.

• Independent thresholds per alias
• Quarantine for borderline messages
• Accept/reject with one click, train the filter automatically