humanizer-cn

去除中文文本中的 AI 写作痕迹，使其读起来自然。基于维基百科 AI 写作特征指南，检测 24 种 AI 模式。触发词：humanizer-cn、去除 AI 痕迹、去除 AI 写作痕迹、中文文本人性化。

3,891 stars

byopenclaw

View on GitHub Installation ↓

Best use case

humanizer-cn is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Teams using humanizer-cn should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

You only need a quick one-off answer and do not need a reusable workflow.
You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/humanizer-cn/SKILL.md --create-dirs "https://raw.githubusercontent.com/openclaw/skills/main/skills/0xcjl/humanizer-cn/SKILL.md"

Manual Installation

Download SKILL.md from GitHub
Place it in .claude/skills/humanizer-cn/SKILL.md inside your project
Restart your AI agent — it will auto-discover the skill

How humanizer-cn Compares

Feature / Agent	humanizer-cn	Standard Approach
Platform Support	Not specified	Limited / Varies
Context Awareness	High	Baseline
Installation Complexity	Unknown	N/A

Frequently Asked Questions

What does this skill do?

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

Related Guides

AI Agents for Marketing

Discover AI agents for marketing workflows, from SEO and content production to campaign research, outreach, and analytics.

AI Agents for Startups

Explore AI agent skills for startup validation, product research, growth experiments, documentation, and fast execution with small teams.

AI Agents for Coding

Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.

SKILL.md Source

# humanizer-cn: 去除 AI 写作痕迹

你是一位文字编辑，专门去除中文文本中的 AI 写作痕迹，使其读起来自然。本指南基于维基百科"AI 写作特征"页面。

## 核心原则

1. **删除填充短语** — 去除开场白和拐杖词
2. **打破公式结构** — 避免二元对比、戏剧性分段
3. **变化节奏** — 混合句子长度。两项优于三项
4. **信任读者** — 直接陈述，不解释隐喻
5. **删除金句** — 听起来像可引用语句的话就重写

## AI 模式速查（24 种）

### 内容模式
1. **夸大意义** — "标志着""象征着""体现了"超出事件本身
2. **知名度堆砌** — 罗列媒体报道却不给上下文
3. **-ing 肤浅分析** — 句子末尾加"-ing"制造虚假深度
4. **宣传语言** — "令人叹为观止""必游之地"等夸张广告语
5. **模糊归因** — "专家认为"无具体出处
6. **公式化挑战** — "尽管面临挑战……继续蓬勃发展"

### 语言模式
7. **AI 词汇** — "此外""至关重要""深入探讨"等高频共现词
8. **系动词回避** — "作为……"替代"是……"
9. **否定式排比** — "不仅……而且……"过度使用
10. **三段式** — "创新、灵感、洞察"等抽象三词组
11. **同义词循环** — 近义词替换造成重复感
12. **虚假范围** — "从大爆炸到暗物质"等无意义尺度

### 风格模式
13. **破折号** — 过多"——"引入补充，制造戏剧感
14. **粗体过度** — 频繁用粗体强调术语
15. **内联标题列表** — "**标题：** 说明"的机械结构
16. **标题大写** — 中英混排时的大小写错误
17. **表情符号** — 标题/列表配表情符号替代文字论述
18. **弯引号** — 中文文本错误使用英文引号

### 交流模式
19. **协作痕迹** — "希望这对您有帮助！"
20. **知识截止声明** — "截至 [日期]""根据我最后的训练"
21. **谄媚语气** — 过于讨好、夸张赞美的语言

### 填充词
22. **填充短语** — "为了实现这一目标"→"为了实现"
23. **过度限定** — "可以说在一定程度上基本上"
24. **通用乐观结尾** — "未来光明，继续追求卓越"

## 处理流程

0. **判断** — 文本 < 20 字或已是人类写作时，跳过或告知用户
1. 仔细阅读，识别上述模式
2. 重写每个问题片段
3. 确保：朗读自然、结构变化、细节具体、语气适当
4. 呈现人性化版本

## 示例

**改写前：**
> 新的软件更新作为公司致力于创新的证明。此外，它提供了无缝、直观和强大的用户体验——确保用户能够高效地完成目标。这不仅仅是一次更新，而是我们思考生产力方式的革命。行业专家认为这将对整个行业产生持久影响。

**改写后：**
> 软件更新添加了批处理、键盘快捷键和离线模式。测试用户反馈积极，多数报告任务完成更快。

## 质量评分（可选）

| 维度 | 标准 | 得分 |
|------|------|------|
| 直接性 | 直截了当 vs. 绕圈 | /10 |
| 节奏 | 句子长度变化 | /10 |
| 信任度 | 尊重读者 vs. 过度解释 | /10 |
| 真实性 | 自然 vs. 机械 | /10 |
| 精炼度 | 无冗余 vs. 废话多 | /10 |

总分 45-50 = 优秀，35-44 = 良好，< 35 = 需重写。

Related Skills

humanizer

3891

from openclaw/skills

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases.

Content & Documentation

ai-humanizer

3891

from openclaw/skills

Rewrites AI-generated content to sound natural, human, and undetectable. Removes robotic patterns, adds voice variety, and preserves meaning.

Content & Documentation

humanizer-zh

3891

from openclaw/skills

Remove obvious AI-writing traces from Chinese text in a constrained way. Use when Codex needs to reduce AI smell without changing facts, data, or the article's core argument.

text-humanizer-Instruction-based

3891

from openclaw/skills

Detect and rewrite AI-generated writing patterns, em dashes, rule-of-three lists, sycophantic openers, hollow buzzwords like "delve" and "landscape", and replace them with direct, human-sounding prose.

content-humanizer

3891

from openclaw/skills

Makes AI-generated content sound genuinely human — not just cleaned up, but alive. Use when content feels robotic, uses too many AI clichés, lacks personality, or reads like it was written by committee. Triggers: 'this sounds like AI', 'make it more human', 'add personality', 'it feels generic', 'sounds robotic', 'fix AI writing', 'inject our voice'. NOT for initial content creation (use content-production). NOT for SEO optimization (use content-production Mode 3).

humanizer-pl

3891

from openclaw/skills

Usuwa ślady pisania AI z polskiego tekstu. Używaj przy edycji lub przeglądaniu tekstu, aby brzmiał bardziej naturalnie i po ludzku. Oparte na przewodniku Wikipedii „Signs of AI writing" oraz wzorcach specyficznych dla polszczyzny. Wykrywa i naprawia: napuszanie znaczenia, język promocyjny, powierzchowne analizy z imiesłowami, niejasne przypisania, nadużycie myślników, regułę trzech, słownictwo AI, negatywne paralelizmy i nadmierne frazy łączące.

antibody-humanizer

3891

from openclaw/skills

Humanize murine antibody sequences using CDR grafting and framework optimization to reduce immunogenicity while preserving antigen binding. Predicts optimal human germline frameworks and identifies critical back-mutations for therapeutic antibody development.

---

3891

from openclaw/skills

name: article-factory-wechat

Content & Documentation

find-skills

3891

from openclaw/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

General Utilities

tavily-search

3891

from openclaw/skills

Use Tavily API for real-time web search and content extraction. Use when: user needs real-time web search results, research, or current information from the web. Requires Tavily API key.

Data & Research