start_experiment · NeuronSearchLab MCP: Risk & Policy

WHAT IT DOES

What start_experiment does on NeuronSearchLab

AI agents invoke start_experiment to trigger actions in NeuronSearchLab. 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 start_experiment needs a policy

This tool executes an operation that initiates an A/B experiment and immediately affects user traffic distribution. While not destructive or financial, it triggers a consequential external operation whose effects depend on experiment configuration. The immediate activation of traffic splitting constitutes Execute-level risk.

From the tool's definition Tool description states 'Start an A/B experiment. Traffic will begin splitting between variants immediately.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

start_experiment Rate-limited

start_experiment 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": {
    "start_experiment": {
      "limits": [
        {
          "counter": "start_experiment_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 NeuronSearchLab, so you can run agents against it safely.

FAQ

Questions about start_experiment

What does the start_experiment tool do? +

Start an A/B experiment. Traffic will begin splitting between variants immediately. The experiment moves from draft to running status. It is categorised as a Execute tool in the NeuronSearchLab MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on start_experiment? +

Register the NeuronSearchLab MCP server in PolicyLayer and add a rule for start_experiment: 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 NeuronSearchLab. Nothing to install.

What risk level is start_experiment? +

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

Can I rate-limit start_experiment? +

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

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

start_experiment is provided by the NeuronSearchLab MCP server (neuronsearchlab/mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.