A new open-source plug-in, called Humanizer, has been released for Anthropic’s Claude Code AI assistant. The plug-in utilizes a list of 24 language and formatting patterns compiled by Wikipedia editors to help AI models write more natural-sounding content, aiming to ‘humanize’ chatbots.
The initiative stems from WikiProject AI Cleanup, a group of Wikipedia editors who have been actively identifying AI-generated articles. The project founder, Ilyas Lebleu, led the effort to tag over 500 articles for review and published a comprehensive list of patterns commonly associated with AI writing in August 2025.
Humanizer functions as a ‘skill file’ for Claude Code, providing specific instructions to guide the AI assistant’s language output. While the plug-in has gained popularity on GitHub with over 1,600 stars, its effectiveness remains a subject of discussion. Initial tests suggest that Humanizer can make AI-generated responses sound more casual, but it may not significantly enhance accuracy or coding capabilities.
As the tech community explores the implications of tools like Humanizer on AI communication, it raises important considerations about the balance between natural language processing and maintaining factual accuracy within AI-generated content.
Source: WIRED