DesktopPiP Connector에 대한 개인정보처리방침
DesktopPiP Connector 제작자: Spirited Systems
DesktopPiP Connector Privacy Policy
Last updated: August 4, 2026
DesktopPiP Connector helps the DesktopPiP Windows app save and restore presets that include browser pages. The connector runs in your browser and communicates with DesktopPiP on the same computer.
The short version
Spirited Systems does not automatically receive your browsing data.
The connector has no accounts, advertising, analytics, telemetry, or remote data service, and it does not sell user data.
To save and restore a browser preset, the connector uses the relevant tab's URL, title, window geometry, and document scroll coordinates.
That information is exchanged with the DesktopPiP app over a connection to 127.0.0.1, the local loopback address on your computer.
The connector stores a connection preference locally and keeps limited restore state in browser-session memory. It does not use cloud or synchronized extension storage.
What the connector does
When you save a DesktopPiP preset, the app supplies the position and size of the captured browser window. The connector matches that geometry to a browser window and returns the active tab's URL, title, browser type, and horizontal and vertical document scroll coordinates.
When you load a preset, DesktopPiP supplies the saved URL, title, scroll coordinates, and desired window geometry. The connector can:
reuse a normal browser window whose active tab already has that URL, or open a new browser window for it;
move or resize that browser window;
restore the saved document scroll position after the page loads and after the same page later reloads during that browser session;
briefly detect and close background, opener-less duplicate tabs created by browser session restore; and
close a single-tab about:blank window that DesktopPiP created solely to wake a suspended browser background worker.
The connector also responds to browser startup, tab, window-focus, and periodic alarm events to maintain its local connection. These events are not used to create a browsing-history profile.
Browser information accessed
The connector uses the following information only to provide browser-preset save, restore, reconnection, and cleanup behavior:
Tab URL and title. Used to identify the active page in a captured window and to find the correct window during restore.
Browser window and tab identifiers. Temporary browser-assigned numbers used to distinguish windows and tabs during restore and cleanup.
Browser window position and size. Used to match the captured window and place a restored window.
Document scroll coordinates. The horizontal and vertical values reported by window.scrollX and window.scrollY.
Browser type and connector version. Used by DesktopPiP to direct requests to the correct connected browser and diagnose compatibility.
To read or restore scrolling, the connector injects a small bundled function into the selected tab. It reads only window.scrollX and window.scrollY and calls window.scrollTo. It does not read page text, images, form fields, passwords, cookies, authentication data, or website storage. It cannot restore scrolling on browser-restricted pages or pages that keep their scroll state in an inner element, iframe, or application-managed canvas.
Local communication and website requests
The connector communicates with a service listening on ws://127.0.0.1:8770 through ws://127.0.0.1:8779. Loopback traffic remains on your computer and is not routed over the internet. The connector sends browser metadata only in response to the small DesktopPiP save/restore protocol; it has no developer-operated endpoint to which it can upload that information.
The local connection does not use a shared secret. DesktopPiP verifies that an incoming connection has a browser-extension origin, but the connector cannot cryptographically distinguish DesktopPiP from another local program that binds one of those loopback ports first. A program already running under your Windows account could therefore impersonate the local app. This does not make the connector reachable from another device or from a website.
Opening a saved URL causes your browser to make the ordinary network requests needed to load that website. Those requests are handled by your browser, the website, and their respective privacy policies. The connector does not add your saved title, scroll coordinates, window geometry, or other preset metadata to those website requests.
Storage and retention
The connector stores the following data on your device:
Connection preference: the last successful DesktopPiP loopback port is stored in browser-managed local extension storage so reconnection is faster. It remains until the extension or its stored data is removed.
Reload restoration: for a tab restored with a saved scroll position, the target URL, tab identifier, and two scroll-coordinate numbers are kept in browser-managed session storage. The entry is removed when that tab closes or navigates to another URL. Session storage is cleared when the browser restarts or the extension is disabled, reloaded, or updated.
Duplicate cleanup: a target URL and tab identifier are held in memory for about 12 seconds after a restore so the connector can recognize a duplicate tab created by browser session restore.
This extension data is not synchronized between devices or stored remotely.
The browser's extension console can contain operational diagnostics, including URLs, titles, scroll coordinates, browser-assigned tab or window identifiers, window geometry, and connection status. The browser controls access to and retention of that console output. Spirited Systems does not receive it automatically.
DesktopPiP stores the URL, title, browser information, geometry, and saved scroll position in the local preset you chose to create. That storage belongs to the desktop app, not the connector. See the DesktopPiP Privacy Policy for the app's storage and deletion details.
Permissions
tabs allows the connector to use tab URLs, titles, identifiers, status, and activation events for matching, reload restoration, reconnection, and narrowly scoped duplicate cleanup.
alarms periodically wakes the background worker so it can reconnect to DesktopPiP. An alarm does not itself read page data.
scripting and HTTP/HTTPS host access allow the bundled scroll function to run in the specific tab being saved or restored.
storage holds the local connection preference and current-browser-session reload-restoration entries described above.
Browser window access allows the connector to find, open, move, and close the narrowly described browser windows. Browser window APIs require no separate manifest permission.
How information is used and shared
Browser information is used only to provide and maintain DesktopPiP's browser preset feature. It is shared only with the local DesktopPiP service (or a local program impersonating that service as described above), unless disclosure is required by law. Spirited Systems does not receive it automatically, sell it, use it for advertising or profiling, or permit people to read it. If you choose to send a preset, screenshot, log, or console output to Spirited Systems for support, the submitted material may be reviewed to answer your request.
DesktopPiP Connector's use of information received from Google APIs adheres to the Chrome Web Store User Data Policy, including its Limited Use requirements.
Remote code
The connector does not download or execute remote code. Its executable code is included in the extension package distributed by the browser store.
Changes to this policy
If the connector's data practices change, this policy and the relevant browser store disclosures will be updated. The date at the top identifies the latest revision.
Contact
Privacy questions can be sent to Spirited Systems at spiritedsystems@gmail.com.