#3 · Primary category: Writing & Content

humanizer

agent-skills ai-writing claude-code codex cursor prompt-engineering writing-tools

Agent skill that removes signs of AI-generated writing from text

Project last updated:08/19/26

GitHub Stars

38.8K

Forks

3.4K

Contributors

15

License

MIT

Why we included this project

Most people who write with an LLM know the feeling: the output is technically fine but reads like it was assembled by a machine, with that even rhythm, inflated claims, and boilerplate phrasing. Humanizer turns Wikipedia's 'Signs of AI writing' guide into a portable SKILL.md file that any agent harness can load, so you can hand it a paragraph, a blog post, or a whole file and get back a rewrite with the AI-isms stripped out. It is not a detector or a standalone app; it is a set of concrete editing instructions your coding agent follows, with before/after examples for each of the 33 patterns it catches. The voice-calibration option is the standout feature: give it a sample of your own writing and it adapts the rewrite to your sentence rhythm instead of producing generic clean output. A no-fabrication rule keeps it honest, since it never invents facts or citations that were not in the source text.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category