ReadYouApp / ReadYouApp/ReadYou
Feature Request: Integrate OpenAI API for Article Summarization & Translation 功能建议:集成 OpenAI API 实现文章概括与翻译
Nobody has claimed this yet.
- Dominant language
- Kotlin
- Stars
- 7.5k
- Forks
- 330
- PR merge metrics
- No merged PRs in 30d
Description
Feature Request
Description
I would like to propose integrating the OpenAI API into Read You to provide the following AI-powered features:
- Article Summarization — Call the OpenAI API to generate concise summaries of lengthy articles
- Article Translation — Call the OpenAI API to translate articles into the user's preferred language
Motivation / Use Case
- Time efficiency: RSS feeds often contain lengthy articles; an AI-generated summary helps users quickly decide whether to read the full content
- Language barrier: Many high-quality RSS sources are in English or other languages; translation makes content accessible to non-native speakers
- Offline reading: Summaries can be pre-generated and cached for offline consumption
- Accessibility: Helps users with reading difficulties by providing condensed versions
Proposed Implementation
Summarization
- Add a "Summarize" button in the article reading view
- Support configurable summary length (short / medium / long)
- Option to auto-summarize all new articles in the background
- Use OpenAI GPT models (e.g.,
gpt-4o-miniorgpt-3.5-turbo) to generate summaries
Translation
- Add a "Translate" option in the article menu
- Support selecting target language from system locale or manual preference
- Option to show original text alongside translation (bilingual view)
- Use OpenAI GPT models for high-quality, context-aware translation
OpenAI API Integration Design
API Key Management
- Users enter their own OpenAI API Key in Settings → AI Services
- Support custom API endpoints (default
api.openai.com, compatible with proxies/reverse proxies) - API Key stored locally using Android Keystore encryption, never uploaded to any server
Model & Parameter Configuration
- Allow users to select models:
gpt-4o-mini(recommended, best cost-efficiency),gpt-4o,gpt-3.5-turbo - Configurable parameters: temperature, max_tokens, etc.
- "Reset to defaults" option available
Usage & Cost Control
- Display estimated token consumption / cost (based on article length)
- Support setting a monthly token limit to prevent unexpected overages
- Optional "Wi-Fi only" mode to save mobile data
Privacy & Security Considerations
- Bring-your-own-key: Read You does not provide a built-in API Key; users must register their own OpenAI account
- Direct requests: The app sends requests directly to the OpenAI API without routing through Read You intermediary servers
- Data minimization: Only article title and body content are sent; user subscription lists, reading history, and other metadata are never included
- Opt-in feature: AI features are entirely optional; if disabled, no related network requests are made
UI/UX Suggestions
- Follow Material You design language
- Display subtle loading animations and progress indicators during AI processing
- Allow users to completely disable AI features in Settings → Privacy
- Display summary results in a collapsible card at the top of the article
Alternatives Considered
- Manual summarization by users (too time-consuming)
- Using external browser translation (breaks the in-app reading experience)
- Third-party RSS services with AI features (requires cloud account, privacy concerns)
Additional Context
- Similar features exist in apps like Feedly (Leo AI) and Inoreader (AI summarization), but these are closed, cloud-based paid services
- Read You adopts a "bring your own API Key" model, remaining free and open-source while giving users full privacy control
- The project already integrates with multiple third-party APIs (Fever, FreshRSS, Inoreader, etc.); the OpenAI API could follow a similar Provider architecture
Thank you for considering this! I'm happy to help with testing or further discussion. ❤️
功能建议
描述
建议为 Read You 集成 OpenAI API,实现以下 AI 智能功能:
- 文章概括 — 调用 OpenAI API 为长文生成简洁摘要
- 文章翻译 — 调用 OpenAI API 将文章翻译为用户指定语言
动机 / 使用场景
- 节省时间:RSS 订阅源中的文章往往篇幅较长,AI 摘要可帮助用户快速判断是否值得精读
- 打破语言壁垒:许多优质 RSS 源为英文或其他语言,翻译功能让非母语用户也能轻松阅读
- 离线阅读:摘要可预先生成并缓存,方便无网络时浏览
- 无障碍阅读:为阅读障碍用户提供精简版内容,提升可及性
建议的实现方案
文章概括
- 在阅读界面添加"概括"按钮
- 支持摘要长度调节(简短 / 适中 / 详细)
- 可选后台自动概括所有新文章
- 使用 OpenAI GPT 模型(如
gpt-4o-mini或gpt-3.5-turbo)生成摘要
文章翻译
- 在文章菜单中添加"翻译"选项
- 支持按系统语言自动识别或手动选择目标语言
- 提供原文与译文对照显示(双语视图)
- 使用 OpenAI GPT 模型进行高质量翻译,支持上下文理解
OpenAI API 集成设计
API Key 管理
- 用户在 设置 → AI 服务 中自行填写 OpenAI API Key
- 支持选择 API 端点(默认
api.openai.com,兼容自定义代理/反向代理) - API Key 本地加密存储(Android Keystore),不上传至任何服务器
模型与参数配置
- 允许用户选择模型:
gpt-4o-mini(推荐,性价比高)、gpt-4o、gpt-3.5-turbo - 可配置 temperature、max_tokens 等参数
- 提供"一键恢复默认"选项
流量与费用控制
- 显示预估 Token 消耗 / 费用(基于文章字数估算)
- 支持设置月度 Token 上限,防止意外超额
- 可选"仅 Wi-Fi 下使用"以节省流量
隐私与安全考量
- 用户自有 Key:Read You 不提供内置 API Key,用户需自行注册 OpenAI 账户
- 请求直发:应用直接向 OpenAI API 发送请求,不经过 Read You 的中间服务器
- 数据最小化:仅发送文章标题和正文内容,不含用户订阅列表、阅读记录等元数据
- 可选功能:AI 功能完全可选,不开启则无任何网络请求
UI/UX 建议
- 遵循 Material You 设计语言
- AI 处理过程中显示轻量加载动画与进度提示
- 允许用户在 设置 → 隐私 中完全关闭 AI 功能
- 摘要结果以卡片形式嵌入文章顶部,可折叠收起
替代方案考量
- 用户手动概括(耗时费力)
- 使用外部浏览器翻译(破坏应用内阅读体验)
- 第三方 RSS 服务的 AI 功能(需绑定云端账户,存在隐私顾虑)
补充说明
- 类似功能已在 Feedly(Leo AI)、Inoreader(AI 概括)等应用中出现,但均为封闭云端付费服务
- Read You 采用"用户自带 API Key"模式,既保持开源免费,又赋予用户完全的隐私控制权
- 项目已集成 Fever、FreshRSS、Inoreader 等多种第三方 API,OpenAI API 可遵循类似的 Provider 架构扩展
感谢考虑此建议!如有需要,我很乐意参与测试或进一步讨论。❤️
Written using Kimi 2.6
使用Kimi 2.6编写
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No implementation files or tests are named. Start by reviewing the existing Fever, FreshRSS, and Inoreader Provider architecture, then trace the article reading view and Settings screens. A complete implementation would need an agreed scope for summarization, translation, key storage, configuration, caching, privacy, and usage controls.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- android, kotlin
- Domain
- ai, mobile
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 28/100