bilibili-transcript

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

• Third-party content exposure detected (low risk: 0.10). The workflow fetches video transcripts and titles from Bilibili via user-supplied video URLs, ingesting outsider-authored video text and captions into the agent.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The script fetches a Whisper model binary from an arbitrary third-party mirror (hf-mirror.com) at runtime without integrity verification or official publisher provenance, constituting a weak operational dependency.