AI prompt injection — hidden directives in white-on-white / HTML comment targeting Copilot/Gemini (EchoLeak CVE-2025-32711)

What this tier means

High-confidence threat indicator — phishing, impersonation, BEC, or scam pattern. Strong contributor to the trash decision.

False-positive guard

Every signal in Gorganizer feeds a multi-module score — never a sole verdict. This is a threat-tier signal — it adds a strong contribution to the trash score. The full pipeline still requires convergence across multiple modules + a margin over the safety floor before deletion happens, and Gmail's trash (30-day recovery) is always used — never permanent delete.

About the scoring engine

Gorganizer's scoring engine emits over 1,800 signals across six modules — headers, sender, subject, body, attachments, and structural metadata. Every email is scored by every module independently; the final verdict requires multiple modules to agree and the trash score to beat the safety floor by a margin.

Sacred safety guards — never delete starred emails, replies, calendar invites, receipts/invoices, or attachments — apply unconditionally regardless of any signal.