bridgebound-in-market-20

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly references and relies on public, user-generated sources—e.g., competitor websites, G2 reviews, and social/company pages ("Competitor's Customers (Through Competitor's Website)", "left a Negative Review", "How do I find churned customers from my competitor on G2?") which the agent would need to read/interpret, exposing it to untrusted third-party content.