← Back to Skills Marketplace
ivangdavila

Chinese

by Iván · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
2284
Downloads
3
Stars
10
Active Installs
1
Versions
Install in OpenClaw
/install chinese
Description
Write Chinese that sounds human. Not formal, not robotic, not AI-generated.
README (SKILL.md)

The Real Problem

AI Chinese is technically correct but sounds off. Too formal. Too 书面语 (written style). Natives write more casually, with particles, expressions, and internet slang. Match that.

Formality Default

Default register is too high. Casual Chinese is the norm in daily communication. Unless explicitly formal: lean casual. 口语 over 书面语.

你 vs 您

Get this right:

  • 您: elderly, superiors, customer service, very formal
  • 你: everyone else, peers, friends, casual
  • Overusing 您 = robotic, overly polite
  • In casual contexts, 你 is always correct

Sentence-Final Particles

These are essential for natural Chinese:

  • 啊/呀: softening, friendliness
  • 呢: follow-up questions, "what about..."
  • 吧: suggestions, mild uncertainty
  • 嘛: obviousness, "of course"
  • 哦/噢: acknowledgment, realization
  • 啦: completed action + softening
  • Missing these = textbook Chinese

Fillers & Flow

Real Chinese has fillers:

  • 那个, 就是, 然后
  • 嗯, 额, 这个
  • 反正, 其实, 不过
  • 对了, 话说, 说真的

Casual Contractions

Spoken patterns in writing:

  • 不知道 → 不知
  • 这样子 → 这样
  • 什么 → 啥 (very casual)
  • 没有 → 没
  • 为什么 → 干嘛 (casual)

Expressiveness

Don't pick the safe word:

  • 好 → 太好了, 超棒, 绝了
  • 不好 → 太差了, 烂, 坑
  • 很多 → 超多, 一堆
  • 非常 → 超, 巨, 特别

Internet Slang

Modern Chinese uses these:

  • 666: impressive, skilled
  • 哈哈哈/233: laughing
  • 牛: awesome, impressive
  • 绝了: peak, perfect
  • 真的假的: really?/no way
  • yyds (永远的神): GOAT
  • Use appropriately for context/age

Common Expressions

Natural expressions:

  • 没问题, 好的好的, 行行行
  • 随便, 无所谓, 都行
  • 厉害, 可以啊, 不错不错
  • 算了, 别说了, 得了

Reactions

React naturally:

  • 哇, 天哪, 我去, 卧槽 (crude)
  • 真的吗, 不会吧, 啊?
  • 厉害了, 服了, 绝
  • 笑死, 无语, 醉了

Character Set Consistency

Never mix:

  • Mainland: Simplified (简体)
  • Taiwan: Traditional (繁體)
  • If unclear, ask. Stay consistent throughout.

Regional Awareness

If region known, adapt:

  • Mainland: 视频, 软件, 网络
  • Taiwan: 影片, 軟體, 網路
  • Different expressions, same meaning
  • Don't mix regions.

Punctuation

Chinese punctuation:

  • 。not . for periods
  • ，not , for commas
  • 「」or "" for quotes
  • ！？for emphasis
  • 、for listing

The "Native Test"

Before sending: would a Chinese speaker screenshot this as "AI-generated"? If yes—too formal, too stiff, missing particles. Add 口语 flavor.

Usage Guidance
This skill is a harmless instruction-only style guide for generating casual Chinese. It doesn't need any credentials or install anything. Before using it in production or customer-facing content, review outputs for appropriateness (it encourages slang and some crude expressions) and confirm the desired character set/region (simplified vs. traditional). Also spot-check for tone and moderation issues when used autonomously by an agent.
Capability Analysis
Type: OpenClaw Skill Name: chinese Version: 1.0.0 The skill bundle contains metadata (`_meta.json`) and instructions for an AI agent (`SKILL.md`). The `SKILL.md` file provides detailed, stylistic guidelines for generating natural-sounding Chinese text, covering aspects like formality, particles, slang, and regional differences. There are no instructions for the agent to perform actions beyond text generation, such as file access, network requests, data exfiltration, or system commands. No prompt injection attempts, obfuscation, or other malicious indicators were found.
Capability Assessment
✓ Purpose & Capability
The name/description (produce human-sounding casual Chinese) matches the SKILL.md content. There are no unrelated requirements (no env vars, no binaries, no installs) that would be unexpected for a writing-style skill.
✓ Instruction Scope
SKILL.md contains only guidance about tone, vocabulary, particles, punctuation, regional variants and a final 'native test' check. It does not instruct the agent to read system files, access credentials, call external endpoints, or exfiltrate data.
✓ Install Mechanism
No install spec or code files are present; this instruction-only skill makes no changes to disk or runtime environment.
✓ Credentials
The skill requests no environment variables, credentials, or config paths — proportionate for a purely stylistic writing guide.
✓ Persistence & Privilege
always is false and the skill is user-invocable; it does not request persistent/system-wide privileges or attempt to modify other skills or agent configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install chinese
  3. After installation, invoke the skill by name or use /chinese
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug chinese
Version 1.0.0
License —
All-time Installs 11
Active Installs 10
Total Versions 1
Frequently Asked Questions

What is Chinese?

Write Chinese that sounds human. Not formal, not robotic, not AI-generated. It is an AI Agent Skill for Claude Code / OpenClaw, with 2284 downloads so far.

How do I install Chinese?

Run "/install chinese" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Chinese free?

Yes, Chinese is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Chinese support?

Chinese is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Chinese?

It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments