start_training · Ultralytics MCP Server MCP: Risk & Policy

WHAT IT DOES

What start_training does on Ultralytics MCP Server

AI agents invoke start_training to trigger actions in Ultralytics 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 IT NEEDS A POLICY

Why start_training needs a policy

This tool executes a training job rather than simply reading data or writing static configuration. Training jobs consume computational resources, run complex algorithms, and their outcomes depend on provided arguments. While not destructive (training can be stopped and rerun) or financial, it clearly fits Execute category—it triggers external operations with effects contingent on parameters.

From the tool's definition Tool description states 'Start a new YOLO training job' which initiates an external operation (machine learning model training) whose effects depend on the specified parameters.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

start_training Rate-limited

start_training 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_training": {
      "limits": [
        {
          "counter": "start_training_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 Ultralytics MCP Server, so you can run agents against it safely.

FAQ

Questions about start_training

What does the start_training tool do? +

Start a new YOLO training job with specified parameters. Returns training job ID and status. It is categorised as a Execute tool in the Ultralytics MCP Server 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_training? +

Register the Ultralytics MCP Server MCP server in PolicyLayer and add a rule for start_training: 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 Ultralytics MCP Server. Nothing to install.

What risk level is start_training? +

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

Can I rate-limit start_training? +

Yes. Add a rate_limit block to the start_training 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_training completely? +

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

start_training is provided by the Ultralytics MCP Server MCP server (metehanyasar11/ultralytics_mcp_server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.