ai-video-editing

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

• Third-party content exposure detected (medium risk: 0.65). Runtime path: the operating user-provided materialUrl/materialText/materialFile are ingested into the skill’s context composer; if materialUrl points to public web content or materialText/uploaded files contain outsider-authored text, that free text can be read and fed into the LLM (indirect prompt injection risk).