grill
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access mechanisms were detected in the skill instructions or scripts.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests user input to build a "design tree" and determine subsequent actions for sub-agents. However, the interaction follows a standard user-in-the-loop pattern.
- Ingestion points: User answers to interview questions provided via the AskUserQuestion tool (SKILL.md).
- Boundary markers: No explicit delimiters or warnings for the agent to ignore instructions embedded in user answers are defined.
- Capability inventory: The skill utilizes the AskUserQuestion tool and has the capability to dispatch sub-agents to explore the filesystem and environment (SKILL.md).
- Sanitization: No sanitization or validation of user-provided answers is specified before they influence the agent's logic.
Audit Metadata