youtube-transcript
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external YouTube transcripts which could contain embedded instructions intended to override agent behavior.\n
- Ingestion points: Transcripts are fetched in
scripts/get_transcript.pyand passed to the agent context.\n - Boundary markers: There are no explicit markers or 'ignore' instructions wrapping the external content returned by the script.\n
- Capability inventory: The agent has instructions to write transcripts to files on the local system.\n
- Sanitization: The transcript text is not sanitized before being returned to the agent context.\n- [COMMAND_EXECUTION]: The skill runs a Python script via
uv runto perform the transcript extraction.
Audit Metadata