kubectl · kube- MCP: Risk & Policy

WHAT IT DOES

What kubectl does on kube-MCP

AI agents invoke kubectl to trigger actions in kube-MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why kubectl needs a policy

kubectl enables execution of arbitrary commands against a Kubernetes cluster. While the sibling tools (delete_pod, get_pods, etc.) are scoped operations, this generic 'Execute kubectl commands' tool permits running any kubectl subcommand, which can modify state (Write), destroy resources (Destructive via kubectl delete), or execute code in containers (Execute via kubectl exec).

From the tool's definition Tool description states 'Execute kubectl commands' — kubectl is a command-line tool that runs arbitrary operations against Kubernetes clusters, including deployments, scaling, port-forwarding, and code execution via exec.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

kubectl Rate-limited

kubectl stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "kubectl": {
      "limits": [
        {
          "counter": "kubectl_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches kube-MCP, so you can run agents against it safely.

FAQ

Questions about kubectl

What does the kubectl tool do? +

Execute kubectl commands. It is categorised as a Execute tool in the kube-MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on kubectl? +

Register the kube- MCP server in PolicyLayer and add a rule for kubectl: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches kube-MCP. Nothing to install.

What risk level is kubectl? +

kubectl is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit kubectl? +

Yes. Add a rate_limit block to the kubectl rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block kubectl completely? +

Set action: deny in the PolicyLayer policy for kubectl. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides kubectl? +

kubectl is provided by the kube- MCP server (siddjoshi/kube-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.