domain-grill
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted engineering artifacts and repository files such as
CONTEXT.mdand source code to drive an automated interview process. - Ingestion points: Technical design documents provided by the user,
CONTEXT.md(orCONTEXT-MAP.md), existing Architecture Decision Records (ADRs) indocs/adr/, and project source code. - Boundary markers: The instructions lack specific delimiters or instructions for the agent to ignore potential commands embedded within the analyzed technical documentation or source code.
- Capability inventory: The skill is authorized to read project files and write new Architecture Decision Records (ADRs) to the
docs/adr/directory based on the interview outcomes. - Sanitization: The instructions do not specify any sanitization, filtering, or escaping of the technical content before it is processed by the agent.
Audit Metadata