Skip to content

epoko77-ai/im-not-ai / skills/humanize-korean

humanize-korean

Verified

AI(ChatGPT·Claude·Gemini 등)가 쓴 한글 텍스트를 "사람이 쓴 글처럼" 윤문해주는 오케스트레이터 스킬. 번역투·영어 인용 과다·기계적 병렬·관용구·피동태 남용·접속사 남발·리듬 균일성·이모지/불릿 과다 등 10대 카테고리 85개 AI 티 패턴을 탐지·분류해 내용은 한 글자도 건드리지 않고 문체·리듬·표현만 자연스러운 한국어로 재작성한다. shim의 route_hint(light|standard|heavy)로 경로를 정해 잘 쓴 글은 1콜, 표준은 2콜, 중증·장문만 3+콜(진단→겨냥 윤문→finalize)로 처리한다. 트리거 — "AI 티 없애줘", "AI 같은 글 자연스럽게", "GPT/ChatGPT 문체", "AI 번역투 고쳐", "사람이 쓴 것처럼 윤문", "AI 윤문", "ChatGPT 티 제거", "한글 AI 탐지·윤문", "AI 글 사람처럼", "번역투 제거", "영어 인용 많은 글 윤문", "AI 글 티 안 나게", "휴머나이저", "humanize Korean", "AI detector bypass 한글". 후속 작업 — "특정 카테고리만 다시", "윤문 강도 조정", "장르 바꿔서", "이 문단만", "2차 윤문" 도 모두 이 스킬. 단순 맞춤법·오탈자 교정은 직접 처리, 번역은 번역 스킬, 내용 추가·삭제를 동반한 재작성은 별도 집필 스킬.

references/baseline_v2.json

View on GitHub
{
  "version": "v2.0-baseline-diff-2026-05-07",
  "source": "Placeholder. Toral 2019 simplification·normalisation·interference 정의 + 보고서 T1~T8 ko_manifestation 추정치. 비번역 한국어 (Sejong corpus·국립국어원 모두의 말뭉치 등) 정밀 측정은 별도 calibration 회차에서 수행.",
  "notes": "metric-engineer(Phase 3b) 산출. ALL cells carry _placeholder: true. Phase 6 integrator는 본진 baseline.json에 merge할 때 placeholder 플래그를 반드시 보존하고, calibration 완료 셀만 플래그를 해제한다.",
  "calibration_due": true,
  "genres": {
    "essay": {
      "lexical_diversity_ttr": {"mean": 0.62, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "simplification", "interpretation": "low = AI-like (repetition)"},
      "lexical_density": {"mean": 0.30, "stdev": 0.07, "_placeholder": true, "calibration_due": true, "axis": "simplification", "interpretation": "low = AI-like (function-word heavy)"},
      "ending_diversity": {"mean": 0.55, "stdev": 0.12, "_placeholder": true, "calibration_due": true, "axis": "simplification", "interpretation": "low = monotonic endings"},
      "normalisation_score": {"mean": 0.50, "stdev": 0.15, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "interpretation": "high = AI-like (-한다/-된다/-이다 concentration)"},
      "da_streak_rate": {"mean": 0.4, "stdev": 0.6, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "unit": "streaks_per_doc", "interpretation": "high = AI-like"},
      "inanimate_subject_rate": {"mean": 0.10, "stdev": 0.06, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T1", "interpretation": "high = AI-like (translationese)"},
      "by_passive_count": {"mean": 0.5, "stdev": 1.0, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2a", "unit": "per_doc", "interpretation": "high = AI-like"},
      "double_passive_count": {"mean": 0.2, "stdev": 0.6, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2b", "unit": "per_doc", "interpretation": "high = strong S2 signal"},
      "pronoun_density": {"mean": 0.012, "stdev": 0.010, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T3", "interpretation": "high = AI-like (he/she literal mapping)"},
      "deul_overuse_rate": {"mean": 0.005, "stdev": 0.008, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T4", "interpretation": "high = AI-like (-들 over-use)"},
      "relative_clause_nesting": {"mean": 0.3, "stdev": 0.5, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T5", "unit": "sentences_per_doc", "interpretation": "high = AI-like (left-modifier overload)"},
      "have_make_literal_count": {"mean": 0.4, "stdev": 0.8, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T6", "unit": "per_doc", "interpretation": "high = AI-like"},
      "double_particle_count": {"mean": 0.3, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T7", "unit": "per_doc", "interpretation": "high = AI-like"},
      "progressive_aspect_rate": {"mean": 0.10, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T8b", "interpretation": "high = AI-like (~고 있다 literal mapping)"}
    },
    "news": {
      "lexical_diversity_ttr": {"mean": 0.65, "stdev": 0.07, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "lexical_density": {"mean": 0.34, "stdev": 0.07, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "ending_diversity": {"mean": 0.40, "stdev": 0.10, "_placeholder": true, "calibration_due": true, "axis": "simplification", "note": "뉴스는 평서형 단조성이 일부 정상"},
      "normalisation_score": {"mean": 0.75, "stdev": 0.12, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "note": "뉴스 장르는 -이다/-한다 비중이 높음"},
      "da_streak_rate": {"mean": 1.0, "stdev": 1.0, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "unit": "streaks_per_doc"},
      "inanimate_subject_rate": {"mean": 0.18, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T1"},
      "by_passive_count": {"mean": 0.8, "stdev": 1.2, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2a", "unit": "per_doc"},
      "double_passive_count": {"mean": 0.3, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2b", "unit": "per_doc"},
      "pronoun_density": {"mean": 0.015, "stdev": 0.010, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T3"},
      "deul_overuse_rate": {"mean": 0.006, "stdev": 0.008, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T4"},
      "relative_clause_nesting": {"mean": 0.4, "stdev": 0.6, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T5", "unit": "sentences_per_doc"},
      "have_make_literal_count": {"mean": 0.5, "stdev": 0.9, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T6", "unit": "per_doc"},
      "double_particle_count": {"mean": 0.4, "stdev": 0.8, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T7", "unit": "per_doc"},
      "progressive_aspect_rate": {"mean": 0.08, "stdev": 0.06, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T8b"}
    },
    "blog": {
      "lexical_diversity_ttr": {"mean": 0.60, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "lexical_density": {"mean": 0.27, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "ending_diversity": {"mean": 0.65, "stdev": 0.12, "_placeholder": true, "calibration_due": true, "axis": "simplification", "note": "블로그는 종결 다양성 자연 높음"},
      "normalisation_score": {"mean": 0.40, "stdev": 0.18, "_placeholder": true, "calibration_due": true, "axis": "normalisation"},
      "da_streak_rate": {"mean": 0.3, "stdev": 0.6, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "unit": "streaks_per_doc"},
      "inanimate_subject_rate": {"mean": 0.08, "stdev": 0.06, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T1"},
      "by_passive_count": {"mean": 0.3, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2a", "unit": "per_doc"},
      "double_passive_count": {"mean": 0.2, "stdev": 0.5, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2b", "unit": "per_doc"},
      "pronoun_density": {"mean": 0.018, "stdev": 0.012, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T3", "note": "1인칭 사용 빈도 높음"},
      "deul_overuse_rate": {"mean": 0.004, "stdev": 0.007, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T4"},
      "relative_clause_nesting": {"mean": 0.2, "stdev": 0.4, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T5", "unit": "sentences_per_doc"},
      "have_make_literal_count": {"mean": 0.3, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T6", "unit": "per_doc"},
      "double_particle_count": {"mean": 0.2, "stdev": 0.5, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T7", "unit": "per_doc"},
      "progressive_aspect_rate": {"mean": 0.12, "stdev": 0.10, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T8b"}
    },
    "qa": {
      "lexical_diversity_ttr": {"mean": 0.58, "stdev": 0.09, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "lexical_density": {"mean": 0.25, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "ending_diversity": {"mean": 0.50, "stdev": 0.13, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "normalisation_score": {"mean": 0.55, "stdev": 0.18, "_placeholder": true, "calibration_due": true, "axis": "normalisation"},
      "da_streak_rate": {"mean": 0.5, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "unit": "streaks_per_doc"},
      "inanimate_subject_rate": {"mean": 0.15, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T1"},
      "by_passive_count": {"mean": 0.4, "stdev": 0.8, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2a", "unit": "per_doc"},
      "double_passive_count": {"mean": 0.3, "stdev": 0.6, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2b", "unit": "per_doc"},
      "pronoun_density": {"mean": 0.014, "stdev": 0.010, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T3"},
      "deul_overuse_rate": {"mean": 0.007, "stdev": 0.010, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T4"},
      "relative_clause_nesting": {"mean": 0.25, "stdev": 0.5, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T5", "unit": "sentences_per_doc"},
      "have_make_literal_count": {"mean": 0.5, "stdev": 0.9, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T6", "unit": "per_doc"},
      "double_particle_count": {"mean": 0.3, "stdev": 0.7, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T7", "unit": "per_doc"},
      "progressive_aspect_rate": {"mean": 0.10, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T8b"}
    },
    "dialogue": {
      "lexical_diversity_ttr": {"mean": 0.55, "stdev": 0.10, "_placeholder": true, "calibration_due": true, "axis": "simplification", "note": "대화는 반복이 자연스러움"},
      "lexical_density": {"mean": 0.22, "stdev": 0.08, "_placeholder": true, "calibration_due": true, "axis": "simplification"},
      "ending_diversity": {"mean": 0.70, "stdev": 0.12, "_placeholder": true, "calibration_due": true, "axis": "simplification", "note": "대화는 종결어미 자연 다양"},
      "normalisation_score": {"mean": 0.20, "stdev": 0.15, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "note": "대화는 -한다/-된다/-이다 거의 안 씀"},
      "da_streak_rate": {"mean": 0.1, "stdev": 0.3, "_placeholder": true, "calibration_due": true, "axis": "normalisation", "unit": "streaks_per_doc"},
      "inanimate_subject_rate": {"mean": 0.05, "stdev": 0.05, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T1"},
      "by_passive_count": {"mean": 0.1, "stdev": 0.3, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2a", "unit": "per_doc"},
      "double_passive_count": {"mean": 0.1, "stdev": 0.4, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T2b", "unit": "per_doc"},
      "pronoun_density": {"mean": 0.020, "stdev": 0.015, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T3"},
      "deul_overuse_rate": {"mean": 0.003, "stdev": 0.006, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T4"},
      "relative_clause_nesting": {"mean": 0.1, "stdev": 0.3, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T5", "unit": "sentences_per_doc"},
      "have_make_literal_count": {"mean": 0.2, "stdev": 0.5, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T6", "unit": "per_doc"},
      "double_particle_count": {"mean": 0.1, "stdev": 0.3, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T7", "unit": "per_doc"},
      "progressive_aspect_rate": {"mean": 0.15, "stdev": 0.10, "_placeholder": true, "calibration_due": true, "axis": "interference", "type": "T8b"}
    }
  },
  "axis_definitions": {
    "simplification": {
      "source": "Toral 2019; 보고서 line 351",
      "metrics": ["lexical_diversity_ttr", "lexical_density", "ending_diversity"],
      "interpretation": "low values → AI-like (repetition, function-word heavy, monotonic endings)"
    },
    "normalisation": {
      "source": "Baker 1993; 보고서 line 352",
      "metrics": ["normalisation_score", "da_streak_rate"],
      "interpretation": "high values → AI-like (-한다/-된다/-이다 concentration, '-다' streak runs)"
    },
    "interference": {
      "source": "Toury 1995 law of interference; 보고서 line 353",
      "metrics": [
        "inanimate_subject_rate",
        "by_passive_count",
        "double_passive_count",
        "pronoun_density",
        "deul_overuse_rate",
        "relative_clause_nesting",
        "have_make_literal_count",
        "double_particle_count",
        "progressive_aspect_rate"
      ],
      "interpretation": "high values → AI-like (영어 통사 보존)"
    }
  },
  "z_score_thresholds": {
    "comments": "Phase 6 integrator는 v1.6 thresholds(z>1.0 가산)를 그대로 계승. v2.0 placeholder는 calibration 완료 후 본진 baseline.json에 merge."
  }
}
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall