start_server · Llauncher MCP: Risk & Policy

WHAT IT DOES

What start_server does on Llauncher

AI agents invoke start_server to trigger actions in Llauncher. 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_server needs a policy

Starting a server is an Execute action because it initiates and runs an external process (llama-server instance) whose effects depend on the provided arguments (port, model). This is not merely a Read (no data retrieval), Write (not reversibly modifying data), Destructive (reversible via stop_server), or Financial operation.

From the tool's definition Tool name 'start_server' combined with server description indicating 'lifecycle management' and 'control local or remote inference servers' establishes that this tool triggers external operations (starting an inference server process).

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

start_server Rate-limited

start_server 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_server": {
      "limits": [
        {
          "counter": "start_server_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 Llauncher, so you can run agents against it safely.

FAQ

Questions about start_server

What does the start_server tool do? +

Start a model on an empty port. Fails with. It is categorised as a Execute tool in the Llauncher 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_server? +

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

What risk level is start_server? +

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

Can I rate-limit start_server? +

Yes. Add a rate_limit block to the start_server 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_server completely? +

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

start_server is provided by the Llauncher MCP server (shanevcantwell/llauncher). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.