ml_train_incident_classifier

What ml_train_incident_classifier does on ServiceNow MCP Server

AI agents invoke ml_train_incident_classifier to trigger actions in ServiceNow MCP Server. 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 ml_train_incident_classifier is rated High

Training an ML model is a computational operation that executes code with effects dependent on training data and parameters. It is not a simple create/modify operation (Write), but rather triggers a complex external process (Execute). Severity is high because a compromised training pipeline could poison the incident classifier, affecting all downstream incident categorization and routing decisions organization-wide.

From the tool's definition Tool name 'ml_train_incident_classifier' and description 'Trigger training of the incident classification ML solution' indicate execution of a machine learning training pipeline, which runs external code/operations.

The rule that runs ml_train_incident_classifier safely

PolicyLayer is an MCP gateway: it sits between your AI agents and ServiceNow MCP Server, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For ml_train_incident_classifier, this is the rule to start with:

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

Questions about ml_train_incident_classifier