Phishing email extractor and risk scorer

The brief

Employees forward suspicious emails to the security team, which reads each one. Build a tool that extracts sender, links, attachments and persuasion tactics (urgency, authority, rewards) into validated JSON, and scores how risky each email is. Good looks like: a schema, accuracy on labelled emails, and a risk score that ranks real phishing above genuine mail.

What you will build

1. Define the email analysis schema
2. Extract fields and tactics with structured outputs
3. Validate links and sender domains
4. Score risk and check the ranking on labelled emails
5. Build a forward-and-check page

Free, open data to build it with

• Phishing email dataset ↗ Kaggle

Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.