ai-agentic-marketing-workflows

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly instructs agents to scan and ingest untrusted, user-generated content from public social platforms and sites (e.g., "Scan Facebook, Instagram, X/Twitter, Google reviews" in the Sentiment Monitoring Agent and "Monitor trending topics and competitor content" in the Content Pipeline Agent), which the agent must read/classify and which can drive decisions and actions.