Last updated: September 13, 2026
LLMs Mind Reader is a Chrome extension that shows the hidden mechanics behind a ChatGPT response as it streams in — the search queries ChatGPT issues, how it fans a question out into parallel searches, which tool it invokes, the full list of pages it found, and the sources it cites.
LLMs Mind Reader does not collect, transmit, sell, or store any of your data on any server. It has no backend, no analytics, and no tracking of any kind.
The extension reads ChatGPT’s own network response for the conversation you’re already having, directly inside your browser. It parses that response to display search queries, tool calls, search results, and citations in a side panel. All of this processing happens locally, on your device, in real time.
Parsed conversation data is held only in your browser’s memory for the current tab, for as long as that tab stays open. It is never written to disk, never synced, and is discarded automatically when the tab is closed or the browser exits. The extension does not use chrome.storage or any other persistence mechanism.
The side panel displays favicon images for the domains found in ChatGPT’s search results, loaded from Google’s public favicon service (google.com/s2/favicons). This sends only the bare domain name (e.g. “example.com”) to Google to fetch its icon — never any conversation content, search query text, or personal information.
This extension is not directed at children and does not knowingly collect any information from anyone.
If this policy changes, the “Last updated” date above will be revised accordingly.
Questions about this policy can be sent to saraseo86@gmail.com.