gtm-meeting-prep

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

• Third-party content exposure detected (high risk: 1.00). The skill's Step 2 Research flow launches subagents to "Research [COMPANY NAME] and [PERSON NAMES]" and explicitly calls for scraping public/untrusted sources (e.g., "Recent LinkedIn posts", job postings, BuiltWith, earnings calls, press, blog") as part of building the meeting brief, meaning runtime-consumed third-party content can directly influence subsequent actions and recommendations.