email-newsletter
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill does not access sensitive data or environment variables, and it has no mechanism for sending data to external servers. It strictly processes user-provided inputs to generate newsletter content.\n- [COMMAND_EXECUTION]: The skill uses basic shell commands to create directories and save output files locally. This is a functional requirement for its task and does not pose a security risk.\n- [EXTERNAL_DOWNLOADS]: The skill documentation includes installation instructions that reference vendor packages and well-known AI platform import URLs. There are no automated downloads or remote code executions during the skill's operation.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined topics and briefs through a controlled, multi-step drafting process. This structure helps prevent instructions within the user data from being interpreted as commands by the agent.
Audit Metadata