reply-handler
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely composed of text-based instructions and logic for handling communication. No security risks were identified.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data in the form of prospect replies. While a prospect could attempt to embed instructions within their reply to influence the AI, the skill possesses no dangerous capabilities (such as shell access, network requests, or file system modifications) that could be exploited. The surface area for harm is limited to the content of the generated response drafts.
Audit Metadata