launch_sandbox · MCP4Modal Sandbox MCP: Risk & Policy

WHAT IT DOES

What launch_sandbox does on MCP4Modal Sandbox

AI agents invoke launch_sandbox to trigger actions in MCP4Modal Sandbox. 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 launch_sandbox needs a policy

Launching a sandbox environment is an Execute action—it triggers creation of an external computational resource whose subsequent effects depend on what commands are run within it. While the immediate action is creation (Write), the purpose is to enable arbitrary code execution, making Execute the appropriate category.

From the tool's definition Tool name 'launch_sandbox' combined with server description stating it 'enables LLMs and AI assistants to create, manage, and interact with isolated cloud-based Python environments with GPU support on Modal.com'.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

launch_sandbox Rate-limited

launch_sandbox 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": {
    "launch_sandbox": {
      "limits": [
        {
          "counter": "launch_sandbox_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 MCP4Modal Sandbox, so you can run agents against it safely.

FAQ

Questions about launch_sandbox

What does the launch_sandbox tool do? +

launch_sandbox. It is categorised as a Execute tool in the MCP4Modal Sandbox MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on launch_sandbox? +

Register the MCP4Modal Sandbox MCP server in PolicyLayer and add a rule for launch_sandbox: 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 MCP4Modal Sandbox. Nothing to install.

What risk level is launch_sandbox? +

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

Can I rate-limit launch_sandbox? +

Yes. Add a rate_limit block to the launch_sandbox 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 launch_sandbox completely? +

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

launch_sandbox is provided by the MCP4Modal Sandbox MCP server (milkymap/mcp4modal_sandbox). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.