revai

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

• Third-party content exposure detected (high risk: 0.80). The skill's workflow (see "Popular actions" in SKILL.md — notably "Submit Transcription Job" which accepts a publicly accessible media URL and "Get Transcript" which returns the transcribed text) clearly ingests untrusted third-party content (public media/transcripts) that the agent is expected to read and could influence subsequent actions.