content-creation

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md workflow clearly accepts and fetches arbitrary external media URLs (e.g., "recoup content transcribe --url {audioUrl}", "--reference-image {referenceImageUrl}", "--image {faceUrl}", "--mux-audio {url}" and related video analyze steps), meaning untrusted third‑party audio/image/video content is ingested and interpreted by the agent and could therefore contain embedded instructions that influence processing or subsequent actions.