Humanizer ID skill

Skill and CLI auditor for editing academic Indonesian text: removes calques, stiffness and cliches while leaving facts intact.

  • 193 ★ on the repository
  • Last commit
  • MIT
Skill

Minimal access

Why this level

  • The skill and auditor edit and check text without calling external services.
All reasons and checks

apissaj/humanizer-id

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/humanizer-id

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Minimal accessThe tool runs nothing and makes no network calls.

  • The skill and auditor edit and check text without calling external services.

This is third-party code. Review the repository files before installing.

What it does

Humanizer ID is an editing guide for academic and professional Indonesian text, for agents and LLM platforms. The skill hunts six typical problems of machine writing: piled-up nominalizations, calques from English, anticipatory denials, choppy short sentences, monotone paragraph openings and empty transitions without premises. The rules trace back to Indonesian Ministry of Education textbooks and university guidelines, each with a page number. The skill preserves scientific accuracy and makes no attempt to evade detectors. For agents there are `SKILL.md` and prompts, including a separate one for academic work. The repository also holds the Python package `humanizer-id` and the `scripts/audit.py` script, which checks Markdown or text against calque dictionaries and a glossary with `akademik`, `dinas` and `jurnalistik` presets. Dependencies are the standard library only.

Who it is for. Students, researchers and editors who write in Indonesian with AI help and want to clean out calques and machine stiffness.

Good fit when

  • A paper or article reads like a translation from English and the sentences sound unnatural.
  • A business letter must be brought to official style norms.
  • Before submission you want an auditor pass and a list of problems.
  • A journal text must be edited by academic rules with data preserved.

Not a fit when

  • You need editing of Russian, English or another language: the skill is built for Indonesian.
  • You want to fool an AI text detector: the skill does not do that.
  • You need a legal review instead of a stylistic one.

Example request

Edit this Indonesian paragraph with humanizer-id: remove calques and empty transitions, keep every fact and number.

Limitations

The skill covers Indonesian only, and the README and data are in Indonesian. The automatic check relies on calque dictionaries and presets and catches predefined patterns, while factual slips stay with the author. It adds no new facts and promises no AI detector evasion. Hermes, Claude Code, Gemini CLI and the `.agents` directory each use a different copy path. The `scripts/audit.py` CLI runs separately from the agent and changes nothing in the file, it only prints findings.

How to disable. Delete the `humanizer-id` directory from `~/.claude/skills/` (or from the Hermes, Gemini CLI or `.agents/skills/` directory). Remove the Python package with `pip uninstall humanizer-id`.

FAQ

How do I install in Claude Code?

Clone the repository into `~/.claude/skills/humanizer-id` with `git clone`. Gemini CLI and `.agents` use their own directories.

What is the auditor?

The `scripts/audit.py` script or the `humanizer-id` package: it checks text against calque dictionaries and a glossary and prints findings. Install with `pip install humanizer-id`.

Which presets exist?

Three: `akademik` for scholarly work, `dinas` for official letters and `jurnalistik` for journalism. A preset is set with the `--preset` flag.

Does it evade AI detectors?

No. The README states fact preservation and no detector manipulation. The skill fixes style and does not hide where a text came from.

Where do the rules come from?

From official Kemendikbudristek textbooks and university guidelines, with page numbers in `references/`.

Discussion

No questions or reviews yet.

Skills

A skill that edits Indonesian text: removes machine phrasing and English calques while keeping the author's voice

SkillMinimal access299Repository stars

A skill and Node.js CLI that score English text for AI-writing patterns and statistics and suggest fixes on a 0 to 100 scale

SkillMinimal access132Repository stars

An editor skill for Chinese text: removes filler, repetition and template phrasing, keeping facts, certainty and author voice

SkillMinimal access251Repository stars

A skill that removes AI tone from work writing while keeping every fact, commitment and caveat at the same strength.

SkillMinimal access129Repository stars
Russian stack

An editor for AI-written Russian text: removes bureaucratese, cliches and 38 AI patterns while keeping facts intact

SkillMinimal access210Repository stars

A skill that edits German AI drafts: it removes stock phrases, keeps facts and the author's voice and checks 72 language patterns.

SkillMinimal access205Repository stars

Comparison with similar tools

Cards that share the main search query, and how they differ.

ToolOfficialRepository starsMain difference
Humanizer ID (this card)no193
Humanizer-zhno22,682a similar skill for Chinese; here Indonesian, six patterns and rules from state textbooks
Humanizer RUno210a similar skill for Russian; here Indonesian and an offline auditor with presets

README in short

The repository explains why AI-written Indonesian looks stiff and gives six patterns with examples. It describes rule sources, installation for Hermes, Claude Code, Gemini CLI and `.agents`, and a standalone auditor. An extra skill for journal articles is included.

README badge

Report a problem