A sophisticated text naturalization engine that identifies 24 AI writing patterns and rewrites content to sound authentically human using Openclaw Skills.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install peng-ai-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 peng-ai-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?
The Humanizer skill is a specialized tool designed to bridge the gap between mechanical LLM output and natural human prose. By leveraging a comprehensive database of 24 distinct AI patterns and over 500 flagged vocabulary terms, it enables developers to polish agent-generated content within the Openclaw Skills ecosystem. Unlike simple synonym swappers, this skill analyzes the statistical properties of text—such as burstiness and sentence length variation—to ensure the final output mimics the organic rhythm of human writing.
Built for developers and technical content creators, this tool serves as a critical layer in any AI writing pipeline. It effectively strips away significance inflation, promotional language, and chatbot artifacts that often trigger AI detection systems. By integrating this into your Openclaw Skills workflow, you ensure that every piece of content maintains a specific personality and professional tone, making it indistinguishable from human-authored text.
Humanizer Use Cases
- Refining AI-generated blog posts and articles to remove repetitive patterns found in Openclaw Skills outputs.
- Auditing professional emails and documentation to eliminate sycophantic tones or generic chatbot conclusions.
- Scoring existing marketing copy for AI-like statistical signals using the built-in analysis CLI.
- Automatically humanizing technical documentation by replacing inflated vocabulary with concrete, actionable language.
How Humanizer Works
- The skill initiates a deep scan of the input markdown or text file against 24 specific AI markers like significance inflation and copula avoidance.
- It cross-references the content with three tiers of AI vocabulary, identifying over 500 giveaway terms often used by models in Openclaw Skills.
- A statistical analysis engine calculates the text's burstiness and type-token ratio to detect the metronomic uniformity typical of machines.
- Based on the analysis, it provides a priority-based list of suggestions to improve readability and natural flow.
- The humanize engine rewrites problematic sections, preserving the original meaning while adding personality and varying sentence structure.
Humanizer Setup
To deploy the Humanizer skill within your environment, follow these installation steps. This skill is designed for seamless integration with Openclaw Skills compatible agents:
# Install dependencies and the CLI tool
npm install humanizer-skill
# Run a full analysis report on a local file
node src/cli.js analyze -f draft.md
# Score text for AI-patterns (0-100 scale)
echo "Your content here" | node src/cli.js score
Humanizer Data Schema & Taxonomy
The Humanizer skill structures its analysis data to help users track content improvements across the Openclaw Skills framework:
| Attribute | Description | Data Type |
|---|---|---|
| AI_Score | A probability rating (0-100) of how likely the text is AI-generated | Float |
| Pattern_Matches | List of specific markers triggered (e.g., 'Em dash overuse') | Array |
| Burstiness | A measure of sentence length variation and rhythm | Float |
| Vocab_Density | Analysis of Tier 1 vs Tier 2 vocabulary usage | Object |
| Auto_Fixes | A history of changes made during the humanization process | Log |
Humanizer Advanced Features
- Comprehensive 24-pattern detection system covering content, language, style, and communication fillers.
- Statistical auditing tools that measure Type-token ratio and sentence uniformity for objective quality control.
- Multi-tier vocabulary filtering that targets 'dead giveaway' terms frequently found in Openclaw Skills agents.
- Support for an 'Always-on' mode that integrates core humanization rules directly into system prompts for consistent output quality.
- Automated CLI hooks for programmatic scoring and report generation in CI/CD pipelines.
SKILL.md
Loading
Related Openclaw Skills
Youku (优酷) Integration
An AI-driven skill for navigating Youku vast library of classic and exclusive video content.
Project Workflow Scheduler
A sophisticated orchestration framework that transforms complex multi-step projects into a supervised sequence of safe, isolated, and bounded cron work blocks.
Data Analysis Litiao
A sophisticated analytical framework for AI agents to transform raw data into statistically sound, decision-oriented insights.
Google Scholar Search
A powerful search utility for retrieving academic paper metadata, citation counts, and direct PDF links using the Semantic Scholar API.
Product Hunt Launch
A comprehensive strategic framework for planning, preparing, and executing a high-impact launch on Product Hunt to reach the Top 5.
Body Clock
Body Clock is a circadian rhythm and biological clock optimization tool designed to align your daily activities with your natural internal rhythms.
Featured*
