experiment_fork · ML Lab MCP: Risk & Policy

WHAT IT DOES

What experiment_fork does on ML Lab MCP

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

WHY IT NEEDS A POLICY

Why experiment_fork needs a policy

Forking an experiment creates a new experiment derived from an existing one, potentially with modified configuration. This is a reversible write operation (creating new data), not destructive. The blast radius is medium since it could spin up compute resources or training jobs depending on the backend.

From the tool's definition Fork an experiment with optional config changes

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

experiment_fork Rate-limited

experiment_fork 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": {
    "experiment_fork": {
      "limits": [
        {
          "counter": "experiment_fork_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 ML Lab MCP, so you can run agents against it safely.

FAQ

Questions about experiment_fork

What does the experiment_fork tool do? +

Fork an experiment with optional config changes. It is categorised as a Write tool in the ML Lab 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 experiment_fork? +

Register the ML Lab MCP server in PolicyLayer and add a rule for experiment_fork: 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 ML Lab MCP. Nothing to install.

What risk level is experiment_fork? +

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

Can I rate-limit experiment_fork? +

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

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

experiment_fork is provided by the ML Lab MCP server (pushpullcommitpush/ml-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.