rbac-design
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: No malicious prompt injection patterns detected. The skill contains defensive instructions instructing the agent to ignore directives within input artifacts that attempt to bypass safety boundaries.
- [EXTERNAL_DOWNLOADS]: No external code or package downloads identified. References to NIST and industry standards are provided as plain-text documentation links.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted authorization artifacts (IaC, role names). It implements defensive logic within its instructions to treat these inputs as untrusted and ignore embedded directives.
- [NO_CODE]: The skill consists entirely of markdown instructions and metadata for the AI agent; no scripts or executable binaries are included.
Audit Metadata