google-email

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly syncs and fetches Gmail message content (see SKILL.md workflows and scripts like scripts/sync-emails, scripts/email, scripts/emails and the triage flow) and the AI triage examines email bodies/HTML to extract links and make decisions, so untrusted third-party email content can be ingested and materially influence actions.