roast-channel
Pass
Audited by Gen Agent Trust Hub on Nov 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes untrusted data from YouTube (titles, descriptions, and About text), which serves as a potential vector for indirect prompt injection.
- Ingestion points: Workflow steps 5 and 6 instruct the agent to extract information from the 'About text, the descriptions and the openings' of YouTube videos.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore or isolate potential commands embedded within the external YouTube metadata.
- Capability inventory: The skill utilizes the 'TubeLab MCP' toolset to perform searches and fetch channel data; the skill instructions do not specify direct shell execution or sensitive file system operations.
- Sanitization: There are no instructions to sanitize, escape, or validate the content retrieved from YouTube before it is interpolated into the agent's context.
Audit Metadata