grilling
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to automatically gather information from the external environment to resolve questions.
- Ingestion points: Data retrieved by sub-agents from the filesystem and other tools is processed to update the 'design tree' and 'frontier' (SKILL.md).
- Boundary markers: The instructions do not define clear delimiters or warnings for the agent when processing data returned from the environment.
- Capability inventory: The skill grants the agent the ability to trigger sub-agents with filesystem and tool access (SKILL.md).
- Sanitization: There are no instructions for sanitizing or validating the content retrieved from external files or tool outputs before it is incorporated into the agent's context.
Audit Metadata