grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions and configuration files do not contain any malicious patterns. The functionality is restricted to an interactive 'grilling' session to improve designs and the creation of standard documentation like ADRs and glossaries.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data (user-provided plans and designs) to generate documentation. However, this interaction is the primary intended purpose of the skill and occurs within standard agent boundaries without specialized file-system or network capabilities that would pose a significant risk.
- Ingestion points: User input provided during the interview process (SKILL.md).
- Boundary markers: None explicitly defined in the provided snippets.
- Capability inventory: The skill documentation mentions creating files (ADRs and glossaries).
- Sanitization: Standard LLM guardrails apply to the output generation.
Audit Metadata