keyword-research

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

• Third-party content exposure detected (medium risk: 0.65). The required workflow ingests user-uploaded CSV keyword/query text (and possibly competitor Spy/Search Term Rank exports) and then analyzes it, so if the CSV contains outsider-authored keyword data (e.g., competitor app search terms or public export content not authored by the operating user), that free-form text is read by the agent at runtime and fed into the LLM context for analysis.