A professional text-refining skill that removes machine-like patterns and restores natural human rhythm to Chinese writing.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install zh-humanizer
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install zh-humanizer 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 humanizer-zh?
humanizer-zh is a specialized editorial skill within the Openclaw Skills ecosystem designed to strip away the statistical average feel of AI-generated content. Its primary purpose is to make Chinese text sound naturally written by a real person rather than a machine. Unlike simple paraphrasing tools, it focuses on increasing information density and removing specific linguistic artifacts like pseudo-intellectual buzzwords, hollow template structures, and conversational fillers common in LLM outputs. By using this skill, developers and writers can ensure their content is concrete, direct, and credible while maintaining the original facts and intent.
humanizer-zh Use Cases
- Refining AI-generated blog posts and articles to sound more authentic and engaging for human readers.
- Stripping corporate buzzwords and hollow phrases from generated reports to make them more actionable.
- Cleaning up technical documentation by removing conversational AI remnants and excessive bolding.
- Optimizing marketing copy by replacing vague ad-speak with objective, descriptive language.
- Converting rigid AI-generated list structures into natural, flowing narrative paragraphs.
How humanizer-zh Works
- The skill performs a deep scan of the input text to identify specific AI traces based on 21 distinct identification rules.
- It flags issues such as semantic inflation, fake authority citations, and the excessive use of high-frequency AI buzzwords.
- The workflow removes hollow template structures like the predictable Challenge-Opportunity-Future cycle.
- It simplifies sentence structures and optimizes linguistic rhythm to ensure the text is easy to read aloud.
- The skill produces a humanized version, identifies any remaining traces, and provides a final polished revision with optional explanation notes.
humanizer-zh Setup
To integrate this capability into your environment, add the skill via your Openclaw Skills manager:
openclaw add humanizer-zh
No additional API configuration is required beyond the standard LLM setup as the skill uses internal logic rules to guide the editing process.
humanizer-zh Data Schema & Taxonomy
The skill organizes its output into a structured multi-step response to ensure transparency in the humanization process:
| Component | Description |
|---|---|
| Rewritten Text | The initial draft after removing machine patterns. |
| Remaining AI Traces | A summary of any persistent AI-style markers that require manual review. |
| Final Revision | The optimized, human-rhythm compliant version of the content. |
| Modification Notes | Optional context on why specific structural changes were made. |
humanizer-zh Advanced Features
- Comprehensive 21-point identification library targeting everything from pseudo-depth verbs to overused dashes.
- Automatic removal of chatbot-specific leftovers like 'I hope this helps' or 'Great question'.
- Intelligent synonym management that avoids the unnatural word-cycling typical of machine writing.
- Concept-stacking header correction to transform vague titles into value-driven natural language.
- Support for multi-tool execution including Read, Write, and Edit for direct file manipulation within Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
fxtwitter
A lightweight utility to fetch comprehensive tweet data from X/Twitter URLs without requiring authentication or browser automation.
Nano Banana Pro Image Generation & Editing
A high-performance AI tool for generating and modifying images up to 4K resolution using the Gemini 3 Pro Image API.
OpenClaw Logfire Plugin
A production-grade observability plugin providing OTEL GenAI traces and token metrics for Openclaw Skills via Pydantic Logfire.
YouTube Watcher
A specialized skill to fetch, read, and analyze YouTube video transcripts for summarization and information extraction.
Crypto Trader
An automated cryptocurrency trading engine supporting multiple strategies, AI-driven market analysis, and multi-exchange integration.
Football Value Bets
A professional-grade football analysis skill that generates data-backed bet slips using statistical modeling and value analysis.
Featured*
