review-intelligence-digest

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

• Third-party content exposure detected (high risk: 0.90). SKILL.md Phase 1 explicitly instructs running a review-scraper against public, user-generated review pages on G2, Capterra, and Trustpilot and then feeds those scraped reviews into LLM-driven analysis to generate messaging and recommended actions, so untrusted third-party review text could materially influence behavior.