youtube-content
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of 'youtube-transcript-api' from PyPI. This is a legitimate and widely-used open-source library for accessing YouTube transcripts.
- [COMMAND_EXECUTION]: Transcript fetching is performed via a local Python script ('scripts/fetch_transcript.py'). The script securely handles user input by using the 'argparse' library and applying a strict regular expression to extract the 11-character YouTube video ID, preventing shell injection.
- [PROMPT_INJECTION]: The skill processes untrusted transcript data from YouTube, which creates an indirect prompt injection attack surface.
- Ingestion points: Transcripts are fetched and passed to the agent's context by 'scripts/fetch_transcript.py'.
- Boundary markers: The instructions do not define clear delimiters (like XML tags or markers) to distinguish the untrusted transcript content from the agent's core instructions.
- Capability inventory: The skill's functionality is limited to text transformation (summarizing, formatting). It does not have access to sensitive file paths, environment variables, or arbitrary system commands that could be triggered by malicious transcript content.
- Sanitization: The transcript text is not sanitized or filtered for instructions before being processed by the agent.
Audit Metadata