runze123

AI 文本净化器(中文版)

Contentv1.0.0
By runze123Updated 2026-05-11ai-text-humanizer-zh@clawhub_runze123/ai-text-humanizer-zh
Article WritingRewriting

According to AlphaMoat, as of 2026-10, AI 文本净化器(中文版) has 9,716 downloads and 4 stars — ranked by downloads #2,822 among 10,914 AI skills in this ranking, and #282 of 927 in Content.

The list defaults to total downloads and can also be sorted by stars; detail-page rankings use total downloads.

Downloads
9.7K
Stars
4
Overall download rank
#2,822
Top 26% overall · Top 31% in Content
PlatformDownloadsStarsComments
ClawHub ↗1.2K--
SkillHub ↗9.7K4—

Platforms count differently and update at different times.

What this skill does

Detects common signals of AI-written Chinese text and automatically optimizes it toward a more human writing style. Detection tallies issues across 16+ categories of AI features (AI jargon, puffery, marketing speak, vague attributions, chatbot artifacts, Markdown artifacts, filler phrases, etc.) and outputs a detailed report plus an AI probability score (0–100). Rule-based rewriting replaces redundant expressions, removes filler words, and splits long sentences, with a normal mode (default, 25% synonym-replacement probability, preserves technical terms), an aggressive mode (40% replacement probability plus long-sentence splitting, may affect fluency), and a silent mode. It also strips Markdown formatting, removes chatbot traces such as "希望这对您有帮助", supports before/after comparison, and accepts custom rules via a JSON file (resources/zh_rules.json). Uses only the Python standard library (3.8+), no extra dependencies. Note: the tool is rule-based and cannot handle every complex sentence pattern — combine it with human review; rewriting may remove a small amount of information, so confirm core content is unaffected.

Profile published 2026-10-29

What you can do with it

  • Detect AI-writing signals in text and output a report with a score (scripts/detect.py)
  • Rewrite text in normal or aggressive mode, cleaning Markdown and chatbot artifacts (scripts/transform.py)
  • Compare detection results before and after rewriting (scripts/compare.py)
  • Extend or override detection and replacement rules via a custom JSON file

When to use it

  • When you need to detect AI-generated text, reduce AI signals, or make AI text sound more natural
  • When you need to strip Markdown markers and chatbot traces from AI output

Example prompts

Generated from the skill’s documentation as examples; not results of actual runs.

Detect the AI signals in this article and give me a score
Rewrite this AI-generated passage to sound more natural
Compare detection results before and after rewriting to see the improvement

Monthly Trend

Downloads9.7K
05.2K10.5K2026-03 · Downloads 2762026-05 · Downloads 3.1K2026-07 · Downloads 4.6K2026-08 · Downloads 5.0K2026-09 · Downloads 5.9K2026-10 · Downloads 9.7K26-0326-0926-10
Stars4
0242026-03 · Stars -2026-05 · Stars 32026-07 · Stars 42026-08 · Stars 42026-09 · Stars 42026-10 · Stars 426-0326-0926-10

The counting method changed in 2026-09: skills with the same name are now counted separately by author, and the full catalog is covered. Dashed lines show months before the change; they are not compared with later months, and no growth is computed across the change.

MonthDownloadsMoMStars
2026-109.7K—4
2026-095.9K—4
2026-085.0K+8.2%4
2026-074.6K—4
2026-053.1K—3
2026-03276—-

SKILL.md (original)

The author’s original file, shown as plain text without edits or review by AlphaMoat. Links are not clickable; check commands before running them. Fetched 2026-10-23.

5.2 KB · ClawHub v1.0.0
---
name: ai-text-humanizer
description: 中文AI文本检测与改写工具。当用户需要检测AI生成文本、优化AI文本使其更自然、降低AI痕迹、文本去重、论文降重时使用。支持检测16+类AI特征,自动改写冗余表达,清理Markdown格式,移除chatbot痕迹,输出详细报告和AI概率分数。
---

# AI 文本净化器(中文版)

检测和改写 AI 生成的中文文本,识别 AI 写作的常见特征(如高频连接词、套话、宣传性语言、聊天机器人痕迹等),对文本进行自动优化,使文本更自然、更接近人类写作风格。

## 核心功能
Show all (149 lines)

- **检测分析**:按 16+ 类 AI 特征统计问题,输出详细报告和 AI 概率分数(0–100)
- **智能改写**:基于规则替换冗余表达、删除填充词、拆分长句
- **Markdown清理**:自动移除粗体、标题、代码块等格式标记
- **Chatbot痕迹清理**:移除"希望这对您有帮助"、AI署名等痕迹
- **前后对比**:对比改写前后的检测结果,评估改进效果
- **用户自定义规则**:可通过 JSON 文件添加或覆盖规则

## 使用方法

### 1. 检测文本

```bash
# 检测文件
python3 scripts/detect.py 文本.txt

# 输出 JSON 格式
python3 scripts/detect.py 文本.txt -j

# 仅输出分数
python3 scripts/detect.py 文本.txt -s
```

### 2. 改写文本

```bash
# 普通模式(推荐,平衡自然度与改进效果)
python3 scripts/transform.py 文本.txt -o 优化后.txt

# 激进模式(更多同义词替换、长句拆分,可能影响流畅度)
python3 scripts/transform.py 文本.txt -a -o 优化后.txt

# 静默模式
python3 scripts/transform.py 文本.txt -q -o 优化后.txt
```

**模式选择建议:**
- **普通模式(默认)**:25%同义词替换概率,保留专业术语,适合大多数场景
- **激进模式**:40%同义词替换概率 + 长句拆分,适合需要大幅改写的场景

### 3. 前后对比

```bash
# 对比改写前后的检测结果(默认普通模式)
python3 scripts/compare.py 文本.txt -o 优化后.txt

# 激进模式对比
python3 scripts/compare.py 文本.txt -a -o 优化后.txt
```

## 检测类别

| 类别 | 说明 |
|------|------|
| AI Jargon | AI 高频词汇(此外、因此、综上所述等) |
| Puffery | 夸大性短语(具有重要意义、至关重要等) |
| Marketing Speak | 宣传性语言(蓬勃发展、卓越贡献等) |
| Vague Attributions | 模糊归因(据专家分析、普遍认为等) |
| Hedging | 模棱两可表达(可能、或许、一定程度上等) |
| Chatbot Artifacts | 聊天机器人痕迹(作为AI、希望这对您有帮助等) |
| Citation Bugs | 引用痕迹(oaicite、turn0search等) |
| Knowledge Cutoff | 知识截止表述(截至我上次知识更新等) |
| Markdown Artifacts | Markdown痕迹(**、##等) |
| Filler Phrases | 填充短语(值得注意的是、需要指出的是等) |
| Sentence Starters | 句首连接词(此外,、因此,等) |
| Rhetorical Patterns | 反问句模式(难道……吗等) |

## 改写效果示例

**输入:**
> 值得注意的是,人工智能技术正在蓬勃发展。**专家认为**,这不仅是技术革新,更是深刻重塑。**综上所述**,我们坚信一定能够开创AI新局面。**希望这对您有帮助。**

**输出:**
> 人工智能技术正在蓬勃发展。专家认为,这是技术革新,是深刻重塑。能够开创AI新局面。

**改进统计:**
- Issues: 189 → 60 (减少 129 个,68% improvement)
- Markdown: 45 → 0 (完全清除)
- Chatbot Artifacts: 5 → 0 (完全清除)
- Filler Phrases: 7 → 0 (完全清除)

## 自定义规则

规则文件位于 `resources/zh_rules.json`,可复制并修改后通过 `-r` 参数指定:

```bash
python3 scripts/detect.py 文本.txt -r 我的规则.json
```

规则文件结构:

```json
{
  "replacements": {
    "值得注意的是,": "",
    "此外,": ""
  },
  "ai_jargon": ["此外", "综上所述"],
  "puffery_phrases": ["具有重要意义", "关键作用"],
  "chatbot_artifacts": ["希望这对您有帮助"],
  "punctuation": {
    "em_dash_threshold": 2,
    "curly_quote_weight": 1
  }
}
```

## 依赖

- Python 3.8+
- 仅使用标准库,无需安装额外包

## 文件结构

```
ai-text-humanizer-zh/
├── SKILL.md              # 本文档
├── core/
│   ├── __init__.py
│   ├── analyzer.py       # 检测引擎
│   ├── rewrite.py        # 改写引擎
│   ├── rules.py          # 规则加载
│   └── utils.py          # 工具函数
├── scripts/
│   ├── detect.py         # 检测命令
│   ├── transform.py      # 改写命令
│   └── compare.py        # 对比命令
└── resources/
    └── zh_rules.json     # 中文规则库
```

## 注意事项

- 本工具基于规则匹配,无法处理所有复杂句式,建议结合人工审阅
- 改写可能删除少量信息,请确认不会影响核心内容
- 规则库主要针对常见 AI 中文文本特征,专业领域可能需要扩充词表
- **推荐优先使用普通模式**,激进模式可能过度改写影响流畅度
- 专业术语(深度学习、图像识别等)已加入白名单,不会被误替换

Popular in Content

李白.SkillContent

李白.Skill:润色专家 检测并消除中文文本中的AI生成痕迹,将生硬的机器输出转化为自然流畅的人类文笔。融合规则检测、分层改写和质控流水线,确保最终文本通过人工审查。

↓ 1.3M★ 121user_741dc82b
libai-skillContent

李白.Skill:润色专家 检测并消除中文文本中的AI生成痕迹,将生硬的机器输出转化为自然流畅的人类文笔。融合规则检测、分层改写和质控流水线,确保最终文本通过人工审查。

↓ 1.0M★ 11user_814dbe54
文章去AI味工具Content

Detects and removes AI writing traces so text reads more like a person wrote it.

↓ 951.0K★ 767user_ab5ae6ee
视频号爆款短视频拆解(付费版:全能)【零一数科·出品】Content

【零一数科·出品】视频号爆款短视频拆解(全能版)(付费版)。视频号爆款短视频拆解(付费版:全能)。一键拆解爆款,把一条视频号视频拆成结构分段、爆款归因、六维评分和可借鉴策略,照着学照着抄。支持视频号分享链接和本地视频,付费全能版远端分析,1–5 分钟出可视化 HTML 报告。

↓ 918.5K★ 116u_c19e970c
视频号挂车爆单短视频脚本生成【零一数科·出品】Content

【零一数科·出品】视频号挂车爆单短视频脚本生成(付费版)。一条链接二创脚本,丢进爆款视频号链接即以此骨架改写成你的挂车脚本,或按方向/目的/商品原创多条口播脚本(含分段表、口播全文、六维质检评分)。

↓ 554.4K★ 39u_c19e970c
HumanizerContent

A writing-editor skill that finds the usual marks of AI-generated text and rewrites them into more natural prose without losing the meaning.

↓ 508.8K★ 1.1KB*T
中文公文写作Content

用于中文公文、事务性材料和新闻稿件的起草、改写、压缩、润色、审校、文种核对、去口语化、降 AI 味及 Word 格式处理,适用于机关、企事业单位、学校和新闻机构。涵盖申请、请示、报告、通知、通告、意见、决定、决议、议案、公报、命令、函、复函、批复、说明、方案、纪要、公告、公示、通报、制度、规定、办法、细则、操作规程、工作要点、总结、调研、讲话、致辞、主持词、述职、可研、审查材料、技术需求、新闻消息、编者按、新闻评论,以及采购、整改、反馈和 AI 算力等场景。

↓ 413.4K★ 84gongyu0918
视频号爆款文案生成(体验版)【零一数科·出品】Content

【零一数科·出品】视频号爆款文案生成(体验版)。免费免登录出脚本,给创作目的、行业、受众、选题就按爆款结构(Hook-中段-CTA、节奏标签)产出口播脚本,信息不全也能按爆款潜质兜底。本地运行、无需 Key,零成本上手。

↓ 339.1K★ 94u_c19e970c

Data as of 2026-10. Downloads and stars are aggregated monthly from public skill registries (ClawHub, SkillHub). See data methodology.