building-phishing-reporting-button-workflow

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill parses user-reported .eml emails (user-generated/untrusted content) and explicitly submits extracted URLs to third-party reputation services (e.g., VirusTotal via scripts/agent.py and SKILL.md's "Submit URLs to VirusTotal, URLScan.io/PhishTank"), and those external results and email content are used to auto-classify and trigger remediation actions, so untrusted third-party content can influence agent behavior.