youtube-content
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from YouTube transcripts, creating a vulnerability surface for indirect prompt injection attacks.
- Ingestion points: The
fetch_transcript.pyscript retrieves transcripts from external YouTube URLs and returns the full text to the agent for processing (summarization, threading, etc.). - Boundary markers: The skill instructions do not mandate the use of delimiters (such as XML tags or explicit block markers) or negative constraints to prevent the agent from following instructions that might be hidden within the transcript text.
- Capability inventory: The skill uses the
terminalto execute its helper script but does not perform dangerous file-system or network operations using the transcript content itself. Its primary capability is text transformation. - Sanitization: The skill performs no filtering, sanitization, or safety validation of the transcript content before it is processed by the agent.
Audit Metadata