grilling
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided plans and ideas to build a design tree and is instructed to autonomously 'find facts' from the environment (filesystem and tools) to validate these inputs.
- Ingestion points: User-supplied text provided as a plan, decision, or idea (SKILL.md).
- Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are provided.
- Capability inventory: The skill authorizes the agent to use tools and access the filesystem to gather facts (SKILL.md).
- Sanitization: No sanitization or validation of the user's input is required before the agent performs environment-based fact-finding.
Audit Metadata