Every AI browser extension promises the same thing: help, right there, beside the page. The question that separates them is not the feature list. It is what the extension can see, and when.
Ctrl+Shift+E, and can bring the visible page into a task when context matters. It reads the current tab only when you invoke it, and its standing site access is limited to Oppermind’s own domains — both of which you can verify in Chrome’s extension details, and should. If you install nothing else from this article, install the habit of reading that page.
A browser is where most knowledge work actually happens: the ticket, the article, the supplier’s page, the form, the thread. An AI assistant that lives in a separate tab makes you carry all of that across by hand. An assistant in a side panel does not. That is the entire pitch, and it is a good one.
It is also why the permission model matters more for a browser extension than for almost any other kind of software. A side panel that can help with any page is, by construction, a thing that could read any page. The design question is whether it does so only when you ask.
Click the icon or use the shortcut and a panel opens beside the current page. It is the Oppermind chat — the same account, the same Skills and Brand Kits and connectors, the same plan — without leaving the tab. Ask a question, draft a reply, extract the five figures from the report you are reading.
Highlight text on the page and press Ctrl+Shift+E. The selection lands in the panel as context for your next request. This is the feature you will use most, and it is worth understanding why it is better than sending the whole page: it is faster, it is cheaper, it is better for privacy, and it produces a better answer, because the model is reading the paragraph you care about rather than the navigation menu, the cookie banner and the footer.
When the whole visible page is the context — a long article to summarise, a listing to compare, a form to fill from — the extension can bring it into the task. This is the capability that needs the permission discussed below, and it is the one you should expect any honest extension to explain.
If you have Oppermind’s desktop co-pilot installed — a separate Windows application — the extension can connect to it, so a task that starts on a web page can continue as work on your PC. It is optional. Without the desktop application installed, the extension does nothing on your machine.
This is the part most extension write-ups skip. Here is Oppermind for Chrome’s permission model in plain language, and then how to verify it yourself — for this extension or any other.
It reads the current tab only when you invoke it. The extension uses Chrome’s activeTab permission, which grants access to the page you are on at the moment you click the icon, use a shortcut or open the panel — and not otherwise. It is not reading the tabs you have open in the background.
Its standing site access is limited to Oppermind’s own domains. The extension declares host permissions for oppermind.com and its variants, so that the panel can talk to your workspace. It does not declare standing access to all websites.
It stores settings locally and can talk to the desktop application. Storage for your preferences, and native messaging for the optional desktop bridge above.
Now the important bit: do not take our word for it. Open Chrome’s extensions page, click Details on any extension, and read two things — the permissions list, and the “Site access” setting. “On click” is a very different grant from “On all sites”. If an extension’s feature list is “summarise pages” and its site access is “read and change all your data on all websites”, the gap between those two is the thing to think about. That check takes thirty seconds and applies to every extension you have ever installed.
| What you want to know | Where to look | What good looks like |
|---|---|---|
| When can it read a page? | Details → Site access | “On click” or a short list of named sites, not “On all sites” |
| What else can it do? | Details → Permissions | A short list that matches the features it advertises |
| Who publishes it? | The store listing | A named company with a website and a privacy policy |
| What does it send, and where? | The publisher’s privacy policy | Plain statements you can understand without a lawyer |
The side panel is for work that starts on a page: read this, reply to that, pull these numbers out, compare this to the one in the other tab. It is quick, it is contextual, and it removes the copy-paste that makes people stop using assistants.
It is not where you build things. A deck, a model, a document from a brief is a workspace job, and the result opens in an editor with track changes and version history. If the work is a file rather than a page, the guide to Oppermind’s surfaces is sixteen minutes on which one fits what, and covers the Office add-ins and the mobile apps alongside the extension.
Oppermind for Chrome is on the Chrome Web Store. It uses the same account and plan as the web workspace — free with no card, then Starter A$9.95, Pro A$29.99 and Pro Plus A$59.99 a month — and it is one of several surfaces the same subscription runs on: the web workspace, the Office add-ins, the iOS and Android apps, and the desktop co-pilot. Unlimited documents, spreadsheets, decks and designs on every plan.
Whatever you’re here to make, make more of it.
A side panel is convenient, and convenience is how people end up sending things they should not. The selection shortcut exists partly for that reason: sending the paragraph is almost always enough, and it is a habit worth keeping regardless of which extension you use. And the permission model above is ours as at the date shown; extensions get updated, so the check in Chrome’s details page is the authority, not this article.
The useful version opens a side panel beside whatever page you are on and lets you bring that page — or just the text you have selected — into a task without copying it out to another tab. Summarise, extract, reply, rewrite, compare, with the page as context. The difference between a good one and a bad one is mostly what it is permitted to read and when.
No. It uses Chrome’s activeTab permission, which grants access to the current tab only when you invoke the extension — by clicking it, using the shortcut, or opening the panel. Its broad host permissions are limited to Oppermind’s own domains. You can verify both yourself in Chrome’s extension details, and you should, for any extension.
Highlight the text and press Ctrl+Shift+E, or use the shortcut you have assigned. The selection goes to the side panel as context for your next request. Sending the selection rather than the page is faster, cheaper and better for privacy, and it produces a better answer because the model is not reading navigation menus and footers.
The extension can connect to Oppermind’s desktop co-pilot, a separate Windows install, so a task that starts on a web page can hand off to work on your PC. It is optional, it requires the desktop application to be installed, and the extension does nothing on your machine without it.
Open Chrome’s extensions page, click Details on the extension, and read the permissions list and the site access setting. “Read and change all your data on all websites” is a very different grant from “when you click the extension”. If an extension asks for far more than its feature list needs, that gap is the thing to worry about.
Permissions described are those declared by Oppermind for Chrome as at 12 September 2026 and may change with updates; the permissions and site-access settings shown in Chrome’s extension details are authoritative. The desktop bridge requires Oppermind’s desktop application, a separate Windows install. Chrome and Chrome Web Store are trade marks of Google LLC; Oppermind is not affiliated with, endorsed by, or sponsored by Google. This article is general information about software, not advice. Oppermind platform prices are current as at 12 September 2026, are in Australian dollars, and are subject to the plan terms at checkout.
A side panel on any page, a selection shortcut, and permissions you can read for yourself.