DrARaj

Ultimate Humanizer — AI skill for Claude Code

AI community

Claude Code skill that keeps AI writing tells out of your text, built on Wikipedia's Signs of AI writing guide.

How to install Ultimate Humanizer

This entry records only its repository, not the path inside it, so there is no exact command to give. Open DrARaj/ultimate-humanizer and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Ultimate Humanizer does

Claude Code skill that keeps AI writing tells out of your text, built on Wikipedia's Signs of AI writing guide.

Alternatives in AI

  • Humanizer — A Claude Code skill that removes signs of AI-generated writing from text, making it sound more natural and hum 11.3k ★
  • Academic Humanizer — Strip AI-writing tells from papers and grant proposals (NSF/NIH), while keeping scholarly voice and tying clai 1.1k ★
  • Cwc Long Running Agents — Claude Code's built-in /goal command gives you a generator/evaluator loop out of the box: set a completion con 668 ★

README

Ultimate Humanizer: delve, tapestry, pivotal and other AI writing tells struck through

lint selftest release license

Ultimate Humanizer

A Claude Code skill that keeps AI writing tells out of your text. It turns every pattern in Wikipedia's "Signs of AI writing" field guide into a numbered rule, adds the extra patterns from the humanizer and no-ai-slop skills, and ships a lint script that catches the mechanical ones.

Before and after

This paragraph was written to be bad on purpose:

Nestled in the heart of the developer ecosystem, Ultimate Humanizer stands as a testament to the power of thoughtful writing. It's not just a linter, it's a commitment to clarity. Additionally, it helps teams delve into their drafts, highlighting the intricate interplay between tone and meaning. In conclusion, it is a pivotal tool for anyone who writes. I hope this helps!

`lint.py` finds 16 FAIL lines in it (seven R9 vocabulary hits, three R4 promotional phrases, two R1 significance claims, and one each for R3, R11, R22 and R58). This is what the skill returned when asked to edit it:

Ultimate Humanizer is a Claude Code skill. It contains a set of writing rules and a Python lint script that checks a draft against them.

0 FAIL. It also dropped the claims about "helping teams", because nothing in the prompt supported them, and said so in its change note. Both samples are in [`assets/`](assets), and CI lints the clean one on every push.

Wh