The Problem

The Solution: Remote Terminal MCP

How It Works

Features

Install See it in action: Demo Video If you've ever used AI to help manage a Linux server, you know the drill: One command at a time. Copy. Paste. Copy. Paste. Token waste. Some MCP servers try to solve this by letting Claude run commands directly. But they're completely blind - commands execute behind the scenes and you get back a text summary. No terminal. No visibility. No control. Would you let someone manage your production server while blindfolded? I built Remote Terminal MCP - an SSH server management tool where you and Claude share the same terminal session. The key: there's a full interactive terminal in your browser at localhost:8080. This isn't a log viewer - it's a real terminal like WSL, PuTTY, or Termius. Colored output, scroll history, copy/paste, arrow keys, the works. When Claude runs a command, you see it appear in your terminal in real-time. And you can type your own commands at any moment - intervene, correct course, run something manually. Behind the scenes, Remote Terminal uses a dual-stream architecture: Both streams share the same SSH session. You're never in the dark about what's happening on your server. It's a PyPI package, so getting started is simple: Configure Claude Desktop to use it, define your servers in hosts.yaml, and you're ready to go. Full setup instructions on GitHub. Would love feedback and suggestions. What features would you want to see next? Templates let you quickly answer FAQs or store snippets for re-use. Hide child comments as well For further actions, you may consider blocking this person and/or reporting abuse