NYC

youtube-video-analyst

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [Unverifiable Dependencies & Remote Code Execution] (LOW): The skill requires the installation of the youtube-transcript-api library. While this is an external dependency from an untrusted author (Ship Shit Dev), it is a common utility for the skill's primary purpose. Severity is downgraded from MEDIUM because it is essential to the core functionality.\n- [Indirect Prompt Injection] (LOW): The skill processes external YouTube transcripts, creating an attack surface where malicious instructions could be embedded in video content to hijack the agent's behavior.\n
  • Ingestion points: Transcripts retrieved via the fetch_transcript.py script or manual pasting into the prompt context.\n
  • Boundary markers: None. The skill does not use protective delimiters or specific instructions to ignore embedded commands in the transcript data.\n
  • Capability inventory: The skill primarily performs text analysis and template generation; the included Python script is limited to network requests for transcripts and basic file writing of the results.\n
  • Sanitization: No validation or sanitization is performed on the incoming transcript text.\n- [Data Exposure & Exfiltration] (LOW): The fetch_transcript.py script performs network requests to non-whitelisted YouTube domains and includes functionality to write data to local files specified by the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:48 PM