brd-interviewer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a well-structured business consulting tool that does not contain any malicious code, obfuscation, or unauthorized commands. It follows a strict persona and interview framework to generate documentation.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied business descriptions and data, creating a potential surface for indirect prompt injection. However, it incorporates structural defenses, including fixed interview phases and explicit instructions to ignore technical implementation details (HLD) provided by the user.
- Ingestion points: User-provided business ideas in SKILL.md (Phase 0).
- Boundary markers: Explicit 'Boundary Guarding' instructions in SKILL.md to reject technical implementation details.
- Capability inventory: Generation of BRD documents using provided templates.
- Sanitization: The agent is instructed to summarize and confirm user inputs before proceeding, which serves as a manual validation step.
Audit Metadata