runbox_execute_command

What runbox_execute_command does on Runbox Website Builder MCP

AI agents invoke runbox_execute_command to trigger actions in Runbox Website Builder MCP. 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 runbox_execute_command is rated High

This tool allows arbitrary command execution in a Linux sandbox with access to both Node.js and Python runtimes. While sandboxed, the blast radius is critical because: (1) an AI agent could execute malicious commands that compromise the sandbox, exfiltrate data, or pivot to other systems; (2) combined with sibling tools like runbox_write_file and runbox_expose_port, it could be used to establish persistence, create…

From the tool's definition Tool description explicitly states 'Execute a command in an existing nodejs+python Runbox linux debian sandbox'. The tool enables execution of arbitrary commands in a sandboxed Linux environment with multiple runtime capabilities (Node.js, Python).

The rule that runs runbox_execute_command safely

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

{
  "version": "1",
  "default": "deny",
  "tools": {
    "runbox_execute_command": {
      "limits": [
        {
          "counter": "runbox_execute_command_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

Questions about runbox_execute_command