{"id":50741339,"url":"https://github.com/dotoricode/korean-humanizer","last_synced_at":"2026-06-10T17:02:52.478Z","repository":{"id":354353397,"uuid":"1223220200","full_name":"dotoricode/korean-humanizer","owner":"dotoricode","description":"Korean humanizer prompt and skill for removing the usual AI smell from generated writing.","archived":false,"fork":false,"pushed_at":"2026-05-21T09:54:42.000Z","size":404,"stargazers_count":3,"open_issues_count":0,"forks_count":1,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-05-21T17:44:42.133Z","etag":null,"topics":["ai-detection","ai-writing","claude","claude-code","claude-skill","codex","humanizer","korean","korean-language","korean-nlp","korean-writing","llm","prompt","prompt-engineering","style-transfer","writing-tools"],"latest_commit_sha":null,"homepage":"https://github.com/dotoricode/korean-humanizer/wiki","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/dotoricode.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-04-28T05:47:34.000Z","updated_at":"2026-05-21T09:54:46.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/dotoricode/korean-humanizer","commit_stats":null,"previous_names":["dotoricode/korean-humanizer"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/dotoricode/korean-humanizer","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dotoricode%2Fkorean-humanizer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dotoricode%2Fkorean-humanizer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dotoricode%2Fkorean-humanizer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dotoricode%2Fkorean-humanizer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/dotoricode","download_url":"https://codeload.github.com/dotoricode/korean-humanizer/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/dotoricode%2Fkorean-humanizer/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34161284,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-06-10T02:00:07.152Z","response_time":89,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-detection","ai-writing","claude","claude-code","claude-skill","codex","humanizer","korean","korean-language","korean-nlp","korean-writing","llm","prompt","prompt-engineering","style-transfer","writing-tools"],"created_at":"2026-06-10T17:02:51.719Z","updated_at":"2026-06-10T17:02:52.469Z","avatar_url":"https://github.com/dotoricode.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# korean-humanizer\n\n\u003e Korean AI output has a recognizable tell. This removes it.\n\n[한국어](README.ko.md) · [中文](README.zh-CN.md)\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-blue.svg)](LICENSE)\n[![Version](https://img.shields.io/badge/version-v1.0.1-brightgreen.svg)](CHANGELOG.md)\n[![Status](https://img.shields.io/badge/status-stable-brightgreen.svg)](docs/STABILITY-PROMISE.md)\n[![Patterns](https://img.shields.io/badge/patterns-137%2B-brightgreen.svg)](references/ko-ai-signals.md)\n[![Domains](https://img.shields.io/badge/domains-12-brightgreen.svg)](references/ko-ai-signals.md#부록-e-도메인별-카테고리-우선-적용)\n\nKorean text from LLMs tends to leak: stiff formalism, empty intensifiers, filler connectives that don't exist in natural speech. Readers notice. `korean-humanizer` strips those patterns — without touching the meaning.\n\n---\n\n```diff\n# LinkedIn\n- 이번 프로젝트를 통해 다양한 기술적 도전을 경험하고 성장할 수 있었습니다. 이러한 경험은 앞으로의 커리어에 있어서 매우 소중한 자산이 될 것이라 확신합니다. 🚀\n+ 이번 프로젝트에서 여러 기술적 도전을 경험했고, 그 과정에서 성장할 수 있었습니다. 이 경험은 앞으로의 커리어에도 소중한 자산이 될 것 같습니다.\n\n# Email\n- 안녕하세요. 다름이 아니오라 미팅 일정과 관련하여 말씀드리고자 연락드립니다. 부득이한 사정으로 인해 일정 변경이 불가피한 상황이 발생하여 양해를 구하고자 합니다.\n+ 안녕하세요. 미팅 일정 때문에 연락드립니다. 부득이한 사정으로 일정 변경이 필요해 양해를 구하고자 합니다.\n\n# Marketing\n- 🚀 혁신적인 솔루션을 활용하여 다양한 비즈니스 가치를 극대화하고, 이러한 접근을 통해 사용자 경험을 한층 더 고도화할 수 있습니다. ✨\n+ 이 솔루션으로 여러 비즈니스 가치를 더 크게 만들고, 사용자 경험도 한 단계 개선할 수 있습니다.\n```\n\n→ [Try it in 30 seconds](#install)\n\n---\n\n## Install\n\n### Codex\n\n```bash\ngit clone https://github.com/dotoricode/korean-humanizer.git\ncd korean-humanizer\nbash scripts/install-codex-skill.sh\n```\n\n### Claude Code\n\n```bash\nmkdir -p ~/.claude/skills\ngit clone https://github.com/dotoricode/korean-humanizer.git ~/.claude/skills/korean-humanizer\n```\n\nThen ask naturally — `이거 AI 티 빼줘:` followed by your Korean text.\n\n**Other LLMs:** paste [`PROMPT.short.md`](PROMPT.short.md) as a system prompt, or use the full [`PROMPT.md`](PROMPT.md).\n\n---\n\n## Pattern Categories\n\n| # | Category | Examples |\n|---|---|---|\n| 1 | Empty intensifiers | `매우`, `굉장히`, `한층 더` |\n| 2 | Empty adjectives | `다양한`, `혁신적인`, `포괄적인` |\n| 3 | Translation-ese / stiff formalism | `~에 있어서`, `~을 통해`, `다음과 같습니다` |\n| 4 | `것이다` / `이러한` / `해당` overuse | `사실 ~인 것이다`, `이러한 이유로` |\n| 5 | Template openings and closings | `오늘은 알아보겠습니다`, `도움이 되셨길 바랍니다` |\n| 6 | Passive / hidden agents | `고려되어야 합니다`, `활용되어 왔다` |\n| 7 | Filler connectives | `또한`, `뿐만 아니라`, `결론적으로` |\n| 8 | Forced triplets / parallelism | `빠르고 정확하며 효율적` |\n| 9 | Honorific / ending mismatch | mixed `~합니다`, `~해요`, `~다` |\n| 10 | Emoji overuse | repeated section emoji / hype emoji |\n| 11 | Excessive hedging | `~라고 할 수 있습니다`, `~인 것 같습니다` |\n| 12 | Korean LLM vocabulary tells | `활용`, `극대화`, `시사한다`, `도모` |\n\nFull catalog: [`references/ko-ai-signals.md`](references/ko-ai-signals.md) · Quick reference: [`CHEATSHEET.md`](CHEATSHEET.md)\n\n---\n\n## Customization\n\nBan words, set preferences, or define a brand voice — all applied before the catalog runs:\n\n```text\n이거 AI 티 빼줘. 금지=활용,매우; 선호=유용하다→쓸만하다:\n[Korean text]\n```\n\nFor a persistent tone profile, see [`examples/brand-voice-template.md`](examples/brand-voice-template.md).\n\n---\n\n## Contributing · License · Services\n\nContributions welcome — [`CONTRIBUTING.md`](CONTRIBUTING.md) · [Pattern addition](.github/ISSUE_TEMPLATE/pattern_addition.md) · [Bug report](.github/ISSUE_TEMPLATE/bug_report.md)\n\nMIT. Use it, fork it, adapt it.\n\nThird-party community demo by Socialistic/Tinkerland (not an official service — user input is processed by the external operator):\n\n[![Try writing-dotoricode-korean-humanizer-5d759b on Socialistic][socialistic-demo-badge]][socialistic-demo-link]\n\n[socialistic-demo-badge]: https://socialistic.ai/api/embed/writing-dotoricode-korean-humanizer-5d759b\n[socialistic-demo-link]: https://socialistic.ai/en/skill/writing-dotoricode-korean-humanizer-5d759b?utm_source=github\u0026utm_medium=readme\u0026utm_campaign=20260520-writing-koc-creators\u0026utm_content=badge\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdotoricode%2Fkorean-humanizer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdotoricode%2Fkorean-humanizer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdotoricode%2Fkorean-humanizer/lists"}