humanizer
The humanizer skill detects and corrects AI-generated Korean text patterns based on linguistic research, specifically the KatFishNet paper. It analyzes 40 distinct patterns across six categories, including punctuation, spacing, vocabulary, and translation-ese, to make text sound natural while preserving original meaning.
Is humanizer safe to install?
Safe to install: our audit of humanizer's source files found 0 shell commands, 0 external URLs, file reads and writes (low risk). Every command and URL listed appears verbatim in the skill's source. The skill reads and writes local files to process text but does not execute shell commands or make external network requests.
How we audit skills: our security review methodology.
Who is this skill for?
Users who want to refine AI-generated Korean text to remove robotic markers and improve natural flow.
What can you do with it?
- Detecting AI-specific writing patterns in Korean text.
- Rewriting AI-generated content to sound like human-written text.
- Analyzing text for linguistic markers like excessive commas, rigid spacing, and translation-ese.
- Validating the naturalness of Korean text using a grading system.
How good is this skill?
Quality score: 5/10. The skill documentation is highly detailed, providing clear methodology, specific linguistic references, and robust safety mechanisms like the change rate guard and meaning preservation checks.
What does the skill file contain?
# humanizer: 한국어 AI 작문 패턴 감지 및 교정기 ## 소개 당신은 AI가 생성한 텍스트의 특징적인 패턴을 감지하고 교정하는 한국어 글쓰기 분석 전문가입니다. 당신의 전문성은 과학적 언어학 연구, 특히 KatFishNet 논문(ArXiv 2503.00032v4)에 기반하고 있으며, 이 논문은 인간이 작성한 텍스트와 LLM이 생성한 한국어 텍스트 사이의 측정 가능한 차이를 식별했습니다. **중요한 컨텍스트**: - 이 스킬은 **한국어 텍스트에만** 집중하며, 영어 번역이나 다른 언어는 다루지 않습니다 - 검출 패턴은 높은 정확도(쉼표 패턴: 94.88% AUC, 품사 다양성: 82.99% AUC, 띄어쓰기: 79.51% AUC)를 가진 **실증적 언어학 분석**에 기반합니다 - 당신의 목표는 원래의 의미와 의도를 보존하면서 텍스트를 **자연스럽게 인간이 작성한 것처럼** 만드는 것입니다 - 우선순위에 따라 구성된 6개 카테고리의 40가지 고유 패턴을 분석하며, 각 패턴은 S1/S2/S3 심각도로 분류됩니다 ## 작업 설명 실행될 때 다음을 수행합니다: 1. **한국어 텍스트 읽기** - 사용자가 제공한 텍스트(인자, 대화 컨텍스트, 또...
Frequently asked questions
What is the basis for the detection patterns?
The skill uses empirical linguistic research, primarily the KatFishNet paper (ArXiv 2503.00032v4), which identifies measurable differences between human and LLM-generated Korean text.
How does the skill handle the original meaning of the text?
The skill performs a mandatory 6-point self-check to ensure facts, numbers, dates, names, logical relationships, and core conclusions remain unchanged during the rewriting process.
What happens if the skill changes too much of the original text?
The skill includes a 'Change Rate Guard' that calculates the percentage of modified words. If the change rate exceeds 50%, the skill stops and asks the user for confirmation before proceeding.
Related skills
copy-editing
76.3KMarketers, copywriters, and content creators who need to refine, polish, or refresh existing marketing materials
The copy-editing skill provides a structured framework for reviewing and improving existing marketing copy. It uses a seven-sweep process covering clarity, voice, benefits, proof, specificity, emotion, and risk reduction, alongside an expert panel scoring system for quality assurance.
brief
6.3KContent marketers, SEO specialists, and writers preparing content for search engine ranking
The brief skill provides a structured framework for creating editor-ready content briefs. It guides users through SERP analysis, intent mapping, outline generation, E-E-A-T requirement setting, and SEO target definition.
content-research-writer
5.1KWriters, content creators, technical documentarians, and professionals producing blog posts, newsletters, or case studies
The content-research-writer skill assists users in drafting, outlining, researching, and refining written content. It provides section-by-section feedback, manages citations, and helps maintain a consistent writing voice.
brand-voice
3.5KUsers who require consistent writing styles across multiple platforms and want to avoid generic AI-generated copy
The brand-voice skill generates a reusable writing style profile based on provided source material such as social media posts, essays, and documentation. It provides a structured schema for downstream AI tasks to ensure consistent tone across content and outreach workflows.