grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user input during a 'grilling' interview session and records results into project files such as CONTEXT.md, LINKS.md, and ADRs. This creates an attack surface where malicious instructions provided during the interview could be persisted into the project's documentation.
- Ingestion points: User input provided during design interviews and design tree resolution (SKILL.md).
- Boundary markers: Absent. No specific delimiters or instructions to ignore embedded commands are specified for the output files.
- Capability inventory: File system writes to the
.agents/projects/directory within the repository (SKILL.md). - Sanitization: Absent. The skill does not specify validation or filtering of the captured facts and decisions before writing them to disk.
Audit Metadata