Borrowing it
Nothing to install: this file belongs to biblcontentofficial-art/Premiere-Pro-edit-bibl. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/biblcontentofficial-art/Premiere-Pro-edit-bibl/main/.claude/skills/subtitle-editing/SKILL.mdgit clone --depth 1 https://github.com/biblcontentofficial-art/Premiere-Pro-edit-biblWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/biblcontentofficial-art/premiere-pro-edit-bibl/subtitle-editing)<a href="https://agentmods.dev/skills/biblcontentofficial-art/premiere-pro-edit-bibl/subtitle-editing"><img src="https://agentmods.dev/badge/skills/biblcontentofficial-art/premiere-pro-edit-bibl/subtitle-editing/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/biblcontentofficial-art/premiere-pro-edit-bibl/subtitle-editing"><img src="https://agentmods.dev/badge/skills/biblcontentofficial-art/premiere-pro-edit-bibl/subtitle-editing.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00107 | $0.00964 |
| Opus 5 | $0.00053 | $0.00482 |
| Sonnet 5 | $0.00021 | $0.00193 |
| Haiku 4.5 | $0.00011 | $0.00096 |
Grade A, and why
subtitle-editing scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
자막 교정
엔진이 만든 컷 정렬 SRT의 타이밍은 그대로 두고 텍스트 품질과 가독성을 올린다.
입력
output/<base>_cut.srtoutput/_workspace/10_research.md(고유명사·핵심어),glossary.txt(있으면)
교정 4종
1. 고유명사 교정
Whisper 오인식(닉네임·브랜드·전문용어)을 수정. 같은 오류가 반복되면 일괄 치환. 리서치/용어집에 근거. 확신 없으면 원문 유지.
2. 줄 균형 / 분할
- 한 줄 ~16자(한글) 초과 → 자연 끊김에서 분할 (기존 시각 구간 안에서)
- 2줄이면 위아래 길이 균형
3. 읽기속도 (CPS)
글자수 / 표시시간이 너무 빠른 자막(대략 한글 9자/초 초과) → 시간 늘리거나 분할. 단 다음 자막 침범 금지.
4. 맞춤법 / 숫자
띄어쓰기·문장부호 정리, "백만"→"100만", "퍼센트"→"%" 등 채널 표기 통일.
도구: subtitle_polish.py (기계적 교정 자동화)
줄 균형·읽기속도(CPS)·맞춤법·타이밍 무결성·다중포맷은 엔진으로 일괄 처리:
python3 engine/subtitle_polish.py "output/<base>_cut.srt"
→ 교정 SRT(원본 .srt.bak 백업) + .vtt + 비블 스타일 .ass(폰트/외곽선/하단중앙) 생성.
이 도구가 못 하는 **문맥 판단(고유명사 교정·의미 기반 분할)**은 에디터가 직접. 흐름: ① 고유명사 등 문맥 교정을 SRT에 반영 → ② subtitle_polish.py로 마감.
강조 키워드 자막 (선택)
숫자·단위·핵심어를 강조색으로 칠한 ASS를 따로 만들 수 있다 (기획자의 [자막강조] 마커와 연결):
python3 engine/emphasis_subs.py "output/<base>_cut.srt" 구체성 영업사원 치트키
→ <base>_cut_emphasis.ass (숫자/단위 자동 + 지정 키워드를 teal로). 리서치 핵심어를 키워드로 넘기면 효과적.
타이밍 무결성 (절대 규칙)
교정 후 반드시 검증: 모든 자막의 시작<끝, 이전 끝 ≤ 다음 시작(겹침 0). 분할 시 기존 구간을 글자 비율로 쪼갠다. 검증 실패하면 교정 롤백.
출력
output/<base>_cut.srt(교정본으로 덮어쓰기, 원본은 작업 전.srt.bak으로 백업)output/_workspace/40_subtitle_notes.md— 주요 교정(특히 일괄 치환한 고유명사)
왜 이렇게 하는가
- 타이밍은 엔진이 컷에 맞춰 정밀 정렬했으므로 건드리면 어긋난다 → 텍스트만 고친다.
- 고유명사 오인식은 비블 채널 신뢰도에 직결 → 리서치 기반으로 확실히. 단 과교정은 의미 왜곡이라 애매하면 둔다.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 54 lines · 107 tokens per session scan A d896149ff08b
subtitle-editing is a skill published in the GitHub repository biblcontentofficial-art/Premiere-Pro-edit-bibl (21 stars, last pushed 1mo ago), licensed MIT. It adds 107 tokens to every session and 964 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
seedance-vocab-ko
This skill should be used when the user asks for Korean Seedance 2.0 prompt wording, Korean cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Korean.
korean-report-style
Korean-language conventions for technical reports, progress updates, meeting agendas, software handoffs, architecture documents, runbooks, API references, model evaluations, and consortium deliverables — prose register, framing of shortfalls and risks, terminology accuracy, and structural integrity after edits. Use…
seedance-vocab-ko
This skill should be used when the user asks for Korean Seedance 2.0 prompt wording, Korean cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Korean.
translate-polish
A Korean editing skill for making translated writing sound natural while keeping the original information and meaning unchanged.
proofread
A Korean proofreading skill that corrects objective language errors such as spelling, spacing, standard-word usage, foreign-word spelling, punctuation, and clear grammatical mistakes.
k-humanizer
Use when polishing Korean resumes, career descriptions, portfolio summaries, and application writing so they sound natural rather than AI-written across operations, planning, QA, design, marketing, customer service, research, education, and other fields. Preserve facts, meaning, ownership, results, field-specific…