personalization-at-scale

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly directs the agent to "research each prospect across multiple sources" using public, user-generated content such as LinkedIn activity, company news, blog posts, and event pages (see "Research Sources" and "Response Approach"), meaning the agent will ingest and act on untrusted third-party content that could contain injected instructions.