MCP server · Developer tools
Infrastructure MCP server
by radumalica
Connect your AI to your Linux servers, Docker, Kubernetes, Proxmox, Grafana and Cisco gear so it can check things, explain them, and fix them when you ask.
This is a small helper program that you run yourself, on your own computer or on a machine you control. It gives your AI assistant a safe way to look at and work with your Linux servers, Docker containers, Kubernetes clusters, Proxmox hosts, Grafana dashboards and Cisco network gear. It is handy if you look after servers or a small network and want help with the day to day checks.
What is an MCP server? The 30-second version
On its own, your AI can only chat. It cannot see your servers, your containers or your dashboards. An MCP server is a small helper program that gives your AI one new skill or one new connection, and here that connection is to your own infrastructure, the machines and services you run. Once it is connected, you ask your AI a normal question and it uses this helper to go and look things up for you. The helper is the only thing that actually touches your machines, so your AI never gets raw access to them.
What this MCP server does
You ask your AI a plain question, like why a server feels slow today. Your AI picks the right tool from this helper, for example the one that reports disk usage on a machine you named. The helper then talks to that machine using the details from a settings file you filled in earlier. It brings back a tidy result instead of a wall of text. Your AI reads that result and explains what is going on in normal words.
What you can do with it
- See which servers and devices are in your list, and filter them by tag
- Check how much disk, memory and CPU each machine is using
- Look at your Docker containers, their logs and how often they restart
- List Kubernetes pods, their logs and recent cluster events
- See which Grafana alerts are firing right now
- List your Proxmox virtual machines and the tasks run on a node
- Grab a Cisco switch or router configuration and compare it to the last copy
Try asking your AI
- “The archive server feels slow, can you find out why?”
- “Which Grafana alerts are firing right now?”
- “Show me the recent logs from the prometheus container on the archive server”
- “Has anything changed in the config on our main switch since the last backup?”
What it gives back to you
You get a short, readable answer in the chat, written by your AI in everyday words. Behind that, every tool sends back a tidy result, so you might see a list of machines with percentages, a table of containers, or just the log lines that matter. If you asked it to change something, such as restarting a container, it tells you exactly what it did. You never have to read a raw dump yourself.
Before you start
What you need
- The infrastructure-mcp program (a file you download for your system, or one you build from source)
- A settings file called an inventory, listing the machines and devices you want to reach
- Login details for those machines, such as SSH keys or API tokens
- An AI app that supports MCP, such as Claude Code, Cursor or Windsurf
Good to know
This helper can restart containers, start or stop virtual machines and take snapshots, so only let your AI make those changes when you are sure which machine it is working on.
Install it with your AI
Add Infrastructure MCP server to your AI, no technical skills needed
You don't install anything by hand. You copy one prompt, paste it into an AI that can work on your computer, and it checks, installs and connects the server for you, asking you when it needs something.
Sign in to get the install prompt
Members get a ready-made prompt that lets the Claude desktop app check Infrastructure MCP server, install it and connect it for them, step by step. You don't need any technical skills: you copy, paste and answer a few questions. Your connected AI can also find and install any of the 4,361 MCP servers here for you.
Who it's for
People who look after servers, containers or a small office network, like system administrators, IT support staff and developers who get called when something breaks.
