analytics-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes user-generated content from an external source.
- Ingestion points: The skill retrieves the top 50 comments for episodes using the mcp__youtube-data__* tool.
- Boundary markers: The instructions do not include specific delimiters or 'ignore embedded instructions' warnings for the comment data, which could allow adversarial text in a comment to influence the model.
- Capability inventory: The agent possesses Read and Write permissions for the workspace/ directory, which could be targeted by successful injection.
- Sanitization: No explicit sanitization, filtering, or escaping of external comment content is documented before the data is processed.
Audit Metadata