cms-roles-permissions
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely advisory and contains no executable code or scripts. It is designed to provide guidance to developers on Contentstack access control mechanisms.
- [CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or tokens are present. The skill contains multiple explicit instructions to avoid exposing tokens, to use environment variables for credentials, and to refer users to secure secret storage.
- [EXTERNAL_DOWNLOADS]: The skill does not perform any network operations or download external content. It relies on internal advisory instructions.
- [PROMPT_INJECTION]: No malicious instruction-override patterns or jailbreak attempts were detected. The instructions reinforce the agent's safety boundaries, particularly regarding destructive actions and data exposure.
- [DATA_EXFILTRATION]: The skill does not have network access tools (allowed-tools are restricted to Read, Grep, and Glob). It explicitly forbids printing or requesting secrets.
- [COMMAND_EXECUTION]: There are no shell commands or subprocess execution patterns. The skill is limited to file reading and searching within the local environment.
Audit Metadata