remote-mcp

Execute shell commands on remote MCP-enabled machines via Terminator CLI.

1.6k|219|Updated Apr 11, 2025
One-click install
npx skills add https://github.com/mediar-ai/terminator --skill remote-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: remote-mcp
Source: https://github.com/mediar-ai/terminator/tree/main/.claude/skills/remote-mcp
Command: npx skills add https://github.com/mediar-ai/terminator --skill remote-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables secure, auditable remote execution on MCP-enabled machines via the Terminator CLI, removing the need to manually connect to each host and copy commands.

Core Features & Use Cases

  • Remote Command Execution: Run shell/JS/PowerShell commands on a remote MCP endpoint.
  • UI and File Tasks: Check file existence, fetch desktop UI trees, take screenshots for verification.
  • Automation & Activation: Automate sequences and auto-activate when user mentions phrases like "remote MCP".
  • Use Case: Example: connect to five remote MCP endpoints and run a diagnostic script in a single workflow.

Quick Start

Connect to the MCP endpoint of your remote machine and run a simple shell command like a whoami to verify access.

Frequently Asked Questions about remote-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I execute commands on remote MCP-enabled machines?

Remote command execution on MCP endpoints uses the Terminator CLI to run shell, JavaScript, or PowerShell commands directly without manual host connections. Provide your MCP endpoint URL and optional MCP_AUTH_TOKEN to authenticate, then invoke run_command to execute scripts across Windows, Linux, and other MCP endpoints in a single workflow.

What can I do with remote MCP machines beyond running commands?

Beyond command execution, you can inspect desktop UI trees, capture screenshots for verification, check file existence, launch applications, click interface elements, type text, and wait for elements to appear. These capabilities enable comprehensive automation of remote machine interactions through Terminator CLI's get_desktop_elements, take_screenshot, open_application, click_element, type_text, and wait_for_element operations.

Do I need authentication to connect to remote MCP endpoints?

Connection requires an MCP endpoint URL; authentication is optional via MCP_AUTH_TOKEN environment variable. If your remote machine requires authentication, set the token before running Terminator CLI commands to establish secure, auditable access to the MCP endpoint.

Can I automate workflows across multiple remote MCP machines at once?

Yes, automate sequences across multiple remote MCP endpoints in a single workflow by specifying different endpoint URLs for each command invocation. This eliminates manual host-by-host connection and allows you to run diagnostic scripts or execute tasks concurrently on five or more remote machines through Terminator CLI.

What platforms and operating systems does remote MCP support?

Remote MCP execution supports Windows, Linux, and other MCP-enabled endpoints through Terminator CLI. The Skill works across heterogeneous environments, letting you manage shell commands on Linux systems and PowerShell scripts on Windows machines from a unified interface.