Chrome extension

Exeora for Chrome puts the dashboard's Workspace screen in Chrome's side panel, next to whatever tab you are on: the pull request you are reviewing, the preview you are testing, the issue you are writing. It is the same screen, built from the same code, so what you can do in it is what Source Control and terminal describes.

Install#

Add Exeora from the Chrome Web Store, then pin it to the toolbar. Clicking its icon opens the side panel in that window; the panel stays open as you switch tabs.

Signing in#

The panel asks you to sign in with Exeora. Chrome opens a window of its own on the Exeora sign-in page, the same one the dashboard uses, and the first time on an account it also asks you to authorize the extension. There is no password or key to paste: it is an OAuth sign-in with PKCE, and the panel keeps the session it is given.

A restart does not sign you out. Sign out in the panel's header forgets the session in the browser and revokes it on Exeora.

What the panel does#

• Pick a project and one of its workspaces, on any of your machines or on Exeora Cloud.
• See what changed, read the diffs, stage, commit, pull and push.
• Open terminals on the workspace, which keep running while you move between tabs.
• Dashboard in the header opens the workspace you are looking at in the full dashboard, and anything the panel does not show (adding a project, a project's settings, clients, activity) opens there too.

Everything the panel does goes through the same checks as the dashboard. A project's policy applies to its terminals and Source Control exactly as it does there, and the same actions appear in the activity log.

What it can reach#

The extension asks Chrome for three permissions: the side panel, the identity window it signs in through, and its own storage. Its only host permission is exeora.dev, where all of its requests go, and it reads nothing from the pages you visit. The one thing it loads from anywhere else is your profile picture, as an image from the provider you signed in with (GitHub or Google).

It is not an MCP client and does not give an AI anything: your agents connect through a client URL as before. The panel is for you, to see and act on what they did.

Revoking it#

In the dashboard, Settings shows Exeora for Chrome once you have approved it: when, and how many browsers are signed in. Revoking it there signs out every side panel at once and forgets the approval, so the next sign-in asks for consent again.

On a self-hosted gateway#

The extension signs in against one gateway, fixed when it is built, and a gateway only accepts the extension ids it lists. The one in the Chrome Web Store talks to exeora.dev. To use it with your own gateway, build it for that gateway and allow its id: see Self-hosting.

What is stored#

In your browser, a refresh token in Chrome's extension storage and a short-lived access token in memory. On Exeora, that you approved the extension and one session per signed-in browser. The privacy policy has the details.