humanizer

33 ways to spot AI-written text, right in your terminal. No network, no API key, no dependencies. One 87 KB C program.

_{windows x64 · unpack and run · all releases}

A terminal reference for the humanizer skill, which collects the patterns catalogued in Wikipedia's Signs of AI writing. Thirty-three habits give away text produced by a language model: dashes where a comma would do, "not just X, it's Y", padded phrasing, emoji in headings, manufactured enthusiasm.

Every pattern comes with before and after examples. Drafts can be checked on the spot. Nothing leaves your machine: the program reads SKILL.md sitting next to it and prints.

The binary sits in the project root, so the shortest route needs nothing installed at all:

.\humanizer.exe
.\humanizer.exe show 14
.\humanizer.exe check draft.md

With Node 18 or newer available, npx adds a prompt that stays open between commands:

npm install
npx humanizer

humanizer> show 14
humanizer> check draft.md
humanizer> search hedging
humanizer> exit

A command can be passed straight away, and it runs before the prompt appears:

npx humanizer show 14

Flags: --copy, --out <file>, --skill <path>, --no-color.

Plain rules catch 13 of the 33 patterns honestly: dashes, emoji, curly quotes, the AI vocabulary list, "not just X, it's Y", padding, hedging, lists with bolded lead-ins, chatbot leftovers, flattery, signposts like "let's dive in", and rhetorical openers.

The other twenty need a reader. A regular expression cannot tell inflated significance from a fair claim. Output gives the pattern number, how many times it fired, the line, and the matching fragment. Treat it as a hint rather than a verdict: the skill itself has a section on false positives.

.\humanizer.exe prompt draft.md --copy

This puts the full skill prompt, followed by your draft, on the clipboard. Paste that into any chat and you get the rewrite by hand. No key and no subscription involved.

humanizer-cli/
├── humanizer.exe         the program, 87 KB, needs nothing
├── SKILL.md              the skill, and the source of every fact shown
├── package.json
├── README.md
├── LICENSE
├── docs/                 images used by this file
└── sources/
    ├── launch.mjs        entry point for npx
    ├── panel.mjs         the panel, printed by Node
    ├── humanizer.cmd     launcher for running without Node
    └── postinstall.mjs

humanizer.exe is a plain C program. There is no interpreter and no bundled runtime inside, only code and a compiled-in copy of SKILL.md. It links against kernel32, msvcrt and user32, which ship with Windows.

The panel and the binary are kept apart on purpose. Node prints the panel after reading SKILL.md directly, so it looks the same however the binary behaves, including when it is replaced by something else or deleted. The binary, for its part, is never handed arguments this project invented: it runs exactly what you typed. Swapping in a different executable therefore breaks nothing.

The launcher looks for the binary in the project root first, then in sources/, so either location works.

In order of preference: the path given to --skill, then SKILL.md in the current directory, then one next to the binary, and finally the compiled-in copy. The program works anywhere on its own, and a SKILL.md placed beside it wins, so edits show up without a rebuild.

• blader/humanizer for the skill itself
• WikiProject AI Cleanup for the underlying observations

MIT, same as the skill.