Skip to content
cccskills
FreePublished on GitHub

humanize-redo

Refine the latest Korean revision with a second editing pass. Target a specific style issue or paragraph, adjust editing intensity, or undo an individual change, then receive a change comparison and an updated grade.

OriginalShow the original description

가장 최근 윤문 결과를 2차로 다시 다듬는다 — 특정 카테고리·문단·강도 조정도 가능. humanize-korean strict 윤문(Phase B)을 기존 run_id에 재실행해 잔존 finding을 처리한다. 트리거 — "/humanize-redo".

See how to install

Good for

  • Rechecking style issues left after the first revision
  • Revising only translation-like phrasing
  • Adjusting editing intensity for one paragraph
  • Undoing a specific edit

Files included (1)

  • SKILL.md2.0 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

/humanize-redo — 2차 윤문 / 부분 재실행

cwd 기준 가장 최근 _workspace/{run_id}/를 찾아 humanize-korean 스킬의 strict 윤문(Phase B)부터 재호출한다.

사용자 지시

$ARGUMENTS

동작

  1. Glob으로 _workspace/YYYY-MM-DD-*/final.md(또는 01_input.txt)를 매칭해 최신 run_id 식별. 없으면 "이전 실행이 없습니다. /humanize로 시작하세요" 안내 후 종료.
  2. 사용자 지시 파싱:
    • 카테고리 지정("번역투만", "관용구만", "이모지만") → 해당 카테고리 finding만 재윤문
    • 문단 지정("이 문단만", "두 번째 문단만") → 해당 범위 finding만
    • 강도 조정("강도 낮춰"·"보수적으로" → S1만, "강도 높여" → S1+S2+S3)
    • 롤백 요청("이 변경 되돌려줘") → 해당 edit을 content-fidelity-auditor 롤백으로 처리
    • 지시 없음·"2차 윤문해줘" → 잔존 finding 전체 대상 round 2
  3. korean-style-rewriter 재호출 입력: 기존 02_detection.json 또는 05_naturalness_review.json의 잔존 finding + 사용자 지시를 target_filter로 전달.
  4. 산출물은 03_rewrite_v2.md(또는 v3)로 버전 분리. 이전 final.md는 final_prev.md로 백업.
  5. strict Phase C 병렬 검증 → Phase D 최종 출력(변경 비교 표 + 신규 등급).

루프 한도

최대 round 3. 그 이상 미해결이면 hold_and_report로 사람 검토 권고.

참고

  • 풀 파이프라인 신규 실행은 /humanize.
  • 분류 체계: humanize-korean/references/ai-tell-taxonomy.md

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

commit-ko

FreeSummary

Polish Korean commit messages drafted by AI into natural developer language. The skill removes stiff wording and literal translations while preserving the stated changes, commit type, scope, issue numbers, and code identifiers.

  • Polishing Korean commit message drafts
  • Keeping Conventional Commits prefixes intact
  • Reviewing several commit messages together
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

humanize

FreeSummary

Revise AI-written Korean text supplied directly or in a file. Choose the genre and editing intensity, then receive the revised text alongside a change rate, assessment grade, detection counts, and highlights of the main edits.

  • Polishing Korean text drafted by AI
  • Revising text from .txt or .md files
  • Adjusting editing intensity for the genre
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

humanize-korean

FreeSummary

Revise AI-written Korean to improve phrasing, sentence rhythm, and structure while preserving meaning, key terms, and tone. The skill chooses an editing depth, checks the changes, and provides revised text with review results.

  • Polishing a Korean draft written by ChatGPT or Claude
  • Improving repetitive sentence endings and rhythm
  • Reviewing a Korean report while preserving key terms
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

humanize-korean

FreeSummary

Polish AI-written Korean by detecting unnatural phrasing, repetitive structures, and uneven stylistic choices. Preserves meaning, facts, quotations, genre, and formality, then saves the revised text with a summary of changes and self-check results.

  • Polishing Korean blog drafts written by AI
  • Reducing translation-like phrasing in Korean columns
  • Editing Korean reports while preserving formality and facts
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

humanize-korean

FreeSummary

Polish AI-written Korean by identifying unnatural phrasing and repetitive sentence patterns while preserving facts and tone. The skill selects a review depth, checks how much changed, and returns revised text with comparisons and remaining warnings.

  • Polishing an AI-written Korean blog draft
  • Reducing translation-like phrasing in a Korean report
  • Reviewing a revised Korean paragraph more closely
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

humanize-scan

FreeSummary

Check Korean writing for six selected signs of AI-like style and see concrete editing examples with before-and-after counts. The conversation-based review helps decide whether the text needs a full editing pass or only minor adjustments.

  • Deciding whether a Korean draft needs full editing
  • Checking repetitive phrasing and sentence endings
  • Previewing edits before requesting a full revision
epoko77-ai/im-not-ai6.2KUpdated Oct 12, 2026

See all skills by epoko77-ai

Report a problem with this skill