
Summary
This detection rule is designed to identify and mitigate brand impersonation attempts targeting PayPal. It employs a combination of string matching techniques to recognize variations of the brand name 'PayPal' in email sender display names, as well as in the body of messages and attachments. The rule leverages machine learning (ML) to analyze logos in attachments, ensuring that emails containing images that visually resemble PayPal branding are flagged. The detection further scrutinizes machine learning topics in the email body to filter out legitimate professional or governmental communications. Additionally, the rule cross-checks the sender's email domain against a whitelist of legitimate PayPal domains while also considering high-trust sender domains, focusing on those that fail DMARC authentication to increase the precision of detections. Keywords such as 'invoice', 'transaction', and others related to phishing scams are looked for in the content to deepen the analysis. The severity of this rule is set to medium, reflecting the risk posed by potential phishing attempts related to PayPal impersonation. By utilizing diverse detection methods including content, header, and file analysis, this rule aims to provide a comprehensive approach to identifying fraudulent emails.
Categories
- Identity Management
- Web
- Endpoint
- Cloud
Data Sources
- User Account
- Application Log
- Network Traffic
- Image
- File
Created: 2021-02-19