comment-responder
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes external user content.
- Ingestion points: Processes "new comments" from external social media platforms.
- Boundary markers: Lacks explicit delimiters for the processed comment text within the rule processing logic.
- Capability inventory: Performs automated public replies, automated direct messaging (DMing), and logging data to a lead list.
- Sanitization: Includes a safety rule to "defer to a human" if a comment does not match the predefined keyword rules, which serves as a validation step.
Audit Metadata