SnapLocal

Screenshots that stay on your computer.

Manifest V3 11 languages GPL-3.0 No network requests Free, no upsell

Add to Microsoft Edge

Version 1.0.0, free and open source.
Chrome Web Store and Firefox are next.

Four ways to capture

Select area

Drag a rectangle over any part of the page.

Visible part

Exactly what is on screen right now.

Full page

Scrolls the page and stitches it into one image.

Delayed

A countdown first, for menus and hover states.

Selecting a region of a page, with the rest of the page dimmed — Drag to select. The size readout follows the selection.

A real editor

Rectangle, ellipse, arrow, line, pencil, highlighter and text, with colour, thickness and font controls. Undo and redo everything — including the crop, which is non-destructive, so undoing it brings the discarded pixels back.

Copy straight to the clipboard, or save as PNG.

The editor with a highlight, a rectangle, an arrow and a text label — Annotating a capture. Text gets a contrasting outline so it stays readable on any background.

Blur that tells you the truth

Blur and pixelate are here. But blurred text can sometimes be reconstructed, so SnapLocal also gives you Redact: a solid block that cannot be reversed. The first time you blur something, it says so, and offers to convert that region to a redaction.

A privacy tool that quietly lets you leak a password is not a privacy tool.

A capture with one line redacted behind a solid bar and another blurred — Redaction on top, blur below. Only one of them is irreversible.

Private by construction, not by promise

No network requests at all. No analytics, no crash reporting, no CDN, no remote code.
No account, no sign-in.
No host permissions. It can only read a page at the moment you start a capture on it.
Settings stay in this browser, on this computer.
Open source under GPL-3.0. Read the code yourself — and any published fork has to stay open too.

The extension ships unminified, so the code in the store is the code you can read in the repository.

Help translate

The interface is available in English, Português (Brasil), Português, Español, Français, Italiano, Deutsch, Русский, Polski, Türkçe and Bahasa Indonesia. Adding a language means translating one JSON file — no coding, no build step. See the contributing guide.