executing-plans
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a framework for step-by-step task execution that emphasizes human-in-the-loop review and critical analysis of input plans.
- [DATA_EXPOSURE]: No sensitive file access, credential harvesting, or network operations were detected.
- [PROMPT_INJECTION]: The instructions do not contain patterns aimed at bypassing safety filters or overriding core agent behavior. It explicitly commands the agent to stop if instructions are unclear.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external plan files, which constitutes a data ingestion surface. However, it includes defensive instructions such as 'Review critically
- identify any questions or concerns', 'Raise them with your human partner before starting', and 'Stop when blocked, don't guess', which mitigate the risk of obeying malicious instructions embedded in the plans.
Audit Metadata