outlook-emails-lawvable

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly fetches and reads user emails and attachments from Microsoft Graph (see scripts/outlook_oauth.py and SKILL.md/WORKFLOWS.md) and then parses and feeds that untrusted, user-generated content into downstream workflows/skills (e.g., NDA review, classification), so third-party content can directly influence agent actions.