Yunshu SkillsHub skill
Twenty-seven Chinese-language skills for development, thinking, learning, writing, visuals, branding and working with agents
- 789 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- `git-push` publishes commits, tags and releases through `gh`
- The skills work with a second agent and launch Codex CLI for image generation
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/yunshu-skillshubDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- `git-push` publishes commits, tags and releases through `gh`
- The skills work with a second agent and launch Codex CLI for image generation
This is third-party code. Review the repository files before installing.
What it does
Yunshu SkillsHub is a personal collection of 27 skills that the author built up through daily work with AI. The skills sit in seven folders: development, thinking, learning, writing, visualization, brand and agents. The README table says when to apply each one and what comes out. In development there are `issue-pool` for sorting ideas and bugs into `ISSUES.md`, `prd-test-writer` for requirements together with tests, `req-change-workflow` for changing a finished feature with impact analysis, `git-push` with a pre-release check and `dual-agent-collaboration`, where one agent writes and the other only reviews. Thinking is covered by `thinking-partner` and `multi-perspective-analysis`, writing by `writing-assistant` and `thought-mining`. Visualization skills draw illustrations, comics and presentations, and for agents there are `goal-setter` for stating a task with a completion criterion and `memory-init`, which creates `CLAUDE.md`, `MEMORY.md` and a `memory/` folder.
Who it is for. Chinese-speaking developers, writers and product managers who use Claude Code or Codex and want ready workflows in their own language.
Good fit when
- A pile of ideas, feedback and bugs has to become a workable task list
- Requirements of a finished feature change and before editing you need an impact analysis and a regression checklist
- An important change should go to a second agent for review while the first one writes code
- A new project needs agent memory and a `CLAUDE.md` file so context is not retold every time
- An article needs a topic, structure and draft, and for publishing a cover and illustrations in one style
Not a fit when
- The team does not read Chinese: names, tables and skill texts are in Chinese
- You need a narrow industry skill with verified data, for example for taxes or medicine
- The environment has no Codex CLI but you need illustrations, comics or logos: those skills hand drawing over to Codex
Example request
Use thinking-partner: my release plan is tangled, help me work out where it is stuck and suggest the next stepLimitations
Skill texts and the README are in Chinese; the English README is a separate file. The illustration, comic, presentation and logo skills draw through the built-in Codex image generation, so an installed and authorized Codex CLI with a subscription is needed. `dual-agent-collaboration` requires both Codex and Claude Code on the machine, and releasing in `git-push` requires a `gh` login. Presentations come out as 16:9 images, the skill does not create an editable PPTX file. The collection is personal with mixed topics, so install only the skills you need.
How to disable. Delete the folders of unneeded skills from `~/.claude/skills/` or `~/.agents/skills/`. To remove everything at once delete the directory you copied the repository into.
FAQ
How do I pick the needed skills out of 27?
The `npx skills add yunshu0909/yunshu_skillshub --list` command prints the full list without installing. Then name the skills you want in the `--skill` flag, and a quoted asterisk installs all of them at once.
How do I install a skill by hand without npx?
Clone the repository and copy the whole skill folder, for example `cp -R yunshu_skillshub/thinking/thinking-partner ~/.claude/skills/`. For Codex the destination is `~/.agents/skills/`. Do not trim the folder: the templates and scripts inside are part of the method.
What does dual-agent-collaboration do?
One agent makes changes and the second works read-only and checks the result. The loop repeats until both agree on the outcome. Both Codex and Claude Code must be installed on the machine.
Which files does memory-init create?
The skill prepares `CLAUDE.md`, `MEMORY.md` and a `memory/` directory for a new project. The agent then reads them instead of getting the project description in chat every time.
Does the set help with reading articles and studying?
Yes, the learning section has two skills. `article-study` turns an article, document or code into a course of lessons with quizzes and a review of mistakes, and `system-study` builds a knowledge map of an unfamiliar field with sources and disputed points.
Discussion
No questions or reviews yet.
Related
SkillsA skill set for notes and plans in Chinese: iMandalArt cards, BIRD knowledge addresses, a keyword graph and an eight-page A4 booklet
cafe3310's personal skill collection: spec-driven project work, reviews, reports, browser, 3D and video in one repository
A set of 13 SKILL.md skills: research, a product doc, decks, articles and audits for any compatible agent
Ten Chinese-language skills for Claude Code, Codex and OpenClaw: Vibe Coding from idea to code, topic learning, video downloads.
The author's personal skills for reading papers and books, dissecting concepts, writing and PNG cards; instructions in Chinese, installed via skills CLI
Seven skills for research, HTML explainers and source citation, installable all at once or one at a time through npx.
README in short
The README tells how the author built up the skills through real work and gives three installation routes. Then seven sections with tables: when to apply a skill and what it produces.