Glances: One Terminal Dashboard to Actually See What Your System Is Doing

You've probably been there: your machine feels sluggish, the fan is spinning up, but you have no idea what's eating your CPU or filling your disk. Sure, you could open three different tools, wrestle with top's interface, or dig through logs. Or you could open one terminal window and see everything at a glance. That's the promise of Glances, an open-source, cross-platform monitoring tool that pulls your entire system's health into a single, readable dashboard.

What It Does

Glances is a system monitoring tool written in Python. It gives you real-time visibility into the core resources that matter: CPU, memory, disk, and network usage. But it doesn't stop there. It also tracks running processes, logged-in users, and even hardware-level details like temperatures, voltages, and fan speeds. If you're running containers, Glances has you covered there too, with support for management systems like Docker and LXC.

The architecture is worth noting. Glances is built on an open design that allows developers to add new plugins or export modules, so it's not a closed box. It grabs system information through Python libraries and presents it in a clean, easy-to-read interface. You can run it in a standalone terminal mode, or you can spin it up as a web server with a simple flag. In that web mode, it exposes both a user-friendly WebUI and a RESTful API, which opens up a lot of possibilities for scripting and integration.

Why It's Cool

What makes Glances stand out isn't just the volume of data it collects—it's how accessible that data is. Here's what caught my eye:

• It's a Swiss Army knife for monitoring. You get CPU, memory, disk, network, processes, temperatures, and container stats all in one place. That's not a collection of separate tools bolted together; it's a unified view that actually helps you see correlations. Your CPU spikes and your container stats jump at the same time? You can spot that instantly.

• Client/server mode is genuinely useful. You can run Glances on a remote machine and connect to it from your terminal, a web browser, or through an API. The README mentions support for XML-RPC and RESTful interfaces, which means you can query your systems programmatically. For anyone managing a small fleet of servers, that's a huge time-saver.

• It's got a built-in MCP server. This is a recent addition (available in Glances 4.5.1 and higher) and it's a clever move. The Model Context Protocol (MCP) lets AI assistants like Claude or Cursor query Glances directly. So instead of copying and pasting terminal output into your AI tool, the assistant can pull live system stats on its own. That's a practical bridge between traditional monitoring and the new wave of AI-assisted development.

• It exports data, not just displays it. Stats can be sent to files, external time/value databases, or CSV, and you can even push output directly to STDOUT. If you want to feed your monitoring data into another system, Glances doesn't hold it hostage.

• It's customizable. The README notes that you can tailor it to show only the information you care about. That's a nice touch for people who don't want to stare at a wall of numbers they'll never use.

How to Try It

Getting started is refreshingly simple. If you have Python installed, you can install Glances via pip:

$ pip install glances

For the standalone terminal interface, just run:

$ glances

That's it. You'll get a full dashboard in your terminal, no configuration required.

If you prefer a web interface, run:

$ glances -w

Then open http://<ip>:61208 in your browser. This also exposes a RESTful API if you want to interact with it programmatically.

And if you're curious about the AI assistant integration, just add the flag:

$ glances -w --enable-mcp

The MCP endpoint (using SSE transport) will be available at http://<ip>:61208/mcp/sse. You can find more details on client configuration in the documentation.

The project is actively maintained and has a solid community presence, including a subreddit. You can check out the source, contribute, or just read the docs over at the GitHub repository.

Final Thoughts

Glances is one of those tools that feels obvious once you see it—a single entry point for understanding what your system is doing. It's not trying to replace enterprise-grade monitoring platforms; it's aiming to be the thing you reach for when you just need to know what's going on right now, whether that's on your laptop or a remote server. If you're a developer, sysadmin, or just someone who likes to tinker, it's worth having in your toolkit. The learning curve is nearly flat, and the payoff is immediate.

Follow @githubprojects for more developer tools and open source projects.