Not in the top 1,000 by GitHub stars

hannsxpeter/humanizer

Pure-prompt Agent Skill that de-slops AI-sounding prose, rewrites it in a writer's own voice, and conservatively cleans hidden Unicode, with guards against invented facts and over-editing. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and 8 more AI coding tools.

Codex Cursor Gemini CLI GitHub Copilot Writing Engineering

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

`SKILL.md`

Week of Nov 2, 2026 · +0 / −0 lines

Before Local scripts → After Local scripts

e8f1248a7d49 → e8f1248a7d49

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Local scripts

First observation → e8f1248a7d49

No earlier version was observed for comparison.

Compare on GitHub

View on GitHub Add to my skills

Verified commit e8f1248a7d4919689c1f942e356ed95b102d9f07

Version-pinned installation · Codex

git clone --filter=blob:none --no-checkout https://github.com/hannsxpeter/humanizer.git .hotskills/hannsxpeter-humanizer && git -C .hotskills/hannsxpeter-humanizer checkout --detach e8f1248a7d4919689c1f942e356ed95b102d9f07
# Copy each selected skill directory from .hotskills/hannsxpeter-humanizer into .agents/skills/<skill-name>/

Version-pinned installation · Cursor

git clone --filter=blob:none --no-checkout https://github.com/hannsxpeter/humanizer.git .hotskills/hannsxpeter-humanizer && git -C .hotskills/hannsxpeter-humanizer checkout --detach e8f1248a7d4919689c1f942e356ed95b102d9f07
# Copy each selected skill directory from .hotskills/hannsxpeter-humanizer into .cursor/skills/<skill-name>/

Version-pinned installation · Gemini CLI

git clone --filter=blob:none --no-checkout https://github.com/hannsxpeter/humanizer.git .hotskills/hannsxpeter-humanizer && git -C .hotskills/hannsxpeter-humanizer checkout --detach e8f1248a7d4919689c1f942e356ed95b102d9f07
# Copy each selected skill directory from .hotskills/hannsxpeter-humanizer into .gemini/skills/<skill-name>/

Version-pinned installation · GitHub Copilot

git clone --filter=blob:none --no-checkout https://github.com/hannsxpeter/humanizer.git .hotskills/hannsxpeter-humanizer && git -C .hotskills/hannsxpeter-humanizer checkout --detach e8f1248a7d4919689c1f942e356ed95b102d9f07
# Copy each selected skill directory from .hotskills/hannsxpeter-humanizer into .github/skills/<skill-name>/

16GitHub stars

2stars in 7 days

1skill in this repository

14commits in 30 days

Heat

3.7

Star growth, last 7 daysNot available
Star growth, last 24 hoursNot available
Rank movement, last 7 daysNot available
Commits, last 30 daysNot available
Accounts asking about it through MCP, last 7 daysNot available
Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

30-day change not yet available: not enough history.

Individual skills in this repository

humanizer /

How it works

humanizer

When your agent uses it

Humanize and de-slop AI-sounding prose while preserving meaning, specific detail, and genuine human quirks. When a voice sample or style profile is available, rewrite in that writer's actual voice. Remove recurring AI tells such as inflated significance, stock vocabulary, uniform rhythm, excessive hedging, formulaic structure, and formatting artifacts. Also clean suspicious invisible Unicode and copy-paste residue conservatively, without treating them as proof of authorship. Use whenever a user wants to humanize, de-AI, de-slop, or de-robotify text; fix prose that sounds corporate, generic, like ChatGPT, or otherwise off; match a draft to a writer or named author; edit for authentic voice and rhythm; or clean zero-width characters, exotic spaces, bidi controls, and similar hidden text residue.

What it loads into context

Frontmatter

217 tokens

Always available to the agent

SKILL.md

4,233 tokens

When the skill is activated

Reference documents

22,627 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

When to use this
Core principle: variance over substitution
Step 0: Discover the voice (do this every run, before rewriting)
Step 0b: Stance mode (opt-in, off by default)
Step 0c: Density pre-check (match effort to evidence)
Step 0d: Text hygiene preflight (do this every run)
The multi-pass workflow
False-positive guardrails
Anti-signature guidance
Output contract
Scope and intended use
Reference files

Scripts executed separately

.github/scripts/check_drift.py

Files

.continue/rules/humanizer.mddoc2,052 bytes
.cursor/rules/humanizer.mdcasset2,041 bytes
.github/copilot-instructions.mddoc1,652 bytes
.github/dependabot.ymlasset149 bytes
.github/scripts/check_drift.pyruns15,401 bytes
.github/workflows/drift.ymlasset412 bytes
.gitignoreasset176 bytes
AGENTS.mddoc2,754 bytes
CHANGELOG.mddoc9,274 bytes
CONVENTIONS.mddoc1,658 bytes
GEMINI.mddoc1,544 bytes
LICENSEasset1,062 bytes
README.mddoc12,807 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
SKILL.mdcore18,378 bytes
evals/evals.jsonasset9,310 bytes
evals/files/VOICE.mddoc712 bytes
references/do-not-flag.mddoc7,231 bytes
references/examples.mddoc16,868 bytes
references/tell-patterns.mddoc25,261 bytes
references/text-hygiene.mddoc5,592 bytes
references/voice-matching.mddoc5,905 bytes

Read SKILL.md


# Humanizer

Turn AI-flavored or generically smooth prose into writing that reads as the
work of a person. When a writer's voice is available, make it read as the work
of *that* person. The goal is prose a careful human reader would not flag,
with its meaning and its human texture intact. The goal is not to beat any
particular detector, and this skill names and targets none.

The workflow also includes conservative text hygiene. It can remove
suspicious invisible Unicode and normalize copy-paste spacing in the prose it
rewrites, while preserving characters that carry real linguistic or display
meaning. This is not file-provenance or media-watermark tooling.

## When to use this

View the complete source on GitHub

Skills included

humanizer

Repository facts

License: MIT
Created: May 15, 2026
Last pushed: Oct 8, 2026
Forks: 1
Open issues: 0
Project activity: Active

Before you install

Local scripts

Automated signals, not a guarantee

Checked Nov 5, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 10 notes

SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed

No broad tool permissions
No dangerous shell commands
23 network referencesin 3 files: README.md:3 CHANGELOG.md:200 references/text-hygiene.md:111
No hidden or obfuscated text
No instruction overrides
1 executable scriptin 1 file: .github/scripts/check_drift.py:1
1 reference to secretsin 1 file: .github/scripts/check_drift.py:363
No requests to vote in Hot

Show all 25 signals by file

`.github/scripts/check_drift.py`

Executable scriptline 1
```
#!/usr/bin/env python3
```
Access to secretsline 363
```
[Credential-related source omitted]
```

`CHANGELOG.md`

Network accessline 200

[Unreleased]: https://github.com/hannsxpeter/humanizer/compare/v1.3.1...HEAD

Network accessline 201

[1.3.1]: https://github.com/hannsxpeter/humanizer/compare/v1.3.0...v1.3.1

Network accessline 202

[1.3.0]: https://github.com/hannsxpeter/humanizer/compare/v1.2.1...v1.3.0

Network accessline 203

[1.2.1]: https://github.com/hannsxpeter/humanizer/compare/v1.2.0...v1.2.1

Network accessline 204

[1.2.0]: https://github.com/hannsxpeter/humanizer/compare/v1.1.1...v1.2.0

Network accessline 205

[1.1.1]: https://github.com/hannsxpeter/humanizer/compare/v1.1.0...v1.1.1

Network accessline 206

[1.1.0]: https://github.com/hannsxpeter/humanizer/compare/v1.0.0...v1.1.0

Network accessline 207

[1.0.0]: https://github.com/hannsxpeter/humanizer/releases/tag/v1.0.0

`README.md`

Network accessline 3

![version](https://img.shields.io/badge/version-1.3.1-blue)

Network accessline 4

![license](https://img.shields.io/badge/license-MIT-green)

Network accessline 5

![type](https://img.shields.io/badge/type-pure--prompt%20skill-purple)

Network accessline 6

![dependencies](https://img.shields.io/badge/dependencies-none-brightgreen)

Network accessline 7

![patterns](https://img.shields.io/badge/tell%20catalog-32%20patterns-orange)

Network accessline 8

![tools](https://img.shields.io/badge/works%20with-13%20AI%20coding%20tools-teal)

Network accessline 9

[![drift check](https://github.com/hannsxpeter/humanizer/actions/workflows/drift.yml/badge.svg)](https://github.com/hannsxpeter/humanizer/actions/workflows/drift.yml)

Network accessline 19

`humanizer` is an [Agent Skill](https://agentskills.io): a folder whose

Network accessline 25

git clone https://github.com/hannsxpeter/humanizer ~/.claude/skills/humanizer

Network accessline 33

git clone https://github.com/hannsxpeter/humanizer ~/.agents/skills/humanizer

Network accessline 104

This feature adapts the prompt-compatible text-layer ideas from [watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover).

Network accessline 105

[watermarks-remover](https://github.com/guillaumemeyer/watermarks-remover).

Network accessline 224

This skill was built from the voice-preservation logic that powers [Scriveno](https://github.com/hannsxpeter/scriveno) (formerly Scriven), an

Network accessline 225

[Scriveno](https://github.com/hannsxpeter/scriveno) (formerly Scriven), an

`references/text-hygiene.md`

Network accessline 111

Source concept: https://github.com/guillaumemeyer/watermarks-remover

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Signal confidence: 25 facts

Executable script in code of a file that runs.github/scripts/check_drift.py:1High confidence
Access to secrets in code of a file that runs: a variable, placeholder or call, not a stored value.github/scripts/check_drift.py:363Low confidence
Network access in prose of documentationCHANGELOG.md:200Low confidence
Network access in prose of documentationCHANGELOG.md:201Low confidence
Network access in prose of documentationCHANGELOG.md:202Low confidence
Network access in prose of documentationCHANGELOG.md:203Low confidence
Network access in prose of documentationCHANGELOG.md:204Low confidence
Network access in prose of documentationCHANGELOG.md:205Low confidence
Network access in prose of documentationCHANGELOG.md:206Low confidence
Network access in prose of documentationCHANGELOG.md:207Low confidence
Network access in prose of documentationREADME.md:3Low confidence
Network access in prose of documentationREADME.md:4Low confidence
Network access in prose of documentationREADME.md:5Low confidence
Network access in prose of documentationREADME.md:6Low confidence
Network access in prose of documentationREADME.md:7Low confidence
Network access in prose of documentationREADME.md:8Low confidence
Network access in prose of documentationREADME.md:9Low confidence
Network access in prose of documentationREADME.md:19Low confidence
Network access in code of documentationREADME.md:25Low confidence
Network access in code of documentationREADME.md:33Low confidence
Network access in prose of documentationREADME.md:104Low confidence
Network access in prose of documentationREADME.md:105Low confidence
Network access in prose of documentationREADME.md:224Low confidence
Network access in prose of documentationREADME.md:225Low confidence
Network access in prose of documentationreferences/text-hygiene.md:111Low confidence

Works with

Agent compatibility, from the files in this repository
Agent	Status	Evidence	Install to
Claude Code	Compatible	Standard SKILL.md SKILL.md	`.claude/skills/<skill-name>/`
Codex	Native	Manifest AGENTS.md	`.agents/skills/<skill-name>/`
Cursor	Native	Path .cursor/rules/humanizer.mdc	`.cursor/skills/<skill-name>/`
Gemini CLI	Native	Path GEMINI.md	`.gemini/skills/<skill-name>/`
GitHub Copilot	Native	Manifest .github/copilot-instructions.md	`.github/skills/<skill-name>/`
OpenCode	Compatible	Standard SKILL.md SKILL.md	`.opencode/skills/<skill-name>/`

Native The repository ships a manifest, folder or frontmatter field for this agent.
Compatible A standard SKILL.md that this agent can read.
Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.

hannsxpeter/humanizer

History

SKILL.md

Share this page

Heat

Stars over time

Individual skills in this repository

How it works

humanizer

Frontmatter

SKILL.md

Reference documents

The playbook

Scripts executed separately

Files

Skills included

Repository facts

Before you install

.github/scripts/check_drift.py

CHANGELOG.md

README.md

references/text-hygiene.md

How the signals fit together

Works with

You might also use

Nanako0129/sepia

nanaism/yomiyasu

Raymondhou0917/speak-human-tw

madebymustafa/design-taste

yucezerey/turkish-anti-ai-skill

Vladimir-Human/humanizer-ru

`SKILL.md`

`.github/scripts/check_drift.py`

`CHANGELOG.md`

`README.md`

`references/text-hygiene.md`