Screenshots and element tagging

How capture works

When someone chooses to attach a screenshot, the widget reads the page as it is currently rendered in their browser and redraws it into an image. It crops to what they can actually see, so you get the viewport they were looking at rather than a full-page image where the problem is somewhere in the middle.

Nothing is sent to a server to be rendered. The image is produced on their device from their own view of the page, which is why it reflects their screen size, their zoom level, their cookie banner, and the state they had the page in.

Pointing at an element

Before capturing, the reporter can hover the page and click the exact element they are talking about. That element is outlined in the screenshot, and the report records which element it was.

This is the difference between "the button is broken" and a picture with the button circled. Ask your clients to use it; it removes most of the back and forth about which thing on the page they meant.

What ends up in the image

Exactly what was on their screen. If the page was showing a logged-in account, a customer record, or a test order, that is in the screenshot, because it was in the viewport when they pressed capture.

Screenshots are attached to the feedback item and visible to everyone with access to that project. If your client works with genuinely sensitive data, it is worth telling them to close such views before reporting. The rest of the captured context is deliberately narrower, and that is described in What the widget records.

Known issue: ghosted elements in Firefox

On some Firefox installations, small rounded elements such as pill-shaped badges and buttons come out hollow in the screenshot. The shadow around them renders, but the fill colour and the text inside are missing, leaving an outline where the element should be. The rest of the image is correct.

This is a bug in how Firefox rasterizes pages using the graphics card, not in what the widget captures. We confirmed that the page data handed to the browser is complete and correct; the colour is lost in Firefox's own drawing step. It affects every screenshot tool built on this browser capability, not only VibeVaults.

We have not applied a code-level workaround because the available fix would reduce screenshot quality for everyone in order to help an unknown number of Firefox users. We are measuring how often this actually happens before making that trade. If it is hitting your clients, tell us and it moves up the list.

Attachments

Reporters can also attach their own files: images, PDFs, documents, spreadsheets, and plain text. The limit is 10 MB per file and 10 files per report. Total storage depends on your plan.