training_score

What training_score does on 0nmcp

AI agents call training_score to retrieve information from 0nmcp without modifying anything. It is typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.

Why training_score is rated Low

This tool retrieves or evaluates training pair data and applies scoring logic, returning quality metrics. It does not modify, create, delete, or execute arbitrary code—it reads training pairs and produces an assessment output. The low severity reflects that misuse would only expose evaluation results, not cause system or data damage.

From the tool's definition Tool description states it 'Score[s] training pairs against rubrics for quality assessment.' The verb 'score' and 'assessment' indicate evaluation and measurement of existing data, with no modification, creation, or deletion of training pairs themselves.

The rule that runs training_score safely

PolicyLayer is an MCP gateway: it sits between your AI agents and 0nmcp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For training_score, this is the rule to start with:

{
  "version": "1",
  "default": "deny",
  "tools": {
    "training_score": {}
  }
}

Questions about training_score