Privacy policy

The extension reads the current ChatGPT page and may request the current conversation payload from ChatGPT so it can build a navigable outline, including headings that are not mounted in the viewport yet.

The current implementation still has no separate backend service. It uses a minimal background service worker plus an offscreen document and dedicated worker to host the local extension-side runtime and read, create, or remove Chrome-native bookmarks for heading references.

Conversation structure is processed in the active page, and the extension may request ChatGPT's own authenticated conversation endpoints to load heading and prompt labels for the sidebar.

If you bookmark a heading, the extension creates a Chrome bookmark whose URL includes the target heading id, so Chrome can show the bookmark and the extension can locate that heading when the bookmark is opened.

The extension is scoped to ChatGPT surfaces so the content script and stylesheet can run where the outline is needed.

Heading extraction from the conversation payload, tree construction, layout measurement, and scroll targeting happen in your browser session.

Bookmarked heading references are stored as Chrome bookmarks and are not synced to a service we operate.

The extension does not forward conversation content to our own servers or unrelated third-party analytics services for outline generation.

The extension depends on the ChatGPT page and ChatGPT conversation endpoints themselves, and does not integrate with analytics, payment, or unrelated remote processing services.

If the runtime architecture changes in the future, this page should be updated in the same change set that introduces the new data flow.