Ucs Policy Governor

• Outcome: Review
• Latest audit: Jun 16, 2026 at 7:00 PM UTC
• Version:
  0.1.0

Overview

This skill is a legitimate Huawei Cloud UCS governance guide, but it includes broad cloud and cluster-changing operations that are not consistently scoped or warning-gated.

Install only if you intend to let the agent assist with Huawei Cloud UCS administration. Use a dedicated least-privilege IAM user, prefer read-only permissions for audit tasks, avoid production targets for verification examples, confirm IDs before delete or disable operations, and treat generated kubeconfig files as sensitive credentials that should be cleaned up.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (7)

Context-Inappropriate Capability

High

• Confidence: 97% confidence
• Finding: The file is presented as a compliance audit task, but this workflow escalates from read-only governance review into credential generation (`CreateClusterKubeconfig`), direct cluster mutation via `kubectl apply`, and policy disable/re-enable actions. That scope expansion is dangerous because an audit-oriented skill can unexpectedly grant access to cluster credentials and modify live infrastructure, increasing the chance of unauthorized changes or misuse.

Context-Inappropriate Capability

Medium

• Confidence: 90% confidence
• Finding: This scenario adds `RegisterCluster` to a document whose stated purpose is compliance auditing. Cluster registration is an onboarding/change-management action, not an audit action, so including it in this context broadens the agent's operational scope and could let an audit-triggered workflow make infrastructure changes that the user did not intend.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The document includes a direct `DeletePolicyInstance` example as part of a workflow without any warning that the action is destructive or guidance to verify the target instance first. In a policy-governance skill, deleting the wrong policy instance can disable compliance controls or remove governance coverage from clusters or fleet groups, making accidental misuse materially risky.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The document instructs operators to grant powerful write permissions such as create, update, delete, enable, and disable across all resources, but it does not clearly warn that these actions can change enforcement state or remove governance controls. In the context of a policy-governance skill, this can lead users to overprovision access and make high-impact changes without understanding the operational and compliance consequences.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The remediation section documents potentially disruptive commands such as generating cluster access credentials, applying manifests to a cluster, disabling policy, and re-enabling policy without any warning about production impact, rollback, validation, or approval. In an agent skill, omission of these cautions increases the risk that operators execute impactful changes too casually or in the wrong environment.

Missing User Warnings

Low

• Confidence: 92% confidence
• Finding: The document instructs users to export long-lived Huawei Cloud access keys as shell environment variables without warning about common exposure paths such as shell history, inherited child-process environments, crash dumps, and accidental logging. In a cloud administration skill, these credentials grant control-plane access, so even a low-friction documentation pattern can increase the chance of credential leakage and subsequent unauthorized access.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The document includes create, update, enable, disable, and delete commands that directly change UCS policy state and cluster enforcement behavior, but it does not place a prominent upfront warning before the workflow that these steps can affect running clusters, cause policy enforcement changes, or remove policy instances irreversibly. In a verification guide, operators may treat all listed commands as routine validation and execute them in production, increasing the risk of unintended denial, drift, or service impact.

VirusTotal

VirusTotal findings are pending for this skill version.