market-research

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s required workflow is to synthesize “sources” (including secondary analysis like “media, blog posts, roundups” and competitor verification via “URL, access date, and what you actually observed”), which at runtime would commonly involve fetching/ingesting outsider-authored free text (e.g., scraped web pages or review threads) into the LLM context.