Privacy Policy

Patent Citation Tool

Last updated: June 2026

Patent Citation Tool does not collect, store, or transmit any personal data, except as described in the Bug Report Feature section below. The extension runs entirely in your browser and only communicates with two services for its core citation generation functionality.

Information We Collect

During normal citation use, we collect no personal information of any kind. If you voluntarily submit a bug report, the data described in the Bug Report Feature section below is collected. During normal citation use, specifically:

• No personal information is collected or stored during normal citation use
• No browsing history is accessed or recorded
• No analytics or usage tracking is performed
• No user account is required or created
• No cookies are set

Services Used

The extension communicates with the following services to provide citation functionality:

Google Patents CDN (patentimages.storage.googleapis.com)
The extension fetches patent PDF files from Google's public CDN to extract text position data. Only the patent document identifier is included in the request URL. No user data is transmitted.

Cloudflare KV Cache (pct.tonyrowles.com)
The extension fetches pre-computed patent text position maps from this first-party infrastructure operated by the developer. Only patent document identifiers are sent as part of URL requests. No user data is transmitted. No data is logged or retained beyond the cached position maps. The pct.tonyrowles.com domain is wholly owned and operated by the extension developer — it is not a third-party service.

chrome.storage.sync
Three user preferences are synced across Chrome sessions using Chrome's built-in sync storage:

• Trigger mode (floating button or text selection)
• Display mode (tooltip or side panel)
• Patent number prefix toggle (whether to prepend the patent number to citations)

This data is managed entirely by Chrome's infrastructure and is not accessible to the developer. It never leaves Chrome's own sync system.

Citation Webapp (cite.tonyrowles.com)

In addition to the browser extension, the same citation lookup is available as a standalone web page at cite.tonyrowles.com. The webapp provides the identical deterministic citation function without installing the extension. It introduces no new personal data collection:

• No account, login, cookies, analytics, or usage tracking — the same as the extension.
• Patent PDFs are proxied through the developer's first-party Worker (pct.tonyrowles.com) via a dedicated /webapp/pdf route. The webapp never fetches patent documents directly from Google's CDN. Only the patent document identifier is included in the request.
• Parsed patent text position maps are cached in the same shared first-party Cloudflare KV used by the extension, keyed only by patent number. No user data is stored — the cache contains only derived patent-document position data.
• Your citation-format and patent-number-prefix preferences are stored locally in your browser (localStorage) and are never transmitted.
• The webapp authenticates to the Worker using an Origin header check only — no token and no Authorization header is sent in any request.

The citation computation (PDF text extraction and passage matching) runs entirely in your browser. As with the extension, no AI or remote inference is involved — the result is a deterministic column:line lookup.

Data Sharing

We do not sell or trade your data. Reports submitted through the voluntary Bug Report Feature are stored on Cloudflare (KV) and a notification containing report fields is delivered via Discord. Cloudflare and Discord act as service providers (processors) for that feature; data is not sold or shared with any other party. Outside the Bug Report Feature, no data is collected that could be shared.

Permissions Used

The extension requests the following Chrome permissions, used solely for citation generation:

• declarativeContent — Activates the extension icon only on Google Patents pages, not on all sites
• offscreen — Creates a hidden document to perform PDF text extraction without blocking the browser UI
• activeTab — Reads the current patent page URL to identify which patent is open
• storage — Saves your trigger mode, display mode, and prefix preferences via chrome.storage.sync
• contextMenus — Adds a right-click menu option for generating citations from selected text
• clipboardWrite — Copies the generated citation text to your clipboard

Changes to This Policy

This privacy policy may be updated from time to time. The date at the top of this page reflects the date of the most recent revision. Continued use of the extension after any changes constitutes acceptance of the updated policy.

Contact

For questions about this privacy policy, please open an issue at https://github.com/tonyrowles/patent-cite-tool.

Bug Report Feature

The extension includes a voluntary bug report feature that allows you to submit a diagnostic report when you encounter a citation failure. Submission is entirely optional and user-initiated — a report is only sent when you explicitly click the Submit button. No data is collected during normal citation-only operation.

What is transmitted when you submit a bug report:

• Patent number — the US patent number of the page you were viewing
• Patent URL — the Google Patents URL of the patent page
• Selection text (optional) — the text you had highlighted, up to approximately 200 characters. You can remove this before submitting using the [Remove selection text] toggle. This field is not transmitted if you toggle it off.
• Returned citation — the citation result the extension produced (or a no-match flag)
• Confidence tier — the confidence level of the citation result (green / yellow / red)
• Extension version — the installed version of this extension
• Browser and OS — your browser name and operating system
• Selected-node XPath — the DOM path of the element containing your selected text
• Scroll and viewport position — scroll offset and viewport dimensions at the time of the report
• PDF parse status — whether the patent PDF was successfully parsed
• Trigger mode — your configured citation trigger mode (floating button, automatic, context menu, or silent)
• Recent error log — up to 20 recent internal extension errors or warnings, if any occurred
• Free-text note (optional) — any additional description you choose to provide

Data collection categories (Firefox Add-ons): On Mozilla Add-ons (AMO), the fields above are declared under Firefox's standard data-collection categories. All of this data is collected only when you choose to submit a bug report — nothing is collected during ordinary citation use:

• Website activity — the patent page context: patent number, patent URL, selected-node XPath, and scroll/viewport position.
• Technical and interaction data (optional) — extension diagnostics that help reproduce the failure: extension version, browser and operating system, trigger mode, PDF parse status, the returned citation and its confidence tier, the recent error log, and any free-text note you add. This category is declared as optional and is transmitted only as part of a report you submit.
• Website content (optional) — the text you had highlighted. You can exclude it from any individual report with the [Remove selection text] toggle, in which case it is not transmitted.

Because the entire bug report feature is user-initiated, all of this data is opt-in: it is transmitted only when you explicitly click Submit. The extension never collects technical, interaction, website-activity, or website-content data in the background or during normal citation generation. If you include a free-text note, please avoid entering any personal information, as the note is sent verbatim.

No IP address is stored in bug report records. Your IP address is used only for transient server-side rate limiting (a short-lived counter that resets automatically) and is never written to any stored report record.

Retention: Bug reports are retained for 90 days and then automatically deleted.

Purpose: Reports are used solely for maintainer triage — to identify and fix citation failures. They are not used for advertising, analytics, or any other purpose.

Destinations: Submitted reports are stored in Cloudflare KV (the BUG_REPORTS namespace on first-party infrastructure at pct.tonyrowles.com) and a notification is sent to a maintainer-only Discord channel. These are the only destinations. No data is shared with third parties.