brand-copywriter
Pass
Audited by Gen Agent Trust Hub on Mar 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is focused on text generation and adheres to its stated purpose without any malicious behavior detected.
- [PROMPT_INJECTION]: The instructions establish a rigorous process for copy generation but do not attempt to override safety protocols, bypass filters, or extract system-level prompts.
- [DATA_EXFILTRATION]: File access is restricted to local documentation and an optional project context file. No network operations or credential-accessing patterns were found.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any external network requests, downloads, or remote dependency installations.
- [COMMAND_EXECUTION]: No shell commands, subprocess spawning, or dynamic code execution patterns are present in the skill's logic.
Audit Metadata