shadcn.io is not affiliated with official shadcn/ui
Container Registry Dashboard Block for Svelte
Six images in a list beside the chosen one's tags, the layers of the chosen tag drawn as a bar by size with the command behind each, vulnerabilities by severity, and a read-only pull command with a copy button.
Installation
Overview
Look inside your images before you ship them with this container registry dashboard for Svelte with Ark UI. A hairline frame with a plus mark on each corner puts a list of six Manifestry images at the start, each with its newest tag, its size, its pulls in 30 days and its scan result in words. Choosing one fills the detail beside it: a read-only pull command with a Copy Button, a table of its tags with architectures, short digest and the day it was pushed, the layers of the chosen tag drawn as one bar of segments by size with the command that made each, and its vulnerabilities as four counts: Critical, High, Medium and Low. The totals are summed from the layers. It is built from Copy Button, Input Group, Badge and Guides. For platform teams who audit what runs in production.
FAQ
How do I list my own images?
Edit IMAGES: each image has a name, its pulls, the commands that made its layers, and its tags, newest first. A tag has its architectures, a short digest, the pushed time, the sizes of its layers in MB in the order of the commands, and its vulns counts for Critical, High, Medium and Low. Sizes, the bar and the totals follow.
How is the layer bar drawn?
Each segment is as wide as its layer's share of the tag's total, in five foreground weights from solid to faint, and the list under it names the command and size of every layer in order. The bar is described in words as well, so it is not the only way to read it.
What does the copy button copy?
The command in the field above it, docker pull with the registry host, the image and the chosen tag. The field is read-only; the button says Pull command copied to screen readers, and shows a check for a moment. If the clipboard refuses, it says so instead.
Are the scan results real?
No. Counts are typed for each tag and nothing is scanned or pulled. In your app fill vulns from your scanner's report and sizes from the image manifest; the summary under the panel reads from them.