Guide · Agent skill
How to use the Chinese Commit Conventions agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The chinese-commit-conventions skill is a written reference that an AI agent can load to help you write git commit messages in Chinese using the Conventional Commits style. It also includes reference configs for commitlint, husky, commitizen and conventional-changelog. This guide explains what the skill does, how to install it, how to call it, and what to check before you rely on it.
Chinese Commit ConventionsjnMetaCode/superpowers-zh · MIT Download the skill Quick start
- 1Download and unzip the skill. Download the ZIP from this page and unzip it so SKILL.md sits in ~/.claude/skills/ or in your project's .claude/skills/ folder.
- 2Or upload it in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the skill.
- 3Check your tools. Make sure Git, Node.js and npm are installed in the project where you plan to use the tooling.
- 4Type the command. Start your request with /chinese-commit-conventions so the skill loads. It will not run on its own.
- 5Describe your change. Give the agent the change in plain language, including whether it is a feature, fix, or breaking change.
- 6Review the output. Check the type, scope, subject length and footers against your team's rules before you commit.
- 7Apply tooling if needed. If you want commitlint or husky set up, ask the agent to apply the config and review the changes it makes.
What the skill makes the agent do
The skill gives the agent a reference guide for Chinese commit messages. It covers the type keywords (feat, fix, docs, style, refactor, perf, test, chore, ci, revert), with type kept in English and the scope and description written in Chinese. It also shows a full message layout with subject, body and footer, plus rules for breaking-change and issue-linking footers.
It also includes Node.js tooling templates: an npm install line and config for commitlint, a husky commit-msg hook, a lint-staged setup, a commitizen interactive commit option, and a conventional-changelog setup with a Chinese section mapping in .versionrc.js.
The skill is a reference. It does not inspect or change your repository on its own. If you want config files written or edited, you need to ask the agent to apply them.
Install the skill from the ZIP or the Claude apps
Option one: download the ZIP from this page and unzip it into your personal skills folder, ~/.claude/skills/, or into a project's .claude/skills/ folder if you want it limited to one repository. The skill's file is SKILL.md.
Option two: in the Claude apps, open Settings, go to Capabilities, then Skills, and upload the skill there.
- Personal: ~/.claude/skills/
- Project: .claude/skills/ inside the repository
- Claude apps: Settings → Capabilities → Skills
Check what you need before you start
The skill itself needs no scripts to run, but the setup steps it describes use Git, Node.js and npm. You need those installed in the project where you plan to add commitlint, husky, commitizen or conventional-changelog.
The skill is licensed under MIT.
- Git
- Node.js and npm (for the commitlint, husky, commitizen and conventional-changelog steps)
Trigger the skill with an explicit command
This skill only runs when you type /chinese-commit-conventions explicitly. It is set not to trigger automatically from context, so a normal request to write a commit message will not load it. Type the command in your request, then describe what you need in plain language.
/chinese-commit-conventionsExamples of requests to make after the command
Ask for specific output rather than a general topic. The skill's content is most useful when you give it a concrete change.
- /chinese-commit-conventions Write a commit message in Chinese for my change that adds phone-number login with SMS fallback.
- /chinese-commit-conventions My fix changes the order stock deduction to use a lock. Suggest a subject line that is an action phrase under 50 characters.
- /chinese-commit-conventions This change renames the user API response fields. Draft the BREAKING CHANGE footer and say how callers should migrate.
- /chinese-commit-conventions Show me the commitlint.config.js rules for enforcing these types in my Node.js project.
Know the safety notes and limits
The skill runs no scripts and uses no network access. Its guidance is text, so any commands it lists are for you or the agent to run deliberately in your project.
Check the output before committing. The reference includes team rules such as a subject under 50 characters, no full stop at the end, and BREAKING CHANGE for database, public API and config format changes. Those rules are the team's own conventions, so adjust them if your team differs.
- Runs scripts: no
- Uses network: no
- Review any config change before you commit it to the repository
Tips for getting good results
Keep each commit to one change, since the reference treats atomic commits as part of the checklist. Put the team's commitlint and husky config in the repository so everyone gets the same checks.
The reference notes that scope can be Chinese, but commitlint may need its scope-case check turned off for that. Agree on this as a team before rolling it out.
- Use Chinese scope names such as 用户模块 or 订单 if your team reads Chinese fluently.
- Put a mixed Chinese and English term in the subject with a space, for example 添加 Redis 缓存.
- Ask the agent to produce a pre-commit checklist from the reference when you review a change.
What people use it for
- Drafting consistent Chinese commit messages for a team that uses Conventional Commits.
- Writing BREAKING CHANGE footers that include migration steps for an API or config change.
- Setting up commitlint and husky in a Node.js project so non-conforming commits are blocked.
- Adding a commitizen prompt so developers can write commits interactively with npm run commit.
- Configuring conventional-changelog with Chinese section names for release notes.
Frequently asked questions
Is the chinese-commit-conventions skill free to use?
The skill is licensed under MIT, as stated in its SKILL.md. Check the license on the source page for any additional terms.
What do I need to use this skill?
The skill needs no scripts. The tooling steps it covers require Git, Node.js and npm in your project. You also need an agent environment that supports skills, such as Claude with the skill installed.
How do I install the chinese commit conventions skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder. You can also upload it in the Claude apps under Settings, Capabilities, Skills.
Why doesn't the skill trigger when I ask for a commit message?
The skill is designed to run only when you type /chinese-commit-conventions explicitly. It will not load from context, so include the command in your request.
Will the skill change my repository automatically?
No. It is a written reference and does not inspect or modify your repository. If you want config files created or changed, you must ask the agent to apply them.
How does it compare with using commitlint or commitizen directly?
The skill does not replace those tools. It gives the agent Chinese conventions and config templates for commitlint, husky, commitizen and conventional-changelog, which you still install and run in your project.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
This skill suits Chinese-speaking teams that already want Conventional Commits and need templates for commit messages, commitlint, husky and changelog setup. It is less useful if your team does not use Chinese or Conventional Commits, because its rules would clash with existing habits. If you want the agent to change your repository for you, pick a tool that acts on files directly and use this skill only as a reference.
Download the skill