product-conjoint-analysis

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md Phase I "Two-track collection" and references/review_mining.md explicitly instruct collecting and mining open web sources (e.g., Amazon product pages, public reviews, forums, Sellerly) and using those user-generated texts to triangulate attributes and drive model design and recommendations, so untrusted third-party content is ingested and directly influences agent actions.