executing-plans
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from external implementation plans which could contain malicious instructions aimed at influencing agent behavior.
- Ingestion points: The skill reads a plan file from the local workspace during Step 1.
- Boundary markers: The instructions explicitly mandate the agent to 'Review critically' and 'Raise [concerns] with your human partner before starting'.
- Capability inventory: The skill facilitates the execution of arbitrary tasks defined in the plan, which may include file system operations or command execution depending on the plan's contents.
- Sanitization: The skill relies on the agent's internal reasoning and user confirmation rather than programmatic data sanitization.
Audit Metadata