content-research-writer

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

• Third-party content exposure detected (high risk: 0.85). The skill's "Conduct Research" workflow explicitly instructs the agent to "search for relevant information" and "find credible sources" (examples cite public sources like McKinsey and Stack Overflow) — showing it ingests open/public third-party content that will be read and used to influence writing decisions.