yt-comment
Pass
Audited by Gen Agent Trust Hub on Nov 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied YouTube comments, which are untrusted external inputs. This presents an attack surface where malicious instructions within a comment could attempt to influence the agent's output.
- Ingestion points: The skill triages and drafts replies based on 'a pasted comment thread' or 'what the user pastes' in SKILL.md.
- Boundary markers: The instructions do not define delimiters or specific markers to isolate the untrusted comment content from the agent's system instructions.
- Capability inventory: The skill is restricted to drafting text and explicitly requires human review before any content is published ('This skill writes and you publish').
- Sanitization: The skill does not perform any sanitization, filtering, or validation of the input comment text.
Audit Metadata