• 不收集你的 YouTube 账号、观看历史或任何个人身份信息。视频 ID（videoId）仅用于拉取本视频的社区纠错内容，不会被作为个人观看记录留存。
• 查词和平假名注音完全在你的浏览器本地完成，不联网。
• 不将任何数据出售或共享给广告商。
• 不使用 Cookie 进行广告投放或跨站追踪。
• 不对用户数据进行画像分析或行为追踪。

当你主动点击「AI 语法拆解」时，插件会将所选日语句子发送到我们的后端代理（Cloudflare Worker）。代理调用 Cloudflare Workers AI（或我们后续接入的其他 AI 模型）生成中文解释后返回结果，不保存句子内容。

开启翻译功能时，当前视频的日语字幕会发送至代理，由第三方翻译服务（例如腾讯机器翻译等）转换为中文后返回，不保存字幕内容。

高频请求的词条解释在服务器端 KV 缓存，以加速响应并降低 AI 消耗——缓存内容仅为日语词汇及其解释文本，不含任何用户信息。

mimikatsu 包含一套社区众包系统，用于持续优化字幕纠错和词典内容。以下是该系统涉及的数据处理说明：

视频 ID（videoId）

当插件检测到含日语字幕的视频时，会自动向服务器发送当前视频的 videoId 与匿名 install ID，以拉取社区为本视频提交的字幕及读音纠错内容。videoId 作为索引键使用，不会与任何个人身份关联，也不会被保存为观看记录。

日文词语上报（复合词优化）

当你悬停或选中字幕中的日文词语时，所选词语的文本会连同匿名 install ID 上报到服务器，用于帮助识别用户常用的复合词和惯用表达，以优化词库。上下文句子不会被发送。

插件使用 chrome.storage.local 在你的浏览器本地保存：收藏的句子、生词本、每日 AI 使用次数、随机安装标识。这些数据仅存在你的设备上，不同步到任何服务器。mimikatsu 不会主动将这些数据同步到你的 Google 账号。卸载插件时，Chrome 会自动清除这些数据。

本网站不使用追踪 Cookie，不集成第三方广告或分析 SDK。域名与页面托管均使用 Cloudflare，其基础设施会记录标准访问日志（IP、时间戳、请求路径），详见 Cloudflare 隐私政策。

页面从 Google Fonts 加载字体，Google 可能记录字体请求，详见 Google 隐私政策。

服务器端不主动保留用户数据。Cloudflare Worker 日志遵循 Cloudflare 默认日志保留策略。服务器端 KV 缓存仅存储词条内容（日语词汇及其解释文本），不含用户标识。本地数据由你自行管理，卸载插件即清除。

本插件严格遵守 Chrome Web Store《用户数据政策》及 Limited Use 要求：

• 收集的用户数据仅用于为用户提供或改进本插件的核心功能。
• 不会将用户数据转让给第三方用于与插件功能无关的用途。
• 不会将用户数据用于任何形式的广告投放。
• 不会将用户数据用于与本插件功能无关的任何目的。

本插件面向一般成年学习者，不专门针对 13 岁以下儿童，不故意收集儿童个人信息。若发现有儿童数据被意外收集，请联系我们，我们将及时处理。

如你位于欧盟、英国或其他适用地区，你可能享有访问、更正、删除或限制处理个人数据等权利。如需行使相关权利，请通过下方联系方式与我们联系。

由于本插件收集的数据主要为匿名标识（install ID），不绑定真实姓名或账号，大部分数据通过卸载插件即可完全清除。

若本政策发生重大变更，我们将在网站首页公告，并更新页面顶部的「最后更新」日期及版本号。

隐私相关问题请发邮件至 support@mimikatsu.app，我们通常在 7 个工作日内回复。

© 2026 mimikatsu. All rights reserved.

• We do not collect your YouTube account, watch history, or any personally identifiable information. The current video's ID (videoId) is used solely to retrieve community corrections for that video and is not retained as a watch record.
• Word lookup and furigana generation happen entirely in your browser — no network requests.
• We do not sell or share any data with advertisers.
• We do not use cookies for advertising or cross-site tracking.
• We do not build behavioral profiles from user data.

When you click "AI grammar analysis," the extension sends the selected Japanese sentence to our backend proxy (Cloudflare Worker). The proxy calls Cloudflare Workers AI (or other AI models we may integrate in the future) and returns the result. Sentence content is not retained.

When translation is enabled, Japanese subtitle text is sent to our proxy, which calls a third-party translation provider (such as Tencent Machine Translation) and returns the Chinese translation. Subtitle content is not retained.

Frequently requested word explanations are cached server-side in Cloudflare KV to reduce latency and AI costs. Cached content is Japanese vocabulary and explanation text only — no user data.

mimikatsu includes a community crowdsourcing system for continuously improving subtitle corrections and dictionary content. Here is how data is handled:

Video ID (videoId)

When the extension detects a Japanese-subtitle video, it automatically sends the current video's YouTube ID (videoId) together with your anonymous install ID to our server, to retrieve community-submitted subtitle and pronunciation corrections for that video. The videoId is used as an index key and is not associated with personal identity or stored as a watch record.

Word reporting (compound vocabulary optimization)

When you hover over or select a Japanese word in the subtitles, the selected text is reported to our server together with your anonymous install ID, to help identify frequently used compounds and idioms for vocabulary library optimization. Context sentences are never sent.

The extension uses chrome.storage.local to store on your device: saved sentences, vocabulary list, daily AI usage count, and random install ID. This data never leaves your device. Mimikatsu does not synchronize this data to your Google account. Uninstalling the extension automatically clears all stored data.

This website does not use tracking cookies or third-party advertising/analytics SDKs. Both the domain and site hosting are on Cloudflare, whose infrastructure records standard access logs (IP, timestamp, request path) per its privacy policy. Fonts are loaded from Google Fonts; see Google's privacy policy.

We do not actively retain user data server-side. Cloudflare Worker logs follow Cloudflare's standard retention policy. Server-side KV cache stores only vocabulary content (Japanese words and their explanations), not user identifiers. Local data is under your control and is cleared upon uninstall.

This extension complies with the Chrome Web Store User Data Policy, including the Limited Use requirements:

• Data collected is used only to provide or improve the extension's core functionality.
• Data is not transferred to third parties for purposes unrelated to the extension.
• Data is not used for advertising of any kind.
• Data is not used for any purpose unrelated to the extension's core features.

This extension is intended for general adult learners. We do not knowingly collect personal information from children under 13. If we learn that such data has been collected inadvertently, we will promptly delete it upon request.

If you are located in the EEA, UK, or other applicable jurisdictions, you may have rights to access, correct, delete, or restrict the processing of your personal data. Please contact us to exercise these rights.

Because most data collected by this extension is tied to an anonymous install ID (not a real name or account), most data can be fully removed simply by uninstalling the extension.

If we make material changes, we will post a notice on the homepage and update the "Last updated" date and version number above.

For privacy-related questions, email support@mimikatsu.app. We typically respond within 7 business days.

© 2026 mimikatsu. All rights reserved.