Humanize-AI for Openclaw

An advanced writing editor that identifies and removes AI-generated patterns to make text sound authentic and human.

artur-zhdan
v1.0.0
Jan 31, 2026
9
8.3k
0

Install & Download

1. ClawHub CLI

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

npx clawhub@latest install humanize

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 humanize 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 Humanize-AI?

Humanize-AI is a professional-grade writing tool designed to scrub AI-generated artifacts from text, transforming robotic prose into natural, engaging content. Built on the comprehensive guidelines from Wikipedia's WikiProject AI Cleanup, this skill targets specific linguistic fingerprints—such as inflated symbolism, overused 'AI vocabulary' (like 'delve' or 'tapestry'), and formulaic sentence structures—that characterize machine-generated text.

Beyond simply cleaning up text, this skill focuses on injecting 'soul' into writing by varying sentence rhythm, acknowledging complexity, and allowing for personality. It is a critical component for developers and content creators using Openclaw Skills who need to ensure their documentation, blog posts, and communications maintain a high level of human quality and credibility.

Humanize-AI Use Cases

  • Refining technical documentation to sound less like a manual and more like a developer-to-developer conversation.
  • Cleaning up AI-generated marketing copy by removing 'groundbreaking' or 'vibrant' buzzwords.
  • Auditing large-scale content repositories for robotic patterns using integrated grep and glob tools.
  • Humanizing email drafts and reports to remove sycophantic or servile chatbot artifacts.
  • Converting sterile, outline-style summaries into cohesive, narrative prose with a distinct voice.

How Humanize-AI Works

  1. The skill analyzes the provided text against 24 distinct detection patterns, including superficial analysis and negative parallelisms.
  2. It identifies 'soulless' indicators such as monotone sentence structure and the lack of first-person perspective.
  3. The agent rewrites problematic sections, substituting vague 'weasel words' with concrete facts and specific details.
  4. It adjusts the rhythm by mixing short, punchy sentences with longer, more complex ones to mimic human thought patterns.
  5. The final output is presented as humanized text, often accompanied by a log of the specific AI-isms that were removed.

Humanize-AI Setup

To integrate this skill into your workflow, ensure your environment supports the Openclaw Skills protocol and has the following tools enabled: Read, Write, Edit, Grep, and Glob.

# Example of how to trigger the humanize-ai workflow via CLI
openclaw run humanize-ai --input path/to/draft.md

No complex configuration is required, though providing context about the intended audience (e.g., 'technical developers' or 'casual readers') helps the skill better match the desired personality.

Humanize-AI Data Schema & Taxonomy

Humanize-AI operates primarily on text strings and Markdown files. It organizes its internal logic based on the following taxonomy:

Component Description
Content Patterns Identifies inflated significance, undue emphasis on notability, and promotional language.
Grammar Patterns Detects 'AI Vocabulary' (e.g., underscore, pivotal), copula avoidance, and rule-of-three overuse.
Style Patterns Flags em dash overuse, excessive bolding, and title-cased headings.
Communication Artifacts Removes chatbot filler like 'I hope this helps' or knowledge-cutoff disclaimers.

Humanize-AI Advanced Features

  • Multi-agent support for bulk humanizing across entire directories using Glob and Grep tools.
  • Personality injection logic that allows for opinionated, first-person, or nuanced technical tones.
  • Automatic 'Slop' detection based on the latest community observations from Wikipedia's AI cleanup projects.
  • Pattern-based rewriting that replaces 'Elegant Variation' (synonym cycling) with clear, direct terminology.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*