This project provides a Claude Code skill to transform AI-generated text into a more natural, human-sounding style. It is designed for users of Claude Code who wish to refine AI-generated content, offering a practical solution to mitigate the tell-tale signs of machine writing and improve text authenticity.

How It Works

The humanizer operates by identifying and correcting 24 distinct patterns commonly found in AI-generated text, drawing from Wikipedia's "Signs of AI writing" guide and the WikiProject AI Cleanup. It addresses issues across content, language, style, communication, filler, and hedging. The core principle is to counteract the tendency of Large Language Models (LLMs) to produce statistically probable, generalized phrasing by replacing formulaic expressions with more specific, natural, and varied language.

Quick Start & Requirements

• Primary install / run command: Recommended installation involves cloning the repository directly into the Claude Code skills directory: git clone https://github.com/blader/humanizer.git ~/.claude/skills/humanizer. A manual installation option involves copying the SKILL.md file.
• Non-default prerequisites and dependencies: Requires the Claude Code environment. No other specific software dependencies are listed.
• Usage: Invoke via /humanizer [paste your text here] within Claude Code, or prompt Claude directly with a request to humanize text.
• Links: Installation instructions are detailed in the README.

Highlighted Details

• Addresses 24 specific patterns identified in AI-generated writing, categorized into content, language, style, communication, filler, and hedging.
• Provides explicit "Before" and "After" examples for each of the 24 patterns, demonstrating the transformation.
• The methodology is grounded in observations from Wikipedia's AI Cleanup project, offering a data-driven approach.
• Includes a full example showcasing the humanization of a paragraph.

Maintenance & Community

The project is associated with WikiProject AI Cleanup, indicating a maintained effort. Version history is provided, showing updates to examples and the core logic. No specific community channels (e.g., Discord, Slack) or notable contributors are listed in the provided README.

Licensing & Compatibility

The project is released under the MIT License. This license is permissive and generally allows for commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

The effectiveness of the humanizer is dependent on the accuracy and comprehensiveness of the 24 identified AI writing patterns. As a Claude Code skill, its direct application is confined to that specific environment. The README does not detail any known bugs or unsupported platforms.