Privacy Policy · 隐私政策
This policy explains what Gleaner collects, why, and how it is stored.
本政策说明 Gleaner 收集哪些信息、用途，以及如何保存。
Contact / 联系： github.com/Spolora/gleaner-site/issues
English
Who we are
Gleaner is a browser extension by Spolora that helps you save English words, phrases, and sentences while you read, then review them in a side panel and export them to Anki.
What we collect
On your device (default)
When you save a selection, Gleaner stores on your computer:
- the selected text
- surrounding sentence context (when available)
- page title and page URL
- dictionary content you looked up (definitions, phonetics, examples, glosses, etymology when available)
- optional notes you write
- review metadata (for example created time)
- cached pronunciation audio for saved words
- downloaded native-gloss dictionary packs (Chinese / Japanese / Korean), when your browser language needs them
- optional local vocab-level test results
This local data stays in the browser until you clear it or uninstall Gleaner.
If you sign in
Optional Google sign-in syncs your cards and latest vocab-level summary to our cloud
database (Supabase), tied to your account (email / user id from Google). You can keep
using Gleaner without signing in; then that data stays local only.
Anonymous community map (optional)
To show a shared “learners on Earth” map on our website, the extension may send at most
once per day: a country code only (derived from your timezone, not GPS
and not your IP). We do not store a user id with that ping. Counts are
aggregated by country and day. You can turn this off in Settings → Community map. The map
never shows cities or individuals.
We do not sell your data. We do not use your cards for advertising.
Data sent over the network
To look up meanings and related content, Gleaner may send the selected word or phrase (and similar short queries) to third-party services:
| Purpose | Provider examples |
|---|---|
| English definitions / etymology | Wiktionary, Datamuse |
| Native word gloss (on device) | CC-CEDICT (MDBG), JMdict (GitHub), Kengdic (GitHub) |
| Phrase / sentence translation; pronunciation audio | Youdao |
| Optional mnemonic images | Wikimedia / Wikidata / Wikipedia |
| Account & sync | Google (sign-in), Supabase (cards & vocab level) |
| Anonymous country presence (optional) | Supabase (country code + day count only) |
We do not intentionally send your full browsing history. Lookups happen when you select text or open related features.
Permissions
Gleaner asks for access needed to read the current page selection, show the side panel, store settings, and (where available) complete sign-in. Host access is used so dictionary and sync requests can run from the extension.
Retention and deletion
- Local data: remove it by clearing extension storage or uninstalling Gleaner.
- Cloud cards / vocab profile: sign in and delete in the product, or contact us to request account-related deletion.
- Community map counts are anonymous aggregates and are not tied to an account.
- Third-party services keep data under their own policies.
Children
Gleaner is not directed at children under 13. Do not use it if you are under 13.
Changes
We may update this policy. The “Last updated” date above will change when we do. Continued use after an update means you accept the revised policy.
Contact
Questions about privacy: Open an issue on GitHub.
中文
我们是谁
Gleaner 是 Spolora 推出的浏览器扩展：你在阅读时划选英文单词、短语或句子并收藏，在侧边栏复习，并可导出到 Anki。
我们收集什么
默认仅保存在本机
当你收藏内容时，Gleaner 会在你的电脑/浏览器中保存：
- 划选的文本
- 上下文句子（如能提取）
- 页面标题与网址
- 查阅到的词典信息（释义、音标、例句、母语释义、词源等）
- 你填写的笔记
- 复习相关时间信息
- 已收藏单词的发音缓存
- 按浏览器语言下载的母语词典包（中文 / 日文 / 韩文，如需要）
- 可选的词汇水平测试结果（本机）
在你清空扩展数据或卸载之前，这些内容留在本机。
如果你选择登录
可选的 Google 登录会把卡片与最近一次词汇水平摘要同步到我们的云端数据库（Supabase），并与你的账号（来自 Google
的邮箱 / 用户 ID）关联。不登录也可以正常使用，这些内容只留在本机。
匿名社区地图（可选）
为在官网展示「地球上的学习者」地图，扩展最多每天发送一次：仅国家代码（由时区粗略推断，不是
GPS，也不使用你的 IP）。我们不会把用户 ID 与该心跳一起存储；按国家与日期聚合计数。可在设置中关闭「社区地图」。地图不会显示城市或个人。
我们不出售你的数据，也不把卡片用于广告。
会发到网络上的内容
为了查词和相关功能，扩展可能把你划选的词或短语（及类似短查询）发送给第三方服务：
| 用途 | 服务示例 |
|---|---|
| 英文释义 / 词源 | Wiktionary、Datamuse |
| 单词母语释义（本机） | CC-CEDICT（MDBG）、JMdict（GitHub）、Kengdic（GitHub） |
| 短语 / 句子翻译；发音音频 | 有道 |
| 可选助记图片 | Wikimedia / Wikidata / Wikipedia |
| 账号与同步 | Google（登录）、Supabase（卡片与词汇水平） |
| 匿名国家活跃（可选） | Supabase（仅国家码 + 日计数） |
我们不会故意上传你的完整浏览历史。请求发生在你划词或使用相关功能时。
权限说明
扩展需要读取当前页的选中文本、打开侧边栏、保存设置，并在支持的浏览器中完成登录。主机访问权限用于词典查询与同步请求。
保存与删除
- 本机数据：清空扩展存储，或卸载 Gleaner。
- 云端卡片 / 词汇水平：登录后在产品内删除，或联系我们处理账号相关删除。
- 社区地图计数为匿名聚合，不绑定账号。
- 第三方服务按其各自政策处理数据。
儿童
本产品不面向 13 岁以下儿童。未满 13 岁请勿使用。
变更
我们可能更新本政策，并以文首“更新日期”为准。更新后继续使用即视为接受修订内容。
联系我们
隐私相关问题：请在 GitHub Issues 留言。