kernel_push · Kaggle- MCP: Risk & Policy

WHAT IT DOES

What kernel_push does on Kaggle-MCP

AI agents use kernel_push to create or update resources in Kaggle-MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Kaggle-MCP environment.

WHY IT NEEDS A POLICY

Why kernel_push needs a policy

This tool creates or modifies kernel artifacts on the Kaggle platform, which are reversible write operations. While it does not delete data (ruling out Destructive), does not execute arbitrary code in the user's environment (ruling out Execute in the traditional sense), and does not handle financial transactions (ruling out Financial), it does allow creation and modification of published content.

From the tool's definition Tool description states 'Push a new version of a kernel or create a new kernel.' The verbs 'push' and 'create' indicate data creation/modification operations. Kernels on Kaggle are code notebooks/scripts that are published to the platform.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

kernel_push Rate-limited

kernel_push stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "kernel_push": {
      "limits": [
        {
          "counter": "kernel_push_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL →

Instant setup, no code required.

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

FAQ

Questions about kernel_push

What does the kernel_push tool do? +

Push a new version of a kernel or create a new kernel. It is categorised as a Write tool in the Kaggle-MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on kernel_push? +

Register the Kaggle- MCP server in PolicyLayer and add a rule for kernel_push: 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 Kaggle-MCP. Nothing to install.

What risk level is kernel_push? +

kernel_push is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit kernel_push? +

Yes. Add a rate_limit block to the kernel_push 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 kernel_push completely? +

Set action: deny in the PolicyLayer policy for kernel_push. 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 kernel_push? +

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