shell_exec

What shell_exec does on Hercules MCP

AI agents invoke shell_exec to trigger actions in Hercules 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 shell_exec is rated High

shell_exec executes shell commands whose effects are determined entirely by user input. In the context of a penetration testing tool, this permits running any command within the container—including reconnaissance, exploitation, lateral movement, and system compromise. Effects depend on arguments and are not reversible in some cases (e.g., file deletion, system modification).

From the tool's definition Tool named 'shell_exec' with no description; context indicates it is part of a penetration testing framework exposing 'offensive security tools' in a Kali Linux container. The name 'shell_exec' directly indicates execution of arbitrary shell commands.

The rule that runs shell_exec safely

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

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

Questions about shell_exec