start-gpu-instance

THE RISK

High Risk

Why start-gpu-instance needs a policy

Starting a GPU instance triggers external infrastructure and initiates resource consumption, which are operational effects dependent on which instance is targeted. This is an Execute-category tool rather than Write because it doesn't modify instance configuration but rather triggers a state change with external computational consequences.

From the tool's definition Tool name 'start-gpu-instance' indicates activation of computational resources. Server description states it supports 'GPU instance operations (list, create, start, stop, etc.)', and this tool performs the 'start' operation on GPU instances.

Documented attack patterns abuse exactly the kind of access start-gpu-instance gives an agent:

POLICY

How to control start-gpu-instance

PolicyLayer is an MCP gateway — it sits between your AI agents and Novita MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for start-gpu-instance:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "start-gpu-instance": {
      "limits": [
        {
          "counter": "start-gpu-instance_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

start-gpu-instance 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.

Create a free account and register Novita MCP Server — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

RATE-LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Novita MCP Server tools

Destructive delete-container-registry-auth delete-container-registry-auth Destructive delete-gpu-instance delete-gpu-instance Destructive delete-network-storage delete-network-storage Destructive delete-template delete-template Execute restart-gpu-instance restart-gpu-instance Execute stop-gpu-instance stop-gpu-instance Write create-container-registry-auth create-container-registry-auth Write create-gpu-instance create-gpu-instance

All 20 Novita MCP Server tools →

Execute tools on other servers

Android Forensics ADB MCP Server adb_shell_command EMBA-MCP run_emba_scan Frida Game Hacking MCP remove_breakpoint Procmon request_elevation

Go deeper

High-risk tools in Novita MCP Server → Every high-risk tool this server exposes, in one view.
All high-risk MCP tools → Every high-risk tool across the catalogue, and the policy pattern that contains them.
The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Agent sandbox → Bounding what an agent can execute and where the boundary should sit.
Least privilege for MCP → Granting agents the minimum tool surface the task needs.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about start-gpu-instance

What does the start-gpu-instance tool do? +

start-gpu-instance. It is categorised as a Execute tool in the Novita 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-gpu-instance? +

Register the Novita MCP Server MCP server in PolicyLayer and add a rule for start-gpu-instance: 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 Novita MCP Server. Nothing to install.

What risk level is start-gpu-instance? +

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

Can I rate-limit start-gpu-instance? +

Yes. Add a rate_limit block to the start-gpu-instance 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-gpu-instance completely? +

Set action: deny in the PolicyLayer policy for start-gpu-instance. 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-gpu-instance? +

start-gpu-instance is provided by the Novita MCP Server MCP server (novitalabs/novita-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Novita MCP Server tool call.

Start from Novita MCP Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →