lead-magnets

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow (e.g., “Survey to Lead Magnet”) ingests user-provided survey free-text responses (pain language) into the LLM context; this free-form text is authored by third-party respondents, i.e., an outsider source via the “survey export or pasted responses” input path.