AGENT LAB: SKILLS

youtube-transcribe-skill

Warn

Audited by Snyk on Feb 15, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (high risk: 1.00). The skill explicitly downloads or scrapes transcripts from public YouTube video pages—via yt-dlp (--write-sub / --write-auto-sub) or browser automation (mcp__chrome__evaluate_script reading the page DOM)—thereby ingesting untrusted, user-generated third-party content that could carry indirect prompt injections.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 15, 2026, 09:14 PM