OrigRead-Desktop
Modern desktop RSS reader for Windows, macOS & Linux, featuring side-by-side reading, verifiable AI citation, Web Search, and MCP tool integration.
Install / Use
claude mcp add ZGMFX01A -- npx -y github:ZGMFX01A/OrigRead-DesktopIf the server publishes to npm under a different name, use that package instead — check the repo README.
MCP Server
Model Context Protocol server
Quality Score
Category
Development & EngineeringSupported Platforms
Our assessment of OrigRead-Desktop
OrigRead-Desktop scores 83/100 on our quality scale, 3016th of 4,639 Development & Engineering skills we index.
Its MCP Server is 12 KB long, well organised into 13 sections with 2 code examples: a thorough specification that gives an agent plenty to work with.
It has 10 GitHub stars, so there is little community track record yet; judge it on its content.
Maintenance, license and trust
- The repository was last updated 4 days ago, so OrigRead-Desktop is actively maintained.
- It is released under AGPL-3.0, a copyleft license: you can use it, but modified versions you distribute must carry the same license.
- Its trust signals score 97/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
OrigRead-Desktop compared with similar skills
All 4 of these similar skills score higher than OrigRead-Desktop; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| OrigRead-Desktop (this skill)by ZGMFX01A | 83 | 10 | 4d ago | MCP Server |
| taste-skillby Leonxlnx | 100 | 108.9k | today | CLAUDE.md |
| Agent-Reachby Panniantong | 100 | 108.1k | 2d ago | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 78.7k | today | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 69.1k | today | CLAUDE.md |
Frequently asked questions
- How do I install OrigRead-Desktop?
- Run
claude mcp add ZGMFX01A -- npx -y github:ZGMFX01A/OrigRead-Desktop. The install tabs above show the steps for each supported agent. - Which AI agents does OrigRead-Desktop work with?
- It is written for Claude Code and Claude Desktop, as a MCP Server file. Other agents that read the same format can often use it too.
- Is OrigRead-Desktop safe to use?
- It is AGPL-3.0-licensed and scores 97/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is OrigRead-Desktop still maintained?
- The repository was last updated 4 days ago, so OrigRead-Desktop is actively maintained.
Skill content
View source on GitHubInterface languages
OrigRead Desktop includes interface translations for Spanish, Brazilian Portuguese, German, French, and Japanese, alongside English and Simplified Chinese.
To use one of the five additional languages, choose Follow system in the language setting and use the corresponding system language. The manual language list currently offers only English and Simplified Chinese.
Interface language is separate from article translation. The README and standalone user guides are available in English and Simplified Chinese.
A place for the things you want to read
Your favorite blogs, the news you follow, that occasional column worth waiting for—all in one place. OrigRead brings your chosen sources into a timeline you can read at your own pace.
On the desktop, the article and AI can sit side by side. Read the original while asking questions, comparing reports, and checking citations. Hide the list to settle into a long piece, adjust the type and reading width, or copy an article into your notes. The original stays within reach, from the first headline to the next question.
<p align="center"> <img src="assets/readme/screenshots/en-US/ai-citations.png" width="1000" alt="An article beside its AI analysis, with cited evidence highlighted in the original text" /> <br /><sub>Article and AI together · Read a conclusion, check its source</sub> </p>Citation: see the evidence behind an answer
An AI answer can leave you with another question: did the author really say that? What was the context? Do these two reports rely on the same evidence? Citation connects answers to the original passages, making verification part of reading.
Follow a citation to the author's words. Click an article reference to locate and highlight the relevant text. If the evidence comes from another attached article, OrigRead can open it while keeping the discussion available. With the article and answer in the same window, you can check the context and keep asking questions.
For example, attach two reports about the same event and ask, “Where do their accounts differ?” Follow the references to see whether the disagreement comes from facts, perspective, or wording. AI helps organize the evidence; citations let you judge it for yourself.
Saved answers retain the sources used at the time. Changing attachments later does not replace an old answer's evidence. Search and tool results also retain their sources. If an article changes or a reference can no longer be located, inspect its source information. A citation does not guarantee that AI has interpreted the text correctly.
See the Citation chapter for a walkthrough.
Follow sites beyond RSS
A favorite site without a subscribe button does not always need another daily browser visit. OrigRead can turn regularly updated website content into a source you can follow. Paste a home or section URL to look for RSS / Atom and matching RSSHub routes. When there is no ready-made feed, OrigRead can look for article lists in web pages and public JSON/API data. This includes WordPress article APIs and data embedded in some Next.js and Nuxt pages.
You do not need to choose a parsing method upfront. OrigRead checks article counts, titles, links, and dates to rank the candidates, then lets you choose the section you actually want. Browser rendering offers another way to try pages whose articles appear only after scripts run.
For sites that need special handling, parsing rules tell OrigRead where to find articles. Import or export rules, or ask AI to help create one, then inspect the articles it actually finds before saving. Routine discovery and parsing need no AI setup. Reliable updates still depend on the site's access conditions and structure; a redesign may require a rule update.
Browse the built-in source directory for something new, or import OPML to bring your subscriptions. See the user guide for adding sources and handling parsing problems.
<table> <tr><th width="50%">Find the section you want</th><th width="50%">Make reading comfortable</th></tr> <tr> <td align="center"><img src="assets/readme/screenshots/en-US/source-discovery.png" width="480" alt="Source discovery: inspect an available subscription candidate" /></td> <td align="center"><img src="assets/readme/screenshots/en-US/settings.png" width="480" alt="Reading settings: font, background, line height, and reading width" /></td> </tr> </table>From finding an article to understanding it
Give long articles room
When a feed supplies only an excerpt, try fetching full text. Open the original website inside the app for comments, charts, or interactive content. Adjust fonts, size, background, and reading width, or import a local font you enjoy reading.
Place AI on either side of the article and drag the edge to resize it. Keyboard controls cover common actions such as moving between articles, starring, searching, and entering focus reading.
Read it, understand it, keep it
Read translated text or a bilingual view using Microsoft Translator, DeepL, Google Cloud, a DeepLX / DLX-compatible service, or an AI model. Switch to TTS when you would rather listen.
Share an article as Markdown on the clipboard, optionally including its body and any translation or summary currently open. The original URL stays attached. Paste it into your notes so the source is still there when you return to it.
Let AI follow your reading
Start a long article with a summary, ask about the current article, or select a passage to discuss. Attach related articles to compare perspectives. Use web search when you need background or recent developments.
OrigRead supports OpenAI-compatible cloud providers, self-hosted services, and local model services. Quick Messages save recurring questions, Skills preserve reusable methods, and Custom Instructions hold response preferences. Remote and local MCP services can supply additional tools, with confirmation before tool execution.
Configure these as they become useful. Everyday subscriptions, extraction, and reading work without AI. See the AI, search, and tools guide for advanced setup.
Download and get started
Choose a package for your system from GitHub Releases:
| System | Package |
| --- | --- |
| Windows 11 · x64 | .exe installer |
| macOS 13+ · Apple Silicon | .dmg |
| Linux · x64 | .AppImage; Ubuntu / Debian can also use .deb |
Keep the default Local account, add a source or import OPML, and start reading. The app can check for updates. See installation and updates if you need help installing.
For phones and tablets, visit OrigRead Android. Android and desktop install and update independently. The migration guide explains what you can move between them.
Your subscriptions, in your hands
A Local account stores data on your computer and supports Website, JSON/API, and RSSHub sources. If you already use FreshRSS, a Google Reader-compatible service, or a Fever-compatible service, connect that account to sync the subscriptions and reading states it supports.
Routine parsing, full-text extraction, and filtering run locally. AI and cloud translation send relevant content to the service you configure. Configuration backups carry subscriptions, rules, and settings. Credentials are excluded by default and can be included in a password-encrypted export. Configuration backups exclude article bodies, read and starred history, and summary and translation caches.
Feedback and discussion
Found a problem or have an idea? Open an issue. For parsing problems, include the URL, app version, and steps to reproduce. The project currently does not accept pull requests; use issues for feature suggestions, translation corrections, and documentation feedback too.
<details> <summary>Build from source</summary>Requirements: Node.js 24+ and npm 11+.
npm ci
npm run typecheck
npm test
npm run build
Package for the target platform:
npm run package:win
npm run package:mac
npm run package:linux
See package.json and electron-builder.yml for build scripts and package settings.
For a GitHub release, explicitly push a version tag in the existing vX.Y.Z format. The Build Desktop workflow runs only for v* tag pushes; ordinary branch pushes and pull requests do not start it, and there is no manual dispatch entry. It applies the tag version, runs the Windows, macOS, and Linux checks, builds the installers, and publishes the release after all three platform jobs succeed.
For checks without a release, run Verify Desktop manually from GitHub Actions. It installs dependencies and runs type checks, unit tests, the app build, and the complete Electron runtime suite on the same three platforms, without packaging installers or publishing a release. Ordinary branch pushes and pull requests do not start either workflow. Failed jobs in both workflows upload available test-results diagnostics as platform-specific artifacts retained for seven days.
Project relationship and license
OrigRead Desktop and OrigRead Android share a product direction, with independent repositories and releases. Thank you to everyone who helps with feedback, translations, and code.
Desktop is distributed under the GNU Affero General Public License v3.0 only (AGPL-3.0-only). See LICENSE.
Star history
<a href="https://www.star-history.com/?repos=ZGMFX01A%2FOrigRead-Desktop&type=timeline&logscale=&legend=top-left"> <picture> <source media="(prefers-color-scheme: dark)" srcset="https://api.star-history.com/chart?repos=ZGMFX01A/OrigRead-Desktop&type=timeline&theme=dark&logscale&legend=top-left&sealed_token=9yvZ…[redacted]Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
108.9kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Agent-Reach
108.1kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
78.7kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ai-job-search
69.1kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
