requirement-quality-review
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external, untrusted requirements and change briefs, creating a surface for indirect prompt injection.\n
- Ingestion points: The skill accepts user-provided objectives, scope, material, and evidence in prompts/requirement-quality-review.md.\n
- Boundary markers: The skill instructions enforce a structured audit (Known facts, Missing info, Conflicting info) and use RQ-## finding IDs to separate analysis from the source text.\n
- Capability inventory: The skill is primarily instructional and does not demonstrate dangerous capabilities such as file-system writes, network access, or shell command execution in the provided files.\n
- Sanitization: Employs logical constraints that prohibit the agent from making final business decisions or assigning scores based on provided input.\n- [SAFE]: The skill uses defensive prompt engineering and explicitly limits the AI's authority by requiring human sign-off and prohibiting arbitrary scoring. Configuration in evals/local-rules.json further restricts permissions by disabling escalations.
Audit Metadata