Trust Assessment
k8s-capi received a trust score of 67/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 3 findings: 0 critical, 2 high, 1 medium, and 0 low severity. Key findings include Arbitrary Kubernetes Manifest Application, Exposure of Cluster Kubeconfig, Unrestricted Scaling of Machine Deployments.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 63/100, indicating areas for improvement.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Arbitrary Kubernetes Manifest Application The skill exposes the `kubectl_apply` tool, which allows the application of arbitrary Kubernetes manifests. A malicious prompt could instruct the LLM to generate and apply manifests for privileged pods, RBAC roles, or other resources that could compromise the cluster, exfiltrate data, or cause a denial of service. The examples provided demonstrate creating `Cluster` and `MachineDeployment` resources, which are high-privilege operations in a Kubernetes environment. Implement strict validation and sanitization of manifests before applying them. Consider a human approval step for `kubectl_apply` operations, especially for sensitive resource types or namespaces. Restrict the LLM's ability to generate arbitrary manifests, perhaps by providing templates or whitelisting allowed resource types and fields. | LLM | SKILL.md:109 | |
| HIGH | Exposure of Cluster Kubeconfig The `capi_cluster_kubeconfig_tool` explicitly returns a kubeconfig, which grants direct administrative access to a Kubernetes cluster. If an attacker can prompt the LLM to execute this tool and then reveal the output, it constitutes a direct credential harvesting and data exfiltration risk, allowing unauthorized access to the cluster. Implement strong access controls and redaction policies for the output of this tool. Require explicit user confirmation or multi-factor authentication before revealing sensitive credentials like kubeconfigs. Ensure the LLM's output is not logged or stored insecurely when this tool is used. | LLM | SKILL.md:37 | |
| MEDIUM | Unrestricted Scaling of Machine Deployments The `capi_machinedeployment_scale_tool` allows modification of the `replicas` count for machine deployments. An attacker could prompt the LLM to scale down critical worker nodes to zero, causing a denial of service, or scale up excessively, leading to significant cloud cost overruns. Implement guardrails for scaling operations, such as minimum and maximum replica counts, or require human approval for scaling actions outside predefined safe ranges. Monitor for unusual scaling activities. | LLM | SKILL.md:79 |
Scan History
Embed Code
[](https://skillshield.io/report/50faa391790be994)
Powered by SkillShield