Tabby Server Stats Plugin
A plugin for
Tabby Terminal that displays real-time server statistics (CPU, RAM, Disk, Network) and
custom metrics when connected via SSH / Local Shell.
!
Preview image
Features
*
Real-time Monitoring: Displays CPU usage, RAM usage, Disk usage, and Network upload/download speeds out of the box.
*
Custom Metrics Engine: Define your own metrics using shell commands (e.g., GPU usage, Temperature, Docker container count).
*
Progress Bars: Visual bars for percentage-based data.
*
Text Values: Display raw data with units (e.g., "45°C", "3 Users").
*
Preset Library: One-click import for common metrics (GPU, Uptime, Temperature, etc.) from the community repository.
*
Flexible UI:
*
Bottom Bar Mode: An unobtrusive bar at the bottom of the terminal (docked inside the pane, won't overlap sidebars).
*
Floating Panel Mode: A draggable widget that floats over the content.
*
Highly Customizable:
*
Drag & Drop Sorting: Easily reorder metrics in the settings.
*
Visual Customization: Change chart colors, opacity, and layout (Vertical/Horizontal).
*
Multi-language Support: Interface available in English and Chinese.
*
Zero Dependency: Uses standard Linux commands via the SSH channel. No agent installation required on the server.
Installation
1. Open
Tabby Settings.
2. Go to
Plugins.
3. Search for tabby-server-stats.
4. Click
Install.
Usage
The stats will automatically appear when you connect to a Linux server via SSH or Local Shell.
You can toggle visibility using the "Activity" icon in the toolbar.
$3
Go to
Settings \-\> Server Stats to manage your metrics.
####
1\. Using the Preset Library (Recommended)
1. Click the
"Fetch from GitHub" button in the settings panel.
2. Browse the list of community presets (e.g., NVIDIA GPU, CPU Temp, Uptime).
3. Click
Add next to the metric you want.
4. It will immediately appear in your status bar.
####
2\. Adding Manually
You can define any metric by providing a shell command.
*
Label: Name of the metric (e.g., "GPU").
*
Command: A shell command that outputs a
single number or string.
Example (NVidia GPU)*:
nvidia-smi --query-gpu=utilization.gpu --format=csv,noheader,nounits
Example (Active Users)*:
who | grep -c pts
*
Type:
*
Progress Bar: Requires the command to return a number between 0-100.
*
Text Value: Displays whatever the command outputs.
License
MIT