gtm-meeting-prep

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime research steps (“Research [COMPANY NAME] and [PERSON NAMES]” and “Research the industry and competitive context”) explicitly pull in external web/third-party information (e.g., news, press, job postings, LinkedIn activity, BuiltWith, competitor context) and then feed the resulting free-form briefs into the agent LLM context, creating an indirect prompt-injection surface from outsider-authored text.