content-strategy

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

• Third-party content exposure detected (high risk: 1.00). The skill's "Content Ideation Sources" workflow explicitly instructs using web search and forum research to fetch and extract public, user-generated content (e.g., "Reddit," "Quora," "Hacker News," "Product Hunt") and to mine posts/comments for FAQs and pain points that will directly inform content decisions, which exposes the agent to untrusted third-party content that can alter its actions.