A writing editor skill designed to identify and eliminate the telltale signs of machine-generated text to restore natural human rhythm and voice.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install jz-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 jz-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 sophisticated editing engine developed to strip away the algorithmic fingerprints left by large language models. Built upon the comprehensive standards of Wikipedia's AI Cleanup project, it moves beyond simple synonym replacement to address structural and stylistic patterns that make writing feel sterile. By identifying inflated symbolism, promotional fluff, and repetitive sentence structures, this skill ensures your content maintains a professional yet authentic human quality.
In the context of Openclaw Skills, Humanizer functions as an essential layer for developers and writers who need to ensure their output resonates with readers. It doesn't just delete problematic phrases; it injects soul into the writing by introducing rhythm variance, acknowledging complexity, and adding subjective nuances that machines typically avoid. The result is text that is clear, direct, and indistinguishable from high-quality human craftsmanship.
Humanizer Use Cases
- Refining AI-generated technical articles to remove promotional language and significance inflation.
- Humanizing automated email templates or chatbot responses to improve user engagement.
- Cleaning up academic or wiki contributions to meet strict neutral-tone requirements.
- Transforming soulless, repetitive reports into punchy, varied prose with a distinct personality.
- Stripping out chatbot artifacts like knowledge-cutoff disclaimers and servile filler phrases.
How Humanizer Works
- The skill scans the input text to detect over 20 different AI-specific patterns, including superficial -ing endings and weasel words.
- It evaluates the sentence rhythm and structure to identify monotonous, machine-like repetition.
- The agent rewrites problematic sections, replacing vague attributions with specific details and removing overused AI vocabulary.
- It injects personality by varying sentence lengths and adding a sense of perspective where appropriate.
- The tool generates a humanized version of the text along with an optional summary of the specific AI-isms that were corrected.
Humanizer Setup
To integrate this skill into your workflow, ensure your agent has access to the standard file manipulation tools. Openclaw Skills usually require the following configuration:
# Ensure your agent directory includes the humanizer skill
mkdir -p skills/humanizer
cp humanizer.md skills/humanizer/SKILL.md
Once the skill is active, you can invoke it by asking your agent to humanize a specific file or text block. No additional API keys are required as the logic is built into the prompt-based pattern recognition engine.
Humanizer Data Schema & Taxonomy
The Humanizer skill organizes its analytical and creative output based on the following metadata taxonomy:
| Attribute | Description |
|---|---|
| Detected Patterns | A list of specific AI tells found in the source text (e.g., em dash overuse, Rule of Three). |
| Voice Profile | The detected intended tone, such as formal, technical, or casual. |
| Humanized Text | The final processed content with natural flow and rhythm. |
| Change Log | A detailed breakdown of the transformations applied to each problematic section. |
Humanizer Advanced Features
- Multi-lingual pattern detection, including specialized libraries for Chinese AI-taste vocabulary.
- Copula avoidance correction, replacing elaborate constructions like "serves as" with direct "is/are" statements.
- Automated removal of "sycophantic" artifacts and servile chatbot language.
- Structural rhythm variance algorithms that prevent repetitive sentence length issues.
- Deep integration with Wikipedia's WikiProject AI Cleanup standards for high-fidelity editorial accuracy.
SKILL.md
Loading
Related Openclaw Skills
Agent Guardrails
A mechanical enforcement suite that replaces soft markdown suggestions with hard git hooks and automated validation scripts for AI agents.
context-doctor
Visualize and diagnose context window allocation and token usage for AI coding agents.
AWI (Agentic Web Interface)
AWI is a zero-configuration single-binary tool for AI agents to read webpages and search the internet using an intelligent three-level fallback strategy.
Mission Control
A macOS-native web command center for monitoring agent activity, managing cron jobs, and tracking LLM costs in real-time.
Roundtable
A sophisticated prompting framework that simulates a multi-expert panel to provide diverse viewpoints and consensus for complex decision-making.
Agent Workflow Enforcer
A framework to stabilize AI Agent execution by replacing soft prompts with hard-coded gate scripts and persistent style contexts.
Featured*
