A specialized skill for Openclaw Skills that identifies and removes robotic writing markers to make AI-generated content sound authentically human.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install unclecheng-reduce-ai-perception
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 unclecheng-reduce-ai-perception 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: AI Text De-patterning & Humanization?
The Humanizer skill for Openclaw Skills is a sophisticated writing editor designed to scrub artificial signatures from machine-generated text. Based on the linguistic standards maintained by Wikipedia WikiProject AI Cleanup, this tool goes beyond simple synonym replacement. It addresses the structural, tonal, and grammatical markers that typically reveal AI authorship—such as over-emphasized significance, robotic 'rule of three' lists, and sterile promotional language.
By integrating the Humanizer into the Openclaw Skills ecosystem, developers and content creators can ensure their output resonates with authenticity. The skill doesn't just clean up 'slop'; it actively injects personality, varies sentence rhythm, and acknowledges complexity to ensure the final prose has a human pulse. This is an essential utility for anyone using LLMs who needs their final product to be indistinguishable from high-quality human writing.
Humanizer: AI Text De-patterning & Humanization Use Cases
- Reducing AI detection scores for professional articles and academic drafts.
- Transforming sterile technical documentation into engaging, developer-friendly content within Openclaw Skills.
- Humanizing chatbot responses to make them feel more empathetic and less servile.
- Adapting AI-generated marketing copy into a brand-specific, opinionated voice.
- Cleaning up formulaic 'Challenges and Future Prospects' sections from generated reports.
How Humanizer: AI Text De-patterning & Humanization Works
- The skill performs a deep scan of the input text to identify structural markers like excessive em dashes and title-cased headers.
- It cross-references the content against a database of high-frequency AI vocabulary such as 'pivotal,' 'testament,' and 'underscore.'
- The engine evaluates the tone to identify 'sycophantic' behaviors or overly neutral 'press release' styles.
- Using specialized prompts, it rewrites problematic sections by varying sentence lengths and adding first-person perspectives.
- It removes 'weasel words' and vague attributions, replacing them with specific, concrete details.
- The final output is presented alongside a summary of changes to ensure the human user maintains full editorial control.
Humanizer: AI Text De-patterning & Humanization Setup
To deploy this skill within your Openclaw Skills environment, ensure your agent has file system access for reading and editing documents.
# Add the humanizer skill to your local repository
openclaw skill add humanizer
# Or manually copy the SKILL.md into your skills directory
cp humanizer.md ~/.openclaw/skills/
Humanizer: AI Text De-patterning & Humanization Data Schema & Taxonomy
The Humanizer skill organizes its processing data through a structured analysis of linguistic patterns:
| Data Component | Description |
|---|---|
| Source Text | The raw AI-generated input requiring humanization. |
| Pattern Map | A collection of identified AI-isms and robotic markers found in the text. |
| Personality Layer | Metadata defining the desired voice (e.g., casual, technical, opinionated). |
| Transformation Log | A Markdown-formatted list of specific changes made to the prose. |
| Refined Output | The final version optimized for Openclaw Skills delivery. |
Humanizer: AI Text De-patterning & Humanization Advanced Features
- Platform-Specific Optimization: Includes tailored logic for Kimi, Doubao, and DeepSeek to minimize detection rates.
- Soul Injection: Actively adds opinions and reactions to facts, breaking the sterile neutrality of standard LLMs.
- Copula Correction: Replaces elaborate 'serves as' or 'stands as' constructions with direct 'is/are' verbs for clarity.
- Structural Randomization: Breaks formulaic AI outlines into organic, non-linear narratives.
- Negative Parallelism Removal: Eliminates overused 'Not only... but also' structures common in Openclaw Skills outputs.
SKILL.md
Loading
Related Openclaw Skills
Steve Jobs Perspective
A high-fidelity mental model simulation of Steve Jobs designed to audit products, strategy, and design through the lens of technical and humanistic excellence.
Dandan Test Skill
A diagnostic utility for verifying the publishing lifecycle and CLI functionality of AI agent skills.
MAL Anime Tracker
A command-line tool for managing MyAnimeList watchlists and receiving automated notifications through Openclaw Skills.
Agent Runtime
A comprehensive runtime system for AI agents that manages tool lifecycles, permissions, and context optimization.
Human-Like Memory
A persistent memory system that enables AI agents to search, recall, and store long-term context, user preferences, and historical decisions.
Improvement Evaluator
A specialized tool for measuring real-world AI performance improvements using execution-based task suites instead of simple document structure checks.
Featured*
