alibabacloud-video-forge

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly accepts and processes arbitrary public URLs and user-provided media (see --url parameters in scripts like scripts/mps_audit.py, scripts/mps_transcode.py and the params.md / SKILL.md workflow examples), so it ingests untrusted third-party content (UGC or public web videos) that the agent reads/interprets (media probing, adaptive transcoding, content-moderation) and whose analysis results can materially change downstream decisions (pipeline selection, publish/block actions).