mswnlz/edu-knowlege: a GitHub index of Chinese K-12 study material

What mswnlz/edu-knowlege actually is, and who it is for

The repository description says it collects educational material from kindergarten through primary and secondary school, covering institutions including 学而思, 万维 and 猿辅导, and that it keeps growing. The README adds university material and graduation theses to that scope, and lists the same brands plus 作业帮. So the subject is Chinese-language courseware, not a teaching method and not a learning platform.

The audience follows from that. A parent or tutor looking for exercise sets, a teacher assembling supplementary reading, or a developer building a study-resource site who needs an initial list of sources. The repository is a catalogue, and the README links outward rather than hosting a study application. Anyone expecting a quiz engine, a spaced-repetition scheduler or a student progress tracker is looking at the wrong project, because none of those appear in the file layout.

The repository is also part of a family. The README links to sibling repositories for traditional Chinese medicine, general tools, cross-border commerce, self-media, AI material, general curriculum, film and television, books, and health. The same account publishes all of them. That tells you the maintenance pattern: the same person curates several link collections at once, and this one is the education branch.

How the catalogue is organised: dated Markdown files plus a publishing action

The top level of the repository holds README.md, a LICENSE.txt, a WARP.md, an action.yml, an index.js, a package.json, and a series of files named by year and month: 202505.md, 202506.md, 202507.md, 202508.md, 202509.md, 202510.md and 202607.md. The README links to most of those same files at the top, under a heading that is just the project name.

That naming is the whole architecture. Each dated file is a snapshot of what was collected in that month, and the README is the entry point that points at them. There is no database, no generated index and no search layer. If you want to know what was added in a given month, you open that month's file. If you want cross-month search, you bring your own tooling, because the repository does not provide it.

The JavaScript side is small. package.json names the package aiknowledge-action, version 1.0.0, with main set to index.js and a single dependency, @actions/core. That dependency is the standard library for writing GitHub Actions. Combined with action.yml and the .github/ directory, the picture is a workflow that runs index.js to publish or refresh the catalogue. The README does not describe what the action writes or where. That is the weakest documented part of the project, and anyone forking it should read action.yml and index.js directly rather than trusting the README.

Installing it and reading your first month of material

There is no package to install and no server to start. The README gives no installation section at all. What the README does give is the project homepage, https://doc.869hr.uk, which it describes as a gathering place for material, knowledge, film, documentaries, music, books and media, with the note that it is over 100T and still being organised. Treat that site as the published view and the repository as the source.

The practical first step is to clone the repository and open the newest dated file. The README links 202607.md first, so start there.

git clone https://github.com/mswnlz/edu-knowlege.git
cd edu-knowlege
ls 20*.md

You should see the dated Markdown files listed by the glob, including 202607.md. Open that file in a text editor or a Markdown viewer; the content is a catalogue, so expect links and labels rather than prose.

If you want to run the action locally to see what index.js does, the dependency list is one package, so the setup is short. Note that this mirrors what package.json declares; it is not an install command the README provides.

npm install
node index.js

@actions/core expects to run inside a GitHub Actions environment, so running index.js directly on a workstation may fail on missing action inputs. The repository does not document which inputs it needs. Read action.yml before you spend time on it.

The licence is permissive; the content links are the real risk

The repository ships LICENSE.txt, and the repository metadata identifies it as MIT. MIT covers the code and the Markdown files as they sit in the repository. It does not cover the material the links point to. Courseware from 学而思, 万维, 猿辅导 and 作业帮 is commercial material owned by those companies, and the repository is an index to it, not a rights holder. Nothing in the README claims permission from those publishers.

That distinction matters more here than in a typical open source project, because the value of the repository is entirely in the links. A permissive licence on the catalogue tells you that you may copy and modify the catalogue. It tells you nothing about whether you may redistribute what the catalogue points at. If you plan to build a product on top of this list, the licence question you need answered is not the MIT one.

This is not legal advice, and the repository does not address the question. It is a factual gap you should resolve before shipping anything derived from these links.

Where it breaks down: no manifest, no health check, no versioning

The repository has no releases. The recent releases list is empty, so there is no changelog, no tagged version and no way to pin a known-good state. Updates arrive as new dated Markdown files and edits to existing ones. If a link rots, nothing in the repository reports it. There is no link checker in the top-level layout, and the README does not mention one.

That is the main failure mode. A curated link list degrades silently. The catalogue keeps its shape while the targets disappear, and a reader has no signal about which entries are still live. For a personal bookmark collection this is tolerable. For anything automated, it means you must add your own reachability check, because the project does not ship one.

The second limitation is scope drift. The README describes education from kindergarten through university, and then describes the homepage as a collection that also includes film, documentaries, music, books, media, fitness and nutrition. The same README links to separate repositories for films, books, health and general tools, so the boundary between this repository and its siblings is not enforced by anything in the file layout. A reader looking strictly for K-12 material should expect to filter.

Third, the repository is JavaScript by primary language, but that label is misleading. The JavaScript is one action file with one dependency. The substance is Markdown. Do not pick this project expecting a codebase to contribute to.

Alternatives, and how the approach differs

The obvious alternative is a general search engine. The difference is curation versus coverage. A search engine will find more, including material this repository never lists, but it will not tell you which publisher a resource came from or group it by school stage. This repository's contribution is the grouping and the brand labels, and that is exactly what a search query does not give you.

A second alternative is the sibling repositories from the same account, such as mswnlz/curriculum and mswnlz/AIknowledge. Those are not competitors in the usual sense, they are the same curation method applied to different subjects. If your interest is course material outside the K-12 range, curriculum is the closer fit, and the README here links to it directly. Choosing between them is a question of subject, not of quality.

A third alternative is building your own index. Given that this repository is Markdown files plus a small action, that is a realistic option, and it gives you the link-health checking this project lacks. The trade-off is that you lose the accumulated months of collection, which is the part that takes time to reproduce. Copying the dated files and adding a checker is a legitimate way to use this project without depending on its maintenance.

Maintenance, upgrade cost and what the last push date tells you

The repository is not archived, and the last push was on 2026-07-27. There are no releases, so there is no upgrade path in the versioned sense. Upgrading means pulling the branch and reading whichever dated files are new since your last pull. Because the files are named by month, you can see at a glance which periods you have already processed, and a diff on a dated file shows what changed inside it.

The cost of following the project is therefore low in tooling and non-trivial in reading. Nothing will break when you pull, because there is nothing to compile beyond the single action dependency. What you spend is attention: each new month is a new file to review, and each existing file can change in place.

Two funding notes appear in the README. It states that compute is provided by ZMTO through an open source free VPS programme, and it also carries a DartNode badge for free VPS for open source. The README additionally asks for donations through Alipay and WeChat QR images stored in the repository. Read that as a signal about how the hosting is paid for, not as a statement about the project's roadmap.

Editorial conclusion

Adopt it if you are assembling Chinese-language K-12 or university study material and want a pre-sorted starting point: read README.md, then the dated files such as 202607.md, and check each link before you rely on it. Do not adopt it if you need an application, a searchable database, or a guarantee that any given file still resolves, because the repository contains catalogues and an action, not a content service. Verify first whether the links in the dated file you care about are still reachable, and whether the action.yml workflow in .github/ still runs against the current index.js.

Frequently asked questions

Official sources

1. Issues
2. License: MIT
3. mswnlz/edu-knowlege on GitHub
4. Project website
5. README