# humanizer

> Make AI drafts read like a person — Strips the tells of AI-generated writing out of a draft.

- Page: https://tesign.com/en/item/humanizer/
- JSON: https://tesign.com/en/item/humanizer/index.json
- Korean Markdown: https://tesign.com/item/humanizer/index.md
- Generated: 2026-09-17 01:56 UTC

## RANKS

- All-time #135 · Rising this week #12 · PRODUCTIVITY #10 (As of 2026-09-16 21:00 UTC)

## Numbers

- 48,604 stars — checked on GitHub 2026-09-15 19:34 UTC (+104 observed since the check)
- 7-day +456 observed via GH Archive (as of 2026-09-16 21:00 UTC)
- 24h +90 ★ · 30d +1,546 ★ (as of 2026-09-16 21:00 UTC)
- Star total = the value last checked on GitHub (stars_checked_at) + increases observed via GH Archive since. The 24h · 7d · 30d gains are GH Archive hourly events summed to the reference time (as_of). No score of ours.

## AT A GLANCE

- LICENSE: MIT (permissive) — https://spdx.org/licenses/MIT.html
- USAGE: Use, change and redistribute, commercially too. Keep the notice.
- OPEN SOURCE: YES
- LANGUAGE: Python
- PLATFORM: [unconfirmed]
- CATEGORY: AI · PRODUCTIVITY
- Tags: writing · agent-skills · editing
- How to start: For developers
- SOURCES: GitHub https://github.com/blader/humanizer
- USE PACKAGE: https://skills.sh/blader/humanizer

## ACTIVITY

- Last commit: 2026-09-06 20:26 UTC
- Latest release: v3.0.0 (2026-09-06)
- Contributors: 19
- Open issues (incl. PRs): 16
- Made by: an individual
- Checked on GitHub: 2026-09-15 19:34 UTC

## TESIGN TAKE

A rule set that names 25 AI writing tells and removes them; a sensible first pass before a human edit.

## WHY IT MATTERS

A draft written by a model reads smoothly but keeps its own verbal habits, and readers spot them first. humanizer is an agent skill that finds and removes those traces at the draft stage, attached to Claude Code, Codex or Cursor. It is written in Python under the MIT license and has its own skill distribution page. What counts as an AI tell varies by genre, so read the result yourself.

## BUILD FROM THIS

- An in-house polishing step every draft passes before publishing
- A post-processing rule that pulls drafts back to the brand voice

## WHO IT'S FOR

- Marketer — strips AI tells from a model-written draft
- Blogger — matches the prose to their own voice before publishing
- Docs writer — keeps reference prose plain and neutral

## START IN 5 MINUTES

```
$ npx skills add blader/humanizer --global
# Claude Code 2.1.142+ can install the plugin instead:
# /plugin marketplace add blader/humanizer
# /plugin install humanizer@humanizer
# Use: type /humanizer and paste your text
```

## CAVEATS

- What counts as an AI tell varies by genre, so read the result yourself
- It is designed to ask rather than invent when a sentence lacks a fact
- The README says nothing about how it performs on Korean text

## RECEIPT

- FIRST SEEN BY TESIGN: 2026-09-13 07:43 UTC
- AT SOURCE: 2026-01-18 05:30 UTC
- KEPT: 2026-09-13 11:32 UTC
- Published on TESIGN: 2026-09-13 11:32 UTC
- Text last updated: 2026-09-15 04:36 UTC

## How to read this

- This file was produced by the same build, from the same data, as the tesign.com item page. The text is editorial; the numbers are stored observations.
- null in the JSON means not observed — never zero. The Markdown writes [unconfirmed] for it.
- Star total = the value last checked on GitHub (stars_checked_at) + increases observed via GH Archive since. The 24h · 7d · 30d gains are GH Archive hourly events summed to the reference time (as_of). No score of ours.
- A summary, not legal advice.

[Image] https://tesign.com/img/og/humanizer.en.png