# Browser extension

Save any page, selection or screenshot as a GetItDone task from Chrome or Edge.

The GetItDone browser extension turns whatever is in front of you — the page, a
paragraph you selected, a screenshot of what is on screen — into a task in your
workspace, with the source URL attached.

It supports **Chrome and Edge** today. Firefox is planned; see
[What is not in this version](#what-is-not-in-this-version).

## What it does

**Save the current page.** Click the toolbar icon, or press `Alt+Shift+G`. The
page title arrives pre-filled as the task title and the URL lands in the task
notes. Edit anything, optionally pick a project and a due date, then press
**Add task**.

**Save a selection.** Highlight text on any page, right-click, and choose **Add
selection to GetItDone**. The first line of the selection becomes the task
title; if the selection is longer, the whole of it is kept in the description so
nothing is lost.

**Attach a screenshot.** Press **Attach screenshot of this page** to capture the
visible part of the tab. You see the image before anything is sent, and can
remove it. On submit it is uploaded as a normal task attachment.

Nothing leaves your browser until you press **Add task**.

## Signing in

Choosing **Sign in to GetItDone** opens GetItDone's own sign-in and consent
pages in a browser-controlled window. Your password never reaches the extension:
it receives a scoped access token, and only these four permissions are
requested — read and create tasks, read projects, read your workspace.

You can revoke the connection at any time from **Profile → Connected Apps** in
the web app. Signing out from the extension's options page clears the tokens
stored in that browser and ends the connection on GetItDone's side: the refresh
token can no longer renew it, and the access token that browser already held
stops working at once.

The free plan works. Creating a task costs one API call against your plan's
allowance; adding a screenshot costs two more.

## Privacy

The extension asks for four permissions and one host:

- `storage` — the sign-in tokens, your default project and a short cache of your
  project names.
- `identity` — used only to open the sign-in window.
- `activeTab` — the title and URL of the tab you are on **at the moment you open
  the extension**, and the screenshot when you press the button. It does not run
  on pages in the background.
- `contextMenus` — the one right-click item.
- `https://app.nowgetitdone.com/*` — the only server the extension contacts.

There is no analytics and no error reporting of any kind in this version.

Some pages cannot be read by any extension — `chrome://` pages, the extension
gallery, and PDF viewers. On those the popup says so and you can type the task
yourself.

## What is not in this version

Being explicit, so nothing here reads as a promise:

- **No AI extraction.** The extension attaches a screenshot; it does not turn it
  into several proposed tasks. That capability exists in the web app (Capture)
  and is planned for the extension.
- **No voice capture.**
- **No offline queue** — a task is created when you press the button, or you see
  the error.
- **No task browsing** from the extension, and no in-page sidebar.
- **Firefox is not published yet.** The build supports it; the Add-ons store
  requires a reproducible source-code submission that has not been completed.

## Related

The extension is built on the same [public API](/docs/api) anyone can use —
`POST /v1/tasks` plus the two-phase attachment upload — with OAuth 2.1 and PKCE.
If you want the same capability somewhere else, that API is the way in.