# run_command

Run a bash command on the Ubuntu server as the SSH user.

Agent View of the PolicyLayer registry record for `run_command`. HTML page: https://policylayer.com/tools/mcp-control-plane/run-command

## Facts

- Tool: `run_command`
- Server: Mcp Control Plane (`albertik322-sudo/mcp-control-plane`) — https://policylayer.com/tools/mcp-control-plane.md
- Homepage: https://github.com/albertik322-sudo/mcp-control-plane
- Risk category: Execute (High risk)
- Registry record: grade F, identity unverified
- Server rate-limited: no
- Parameters: 0
- Recommended policy verdict: Rate-limited

## Example call (MCP tools/call, JSON-RPC 2.0)

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "run_command",
    "arguments": {}
  }
}
```

## Why run_command is rated High

This tool permits execution of arbitrary shell commands on a remote Ubuntu server, which is a classic Execute capability. The ability to run any bash command creates a critical blast radius if an AI agent misuses it—an attacker or misbehaving agent could compromise the server, exfiltrate data, modify system state, or pivot to other infrastructure.

From the tool's own definition: "Tool description explicitly states: 'Run a bash command on the Ubuntu server as the SSH user.' Bash commands can execute arbitrary code and trigger external operations with effects determined by the arguments provided."

## Use case

AI agents invoke run_command to trigger actions in Mcp Control Plane. 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.

## Recommended policy (PolicyLayer)

Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches Mcp Control Plane:

```json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "run_command": {
      "limits": [
        {
          "counter": "run_command_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}
```

## Other tools on Mcp Control Plane (285)

- `request_durable_task_cancel` — Destructive — https://policylayer.com/tools/mcp-control-plane/request-durable-task-cancel.md
- `steam_rom_manager_cli` — Destructive — https://policylayer.com/tools/mcp-control-plane/steam-rom-manager-cli.md
- `steam_rom_manager_nuke` — Destructive — https://policylayer.com/tools/mcp-control-plane/steam-rom-manager-nuke.md
- `steam_rom_manager_remove` — Destructive — https://policylayer.com/tools/mcp-control-plane/steam-rom-manager-remove.md
- `activate_window` — Execute — https://policylayer.com/tools/mcp-control-plane/activate-window.md
- `admin_run` — Execute — https://policylayer.com/tools/mcp-control-plane/admin-run.md
- `api_sentence` — Execute — https://policylayer.com/tools/mcp-control-plane/api-sentence.md
- `apply_unified_patch_remote` — Execute — https://policylayer.com/tools/mcp-control-plane/apply-unified-patch-remote.md
- `apt_maintenance` — Execute — https://policylayer.com/tools/mcp-control-plane/apt-maintenance.md
- `call_service` — Execute — https://policylayer.com/tools/mcp-control-plane/call-service.md
- `click` — Execute — https://policylayer.com/tools/mcp-control-plane/click.md
- `click_at` — Execute — https://policylayer.com/tools/mcp-control-plane/click-at.md
- `click_in_window` — Execute — https://policylayer.com/tools/mcp-control-plane/click-in-window.md
- `comfy_generate_from_asset` — Execute — https://policylayer.com/tools/mcp-control-plane/comfy-generate-from-asset.md
- `comfy_interrupt` — Execute — https://policylayer.com/tools/mcp-control-plane/comfy-interrupt.md
- `comfy_restart` — Execute — https://policylayer.com/tools/mcp-control-plane/comfy-restart.md
- `deck_exec` — Execute — https://policylayer.com/tools/mcp-control-plane/deck-exec.md
- `deck_probe` — Execute — https://policylayer.com/tools/mcp-control-plane/deck-probe.md
- `dns_test` — Execute — https://policylayer.com/tools/mcp-control-plane/dns-test.md
- `docker_compose` — Execute — https://policylayer.com/tools/mcp-control-plane/docker-compose.md
- `docker_container_action` — Execute — https://policylayer.com/tools/mcp-control-plane/docker-container-action.md
- `docker_exec` — Execute — https://policylayer.com/tools/mcp-control-plane/docker-exec.md
- `double_click` — Execute — https://policylayer.com/tools/mcp-control-plane/double-click.md
- `drag_to` — Execute — https://policylayer.com/tools/mcp-control-plane/drag-to.md
- `focus_window_win32` — Execute — https://policylayer.com/tools/mcp-control-plane/focus-window-win32.md
- `hold_key` — Execute — https://policylayer.com/tools/mcp-control-plane/hold-key.md
- `hotkey` — Execute — https://policylayer.com/tools/mcp-control-plane/hotkey.md
- `install_flatpak_emulators` — Execute — https://policylayer.com/tools/mcp-control-plane/install-flatpak-emulators.md
- `key_down` — Execute — https://policylayer.com/tools/mcp-control-plane/key-down.md
- `key_up` — Execute — https://policylayer.com/tools/mcp-control-plane/key-up.md
- …and 255 more: https://policylayer.com/tools/mcp-control-plane.md

## For agents

This record is a snapshot. Live verdicts and the full registry:

- Check every server in your MCP config at once: `npx -y policylayer stack`
- Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md)
- Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events`

---

Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=mcp-control-plane · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/mcp-control-plane