iap-agent-platform-troubleshooting
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFE
Full Analysis
- [Strict Read-Only Enforcement]: The skill's operating contract explicitly restricts the agent to read-only operations, such as 'describe', 'list', and 'logging read'. It mandates that all fixes must be printed for the user to review and execute, preventing unauthorized resource modifications.
- [Usage Attribution and Telemetry]: The skill includes instructions to include specific attribution strings in CLI commands and HTTP headers. This allows the vendor to track usage of the troubleshooting skill within their own cloud platform for metrics and improvement purposes.
- [Secure Documentation Links]: All external references provided in the skill point to official Google Cloud documentation domains, ensuring that the agent and user are directed to verified, trusted resources.
- [Principle of Least Privilege]: The skill guides the user to verify necessary read-only roles before proceeding and instructs the agent to stop and report if it encounters a PERMISSION_DENIED error rather than attempting to bypass security controls.
Audit Metadata