huawei-cloud-cce-cluster-management

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill adheres to documented security constraints, specifically avoiding the persistent storage of sensitive credentials (AK/SK). Credentials are required to be provided via environment variables at the process level.
  • [EXTERNAL_DOWNLOADS]: The skill specifies dependencies on official Huawei Cloud SDKs (huaweicloudsdkcore, huaweicloudsdkcce) and standard industry libraries (kubernetes, passlib, requests). These are well-known, trusted packages for cloud and container management.
  • [COMMAND_EXECUTION]: The skill uses a dispatcher to execute Python scripts that interact with cloud APIs. All dangerous operations, such as deleting clusters or node pools, require a confirm=true flag, preventing accidental execution through a built-in preview mechanism.
  • [DATA_EXPOSURE]: The skill creates temporary certificate files in the system's temporary directory to facilitate Kubernetes API authentication. These files are wrapped in finally blocks and deleted immediately after the API call is completed, minimizing the window of exposure.
  • [PROMPT_INJECTION]: Analysis of the instructions and metadata revealed no attempts to override agent behavior or bypass safety filters. Instructional language is focused on task execution and safety confirmations.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 09:10 AM
Security Audit — agent-trust-hub — huawei-cloud-cce-cluster-management