ai-sales-enablement

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime context composer ingests user-supplied free text fields (materialText, and potentially extracted text from materialUrl/materialFile) into the LLM; if those inputs contain outsider-authored content (e.g., pasted customer communications or scraped/linked public pages), that free text becomes LLM context, creating indirect prompt-injection exposure.