Privacy Policy

Overview

LearnByWatching ("the Extension") is a browser extension that helps you learn languages by analyzing subtitle text when you pause videos on YouTube and Nebula. This policy explains what data is collected, how it is used, and your rights.

Data We Collect

Subtitle Text

When you pause a video and trigger an analysis, the subtitle text visible at that moment is sent to the LLM endpoint you configure (by default, a local LLM running on your own machine). This text is not collected or stored by us.

Extension Settings

Your settings (LLM API URL, API key, model name, and language proficiency level) are stored locally using chrome.storage.sync. This data is synced by your browser across your own devices and is never transmitted to our servers.

Translation Cache (Optional)

If you choose to sign in with an account, anonymized analysis results may be stored in a shared cache (keyed by a hash of the subtitle text, model name, and proficiency level) to speed up future lookups. No personally identifiable information is included in the cache.

Data We Do Not Collect

• Browsing history or page content beyond subtitle text
• Video watch history or metadata
• Any data from websites other than YouTube and Nebula
• Device identifiers or IP addresses

Third-Party Services

The Extension communicates only with the LLM endpoint you configure. By default this is a local endpoint on your own machine (http://localhost:…). If you configure a remote endpoint (e.g. an OpenAI-compatible API), your subtitle text will be sent to that service and subject to their privacy policy.

Data Retention

Settings data persists until you uninstall the Extension or clear browser storage. Cached analysis results (if you are signed in) may be retained to improve performance for all users; you may request deletion by contacting us.

Contact

If you have questions about this policy, please open an issue on the project's GitHub repository or contact the developer directly.