CyberDiviner / 赛博算命 Privacy Policy

Last updated: June 3, 2026

English

Summary: CyberDiviner is an entertainment and cultural divination app. Face-reading camera frames are processed for the active session, original face photos are not saved by default, and original face photos are not uploaded by default. Cloud AI features are optional and user-controlled.

1. Scope and entertainment-only use

CyberDiviner / 赛博算命 (“CyberDiviner”, “we”, “our”, or “the app”) is an entertainment and cultural divination app. This Privacy Policy explains how the app handles data when you use features such as oracle readings, Liuyao, tarot, face-reading entertainment, voice input, on-device AI, and optional cloud AI.

CyberDiviner is provided for entertainment, self-reflection, and cultural experience only. It does not provide medical, legal, financial, psychological, diagnostic, or other professional advice. Do not rely on app outputs for important personal, health, legal, financial, or safety decisions.

2. Data processed by the app

2.1 Text input and divination content

You may enter questions, prompts, or other text for divination-style readings. Depending on your settings, this text may be processed locally on your device, processed by the app’s on-device Gemma 3 1B LiteRT / CyberDiviner Gemma 3 1B model if you have downloaded and enabled the offline model, or sent to a cloud AI provider only if you configure and enable cloud AI features.

Cloud AI is optional and user-controlled. If you choose to use a third-party provider, your text may be processed according to that provider’s own terms and privacy policy.

2.2 Camera and face-reading entertainment

CyberDiviner may request camera permission for face-reading entertainment features. Camera frames are used for on-device visual analysis during the active session.

CyberDiviner does not save original face photos by default and does not upload original face photos by default. Face-reading cloud enhancement is disabled by default. If you explicitly enable a cloud enhancement feature, the app may send a text-based, reduced description to the AI provider you selected, not the original photo.

CyberDiviner does not use face data to identify you, authenticate you, or determine your real-world identity.

2.3 Voice input

CyberDiviner may request microphone permission for speech-to-text input. The app requests Android speech recognition with an offline preference where available. Actual speech recognition behavior may depend on your device, operating system, installed speech services, and settings.

CyberDiviner does not store voice recordings.

2.4 On-device AI model files

CyberDiviner may download on-device AI model files, including Gemma 3 1B LiteRT and CyberDiviner Gemma 3 1B variants. These files are stored locally on your device to support offline inference. Model downloads may contact model hosting services such as Hugging Face, hf-mirror, or ModelScope depending on availability and your region or device settings.

CyberDiviner does not use IP geolocation services for model source selection.

2.5 API keys and cloud AI settings

If you configure a cloud AI provider, your API key and provider settings are stored locally in the app’s private data area on your device. CyberDiviner does not operate a server to collect your API key.

You are responsible for the third-party provider and endpoint you configure. HTTPS endpoints are recommended for custom providers.

3. Permissions

4. Data sharing

CyberDiviner does not sell your personal data. CyberDiviner may share data only in these cases:

  1. If you enable or configure a third-party cloud AI provider, the text you submit for that feature may be sent to that provider for app functionality.
  2. If your device speech recognition service processes voice input, speech data handling may depend on your Android system and speech service settings.
  3. Model downloads may contact third-party model hosting services to retrieve model files.

CyberDiviner does not share original face photos by default.

5. Data retention and deletion

Most app data is stored locally on your device. You can delete local data by clearing the app’s data in Android system settings or uninstalling the app. If the app provides in-app history deletion, you may also use that feature to delete stored readings.

Third-party cloud AI providers may have their own data retention policies when you choose to use them. Please review the privacy policy of the provider you configure.

6. Children

CyberDiviner is not directed to children under 13. If you are under the minimum age required in your country or region, please do not use the app without appropriate consent from a parent or guardian.

7. Security

CyberDiviner uses local app storage for settings and local data. Cloud AI requests should use HTTPS where supported by the selected provider. Custom endpoints are user-controlled; using insecure HTTP endpoints may expose your data on the network.

8. International availability

CyberDiviner may be available in multiple countries and regions. Optional third-party AI services and model hosting services may not be available in all regions.

9. Changes to this policy

We may update this Privacy Policy from time to time. The updated version will be posted at the same privacy policy URL with a revised “Last updated” date.

10. Contact

For privacy questions, please contact the developer through the developer contact email listed on Google Play for CyberDiviner, or open an issue at https://github.com/sinonchum/CyberDiviner/issues.

中文

摘要：赛博算命是一款娱乐与文化体验类应用。面相相机画面仅在功能使用期间处理，应用默认不保存原始面部照片，也默认不上传原始面部照片。云端 AI 功能为可选功能，由用户自行配置并主动启用。

1. 适用范围与娱乐用途声明

CyberDiviner / 赛博算命（以下简称“CyberDiviner”“本应用”或“我们”）是一款娱乐与文化占卜体验应用。本隐私政策说明本应用在你使用天机、六爻、塔罗、面相娱乐解读、语音输入、端侧 AI 和可选云端 AI 等功能时如何处理数据。

本应用仅用于娱乐、自我反思与文化体验，不提供医疗、法律、金融、心理、诊断或其他专业建议。请勿将应用输出作为重要个人、健康、法律、财务或安全决策依据。

2. 本应用处理的数据

2.1 文本输入与占卜内容

你可以输入问题、提示词或其他文本以获得占卜式解读。根据你的设置，这些文本可能在设备本地处理；在你下载并启用离线模型后，由本应用的端侧 Gemma 3 1B LiteRT / CyberDiviner Gemma 3 1B 模型处理；或者仅在你配置并启用云端 AI 功能时发送至你选择的云端 AI 服务商。

云端 AI 是可选且由用户控制的功能。如果你选择使用第三方服务商，你提交的文本可能会按照该服务商自己的条款和隐私政策进行处理。

2.2 相机与面相娱乐解读

本应用可能会为面相娱乐解读功能请求相机权限。相机画面用于当前会话中的设备端视觉分析。

本应用默认不保存原始面部照片，也默认不上传原始面部照片。面相云端增强功能默认关闭。若你明确启用云端增强功能，应用可能会向你选择的 AI 服务商发送文本化、降维后的描述，而不是原始照片。

本应用不会使用面部数据识别你的身份、认证你的身份或判断你的真实身份。

2.3 语音输入

本应用可能会为语音转文字输入请求麦克风权限。应用会在可用时请求 Android 离线语音识别。实际语音识别行为可能取决于你的设备、操作系统、已安装的语音服务和系统设置。

本应用不保存语音录音。

2.4 端侧 AI 模型文件

本应用可能会下载端侧 AI 模型文件，包括 Gemma 3 1B LiteRT 和 CyberDiviner Gemma 3 1B 相关模型。这些文件会存储在你的设备本地，用于支持离线推理。模型下载可能会访问 Hugging Face、hf-mirror 或 ModelScope 等模型托管服务，具体取决于可用性以及你的地区或设备设置。

本应用不使用 IP 地理位置服务来选择模型来源。

2.5 API Key 与云端 AI 设置

如果你配置云端 AI 服务商，你的 API Key 和服务商设置会存储在本应用的设备本地私有数据区域中。CyberDiviner 不运营用于收集你的 API Key 的服务器。

你需要自行负责所选择的第三方服务商和接口地址。对于自定义服务商，建议使用 HTTPS 接口。

3. 权限

4. 数据共享

CyberDiviner 不出售你的个人数据。本应用仅可能在以下情况下共享数据：

  1. 如果你启用或配置第三方云端 AI 服务商，你为该功能提交的文本可能会为了实现应用功能而发送至该服务商。
  2. 如果你的设备语音识别服务处理语音输入，语音数据的处理方式可能取决于 Android 系统和语音服务设置。
  3. 模型下载可能会访问第三方模型托管服务以获取模型文件。

本应用默认不共享原始面部照片。

5. 数据保留与删除

大多数应用数据存储在你的设备本地。你可以通过 Android 系统设置清除应用数据或卸载应用来删除本地数据。如果应用提供应用内历史记录删除功能，你也可以使用该功能删除已保存的解读记录。

当你选择使用第三方云端 AI 服务商时，该服务商可能有自己的数据保留政策。请查看你所配置服务商的隐私政策。

6. 儿童

本应用并非面向 13 岁以下儿童。如果你未达到所在国家或地区要求的最低年龄，请在父母或监护人适当同意下使用本应用。

7. 安全

本应用使用设备本地应用存储保存设置和本地数据。云端 AI 请求应在所选服务商支持的情况下使用 HTTPS。自定义接口由用户控制；使用不安全的 HTTP 接口可能会使你的数据在网络传输中暴露。

8. 国际可用性

本应用可能在多个国家和地区提供。可选第三方 AI 服务和模型托管服务不一定在所有地区可用。

9. 本政策的变更

我们可能会不时更新本隐私政策。更新后的版本会发布在同一隐私政策 URL，并注明新的“最后更新”日期。

10. 联系方式

如有隐私相关问题，请通过 Google Play 中 CyberDiviner 开发者联系信息所列的邮箱联系开发者，或在 https://github.com/sinonchum/CyberDiviner/issues 提交 issue。