Word Craft — Privacy Policy

Effective date: 2026-05-10

1. Data we process

• User-triggered text content: when you use selection translation/lookup, sentence analysis, subtitle translation, or reading/chat features, the Extension reads the selected text, the current subtitle sentence, and the questions/prompts you enter in order to generate results.
• Website content and identifiers: to locate subtitles and integrate with the player/page state, the Extension may read the current page URL, video/episode identifiers (e.g., a YouTube v parameter or a Netflix watch id), subtitle track metadata, and similar page context.
• Local preferences and learning data: such as panel preferences, saved vocabulary/favorites, and feature toggles stored locally in chrome.storage.local.

The Extension does not require account registration or login and does not intentionally collect direct identifiers such as your name, email address, or phone number.

2. Third-party services and network requests

Some features rely on third-party services. When you use features that require online services, or when the Extension downloads required resources (for example, optional dictionaries on first use/after installation), the Extension may send network requests to the following domains:

• Google Translate / TTS (GTX): translate.googleapis.com
  • Purpose: selection translation, batch subtitle translation, TTS for exports.
  • Data types: text to translate (selection/subtitles) and language parameters.
• Microsoft Translator (Edge endpoints): edge.microsoft.com, api-edge.cognitive.microsofttranslator.com
  • Purpose: token retrieval and translation when you choose Microsoft as the engine.
  • Data types: text to translate (selection/subtitles) and language parameters.
• OpenSubtitles: rest.opensubtitles.org, dl.opensubtitles.org
  • Purpose: remote subtitle search and download when you use that feature.
  • Data types: search keywords, language parameters, and downloaded subtitle text.
• TED resources: ted.com, *.ted.com, *.tedcdn.com
  • Purpose: fetch page/subtitle resources on TED to power subtitle/learning features.
  • Note: requests to TED resources may include cookies per the browser’s same-origin behavior.
• GitHub Raw (public resources): raw.githubusercontent.com
  • Purpose: download public dictionaries/resources (for example ECDICT data) when enabled/needed and not yet cached.
  • Data types: download requests only; no upload of your learning data to GitHub.
• Supported site (if enabled/used): v.ikanbot.com
  • Purpose: provide subtitle/learning features on that site if you use it.
  • Data types: standard page requests and in-page content needed for functionality.

In addition, the Extension can use Chrome’s on-device Gemini Nano (Prompt API) for certain analysis/generation features. This capability runs locally on your device. However, if you also use online translation or subtitle download features, the above third-party requests may still occur (depending on the features you trigger).

2.1 Datadog (anonymous product and stability analytics)

To understand feature usage and improve stability, the Extension may send anonymous, high-level analytics to Datadog, Inc., with processing in the United States (US3) region, while you use relevant features. Whether this is enabled is decided by the published Extension build you installed; if it is not enabled, no data is sent to Datadog.

• Purpose: product and reliability improvements; not for personalized advertising or cross-site profiling of you as a person.
• Typical contents (high level): for example Extension version, high-level site or scene category, a stable identifier for the current video (not your full browsing URL or full address-bar contents), coarse outcomes such as success/failure and duration ranges, install/update reason where applicable. This analytics is not used to record your screen or capture full-page images.
• Not intended to include: name, email, or account identifiers; full prompts or full model replies; full subtitle transcripts or full-page text; full favorites content; full page URLs (including query strings).
• Anonymous grouping: a random anonymous identifier may be stored on your device to group events from one installation; it is not your Google/Netflix login. Clearing Extension data or uninstalling stops use of that identifier. How the browser stores Extension data is an implementation detail and is not described here.
• Datadog: processing is subject to Datadog’s published privacy and security information.

3. Storage and retention

• Local storage: preferences and learning data are stored locally via chrome.storage.local.
• Retention: data is kept until you clear Extension data or uninstall the Extension.

4. How to delete your data

You can use the Extension’s “factory reset / clear all extension data” functionality to remove locally stored data (including chrome.storage.local data and offline caches). You can also uninstall the Extension from Chrome to remove its data.

5. How we use data

We process data only to provide and improve the Extension’s core functionality, including subtitles, translation/lookup, learning exports, and saving/restoring user preferences and favorites; and, when the anonymous analytics described in section 2.1 is enabled, for anonymous feature analytics and stability analysis.

6. What we do not do

• We do not sell your data.
• We do not use your data for personalized advertising.
• We do not perform cross-site tracking (no ad tracking SDKs, no identity profiling).

7. Security notes

Network requests are sent by your browser and protected by its security mechanisms. Please note that when you use online translation or subtitle download features, the text you submit is processed by the corresponding third-party services and is subject to their own policies.

8. Changes to this policy

If the Extension’s data handling changes, we will update this Privacy Policy and, when required, update Chrome Web Store disclosures accordingly.

9. Contact

If you have questions about privacy, contact: familymr.fan@gmail.com