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 check utility for validation, which is a standard part of a development workflow and does not pose a security threat in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 06:02 PM
Security Audit — agent-trust-hub — write-change-plan