req-elicitation
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill consists entirely of instructional text and methodology guidelines. It does not contain scripts, shell commands, or executable logic.
- [PROMPT_INJECTION]: The skill includes defensive instructions to detect and flag potential injection attempts in the user input it processes (e.g., 'backdoor requirements'). It does not contain any instructions that attempt to bypass agent safety filters.
- [INDIRECT_PROMPT_INJECTION]: While the skill processes untrusted stakeholder data, it implements several mitigation strategies: requirement IDs for traceability, EARS formal syntax constraints, and a dedicated 'Ambiguity Resolution' step that serves as a semantic filter against malicious or vague inputs.
- [SAFE]: The skill follows professional engineering practices (IREB, EARS, ISO 25010) and incorporates 'Misuse Case' analysis to ensure security is addressed by design in the elicited requirements.
Audit Metadata