Chatgpt-Makes-Money: A Chinese-Language List of 66 ChatGPT Side-Income Ideas

What Chatgpt-Makes-Money actually is

The repository is a single README.md file. There is no source directory, no package manifest, no build script and no test folder in the top-level listing. The description calls it a list of ChatGPT side-income techniques, and the README opens with an editor's note explaining that the author and ChatGPT jointly produced 66 tips. The body then walks through numbered items: paid question relay, answering questions inside a paid community, deploying ChatGPT into WeChat or Feishu for a client, applying for API keys on someone else's behalf, building small apps, reselling hardware, running SMS-verification services, paid newsletters, paid groups, selling Plus accounts, mirror sites, self-published books, courses, fairy-tale picture books, prompt libraries, directory sites, mini-programs, prompt engineering, copy editing, reselling document packs, niche consulting, investing, personal finance, automated customer service, and more. The audience is Chinese-speaking readers who already use ChatGPT and want ideas for turning it into income. It is not a library you import and not a service you call.

The mechanism is a numbered list, not a data flow

There is no architecture to describe because there is no program. Each numbered section follows the same shape: a short label, one or two sentences of explanation, and a screenshot hosted on user-images.githubusercontent.com. Item 1, for example, describes selling answers on Xianyu at roughly 0.5 to 1 yuan per answer, with wholesale as an option. Item 13 describes mirror sites that wrap the API and charge a monthly fee, and the README explicitly calls that model "挂羊头卖狗肉" and says it is not recommended. Item 12 notes that Plus accounts are unstable for ordinary users and that the price is being speculated upward. The only link out of the README that points at another repository is the Claude Code anti-ban guide, and the top of the file links to a separate PAC repository for airport recommendations. Everything else is prose and images. If you were expecting a script that automates any of these ideas, the repository does not contain one.

Installing it: there is nothing to install

The README gives no install command, no dependency list, no supported runtime and no release artifact. Because the repository contains only README.md, the only way to "use" it is to read it. Clone it if you want a local copy of the text and the screenshots:

git clone https://github.com/xiaoming2028/Chatgpt-Makes-Money.git
cd Chatgpt-Makes-Money
ls

The `ls` output should be just `README.md`. From there, open the file in any Markdown viewer, or read it on GitHub, and work through the numbered items. There is no configuration step, no environment variable, no port and no first run. If you want to follow the ideas, the work happens outside the repository: signing up for the services mentioned, writing the content, or building the small tools the README alludes to. The repository does not provide those tools.

Where the list stops being useful

Several items describe activities that depend on a specific platform's tolerance at a specific moment. Item 11 warns that paid groups are easy to get shut down. Item 12 says Plus accounts are unstable and that Plus is not very useful for individuals. Item 13 is the author's own example of a model he refuses to recommend. None of these warnings come with a date, a policy link or a fallback. A reader in 2026 has no way to tell from the file alone whether the 2023-era screenshots still reflect how Xianyu, WeChat or the account market behave. The second limitation is structural: the README says the tips were co-written with ChatGPT, and the writing has the flat, enumerating quality you would expect from that process. Item 14 admits the resulting books are "比较泛泛" and not very readable. That is an honest note about the output, but it also describes the list itself. Third, there is no licence file in the repository, so the default copyright position applies and you cannot assume you may republish the text or the screenshots.

Compared with an actual ChatGPT tool repository

A project such as a ChatGPT client, a prompt-management library or a bot framework ships code you can run, an install path, a licence and a changelog. Chatgpt-Makes-Money ships none of those. The comparison that matters is with a curated list such as Awesome ChatGPT or with a prompt-engineering guide: those at least maintain a licence, accept pull requests against a stated contribution format, and separate the maintained entries from the dead ones. Here, the only file is the README, and the only maintenance signal is the push date. The README does invite stars and forks and suggests bookmarking the page, but a fork gives you a copy of the text, not a maintained branch. If your goal is to build something, a code repository is the right place to look. If your goal is to survey how one Chinese-language author framed the monetisation question in early 2023, this file is a compact sample of that conversation, including its self-aware caveats.

Maintenance, licence and what a fork actually copies

The repository is not archived, and the last push was on 2026-05-18. That is roughly four months before today, so the file has been touched recently, but a push to a README does not tell you which items were updated or whether any were removed. There are no releases, so there is no version to pin and no changelog to read. There is no licence file, which means the repository is not open source in the sense most engineers mean: you have the text under default copyright, and reusing it in a product or a course carries risk that a licence would have removed. The README's own instruction is to star the project and use GitHub's fork function to back it up and build on it, but a fork of a README-only repository is a copy of a document. It will not receive the author's future edits unless you pull them, and the author has given no schedule for those edits.

Who this is for, and what to check before relying on it

This is worth reading if you read Chinese, you are curious how the side-income conversation around ChatGPT was framed in that market, and you want a list of starting points rather than a tool. It is the wrong choice if you need code, an API, a licence, a maintained dependency or anything you can put in a build pipeline. It is also the wrong choice if you need current platform policy: the README does not cite any. Before acting on any single item, open the README at the current commit, confirm the item is still there, and treat the screenshot as a historical artifact rather than a current price or policy. Items 11, 12 and 13 already carry the author's own warnings, and those warnings are the most useful part of the file.

Editorial conclusion

Adopt this as a reading list if you read Chinese and want a broad survey of how people in that market talk about ChatGPT monetisation; skip it if you need code, an API, a licence you can rely on, or anything you can install. It is not a software project: the repository contains only README.md, and the README itself says the list was written by the author together with ChatGPT. Before citing it anywhere, open the README and check that the idea you plan to use is still described there, because the file is the whole project and it changes only when the author pushes.

Frequently asked questions

Official sources

1. Issues
2. Project website
3. README
4. xiaoming2028/Chatgpt-Makes-Money on GitHub