grilling
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions define a process for logical stress-testing and decision-tree exploration. No malicious patterns such as credential exfiltration, persistence mechanisms, or obfuscation were detected.
- [INDIRECT_PROMPT_INJECTION]: The instruction to explore the environment (filesystem, tools) to find facts creates a potential surface for indirect prompt injection if the agent encounters malicious instructions within those files. However, this behavior is central to the skill's purpose of context-aware interviewing.
- Ingestion points: Filesystem and tool outputs (SKILL.md).
- Boundary markers: None specified in instructions.
- Capability inventory: Uses available filesystem and tool execution capabilities to gather context.
- Sanitization: Relies on the underlying agent's default safety filters.
Audit Metadata