ai-customer-research

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

• Third-party content exposure detected (high risk: 0.85). Skill runtime ingests user-supplied free text via materialText (and potentially scraped/loaded content from user-supplied materialUrl), which is outsider-authored content the operating user did not author; this text is then composed into the LLM context by the skill’s contextComposer flow.