social-proof-and-testimonials

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

• Third-party content exposure detected (high risk: 0.75). The required workflow is explicitly about curating and transforming “real customer proof” (reviews/testimonials/UGC/case studies) into content, which at runtime would require ingesting outsider-authored review/testimonial text from external customers into the agent’s LLM context (even though the skill claims “injection safety” and “consent first”).