RealChinese for Openclaw

RealChinese is an open-source Chinese AI text humanization engine that detects 29 distinct AI footprints and rewrites content to match 8 major Chinese social media platforms.

thebuddha5566
v1.1.2
Jun 6, 2026
0
586
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install realchinese

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install realchinese using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is RealChinese?

RealChinese is a highly efficient, production-ready Chinese AI text humanization engine designed to bridge the gap between AI-generated drafts and authentic human writing. Unlike basic text replacers, this executable engine actively analyzes text against 29 distinct AI fingerprints—such as high-frequency AI buzzwords, rigid transition templates, and overly uniform sentence structures. It provides a programmatic, highly-reliable approach to polish your digital copy for native Chinese-speaking audiences.

As a prominent utility within the Openclaw Skills registry, RealChinese empowers creators, developers, and marketers to automatically adapt their text for 8 major Chinese platforms. By integrating this engine, teams can easily run automated pipelines that scan drafts and output high-converting, platform-optimized copy that reads naturally and retains high user engagement.

RealChinese Use Cases

  • Platform-Specific Optimization: Automatically rewrite standard LLM outputs to fit the unique voice of major Chinese networks such as Xiaohongshu, Bilibili, or Weibo.
  • AI Detection Auditing: Programmatically scan documents and marketing copy to detect high-frequency AI-generated patterns before publication.
  • Automated Content Workflows: Streamline localized marketing operations by integrating humanization steps directly into your publishing pipelines.

How RealChinese Works

  1. Analyze Input Text: The engine processes raw Chinese text via CLI or the Python API to scan for AI-generated patterns.
  2. Run 29-Point Detection: The system evaluates linguistic features including sentence length standard deviation, 'de' particle density, and excessive transition words to output an AI fingerprint score.
  3. Apply Platform-Specific Style Rules: Select from 8 distinct platform profiles (e.g., emoji-heavy style for Xiaohongshu, structured logic for Zhihu, or fast-paced scripts for Douyin).
  4. Generate Humanized Copy: The engine outputs rewritten text with natural phrasing, irregular structures, and platform-compliant formatting to bypass detection and engage real readers.

RealChinese Setup

Prerequisites

Ensure you have Python installed. You can run the engine directly from your terminal.

Installation & CLI Usage

# Detect AI footprints
python scripts/cli.py detect '这是一个非常有效的方法,能够显著提升工作效率...'

# Rewrite to Xiaohongshu style
python scripts/cli.py rewrite '你的AI文本...' 小红书

Programmatic Integration

from scripts.detector import AIFingerprint
from scripts.rewriter import rewrite

text = '通过优化工作流程,我们可以显著提升团队效率...'
fp = AIFingerprint(text)
print(f'AI Fingerprint Score: {fp.score:.2f}')

result = rewrite(text, platform='小红书', topic='效率提升')
print(result)

RealChinese Data Schema & Taxonomy

Detection Metrics

Indicator ID Detection Category Description
1 High-Frequency AI Vocabulary Detects overused words like '显著', '提升', '优化'
2-3 Template Transitions & Patterns Flags transitional phrases like '此外', '然而', '综上所述'
4 AI-Specific Buzzwords Scans for corporate jargon like '赋能', '抓手', '闭环'
5-6 Internet & Colloquial Slang Evaluates absence of trending slang (e.g., 'YYDS', '显眼包')
7-15 Syntactic Structure Analyzes paragraph uniformity, sentence length variance, and excessive particle '的' density

Platform Rewriting Profiles

Platform Expected Tone Styling Rules Applied
WeChat (公众号) Professional & Engaging Hook headlines, rhythmic paragraph breaks, end-of-article Call-To-Action (CTA)
Xiaohongshu (小红书) Peer-to-Peer Chat Heavy emoji density, modal particles, localized SEO hashtags
Zhihu (知乎) Evidence-Based & Logical Data-backed arguments, structured bullet points, open discussion prompts
Douyin (抖音) Fast-Paced Script Hook in first 3 seconds, short impact sentences, suspense, interactive prompts
Weibo (微博) Casual & Interactive 140-character limit, trending topic hashtags, memes, commentary style

RealChinese Advanced Features

  • Multi-Platform Adaptation: Supports 8 distinct platform personality engines spanning from casual Bilibili slang to geeky Jike products.
  • Extensible Python API: Allows deep integration into developer automation stacks for real-time, programmatic Chinese text optimization.
  • Dual-Engine Execution: Seamlessly combines analytical fingerprint detection with stylistic rewriting models in a single executable pipeline.
  • Zero-Config Deployment: Runs instantly out of the box under an MIT license, making it easy to embed into high-performance open-source workflows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*