write-change-plan
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The instructions do not attempt to override the agent's safety protocols or core instructions.
- [DATA_EXFILTRATION]: No network-related commands or patterns indicative of data exfiltration were found.
- [COMMAND_EXECUTION]: The skill mentions a
suspec checkutility for validation, which is a standard part of a development workflow and does not pose a security threat in this context.
Audit Metadata