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/content-research/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/content-research)<a href="https://agentmods.dev/skills/biblcontentofficial-art/premiere-pro-edit-bibl/content-research"><img src="https://agentmods.dev/badge/skills/biblcontentofficial-art/premiere-pro-edit-bibl/content-research/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/content-research"><img src="https://agentmods.dev/badge/skills/biblcontentofficial-art/premiere-pro-edit-bibl/content-research.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 | $0.00098 | $0.01119 |
| Opus 5 | $0.00049 | $0.00560 |
| Sonnet 5 | $0.00020 | $0.00224 |
| Haiku 4.5 | $0.00010 | $0.00112 |
Grade A, and why
content-research 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
콘텐츠 리서치
영상의 내용을 분석해 편집 판단의 근거를 만든다. 컷편집 엔진이 만든 단어 단위 전사를 읽고 구조화한다.
입력 확보
- 먼저 읽기 좋은 전사본을 만든다:
python3 engine/transcript_export.py "output/<base>_words.json"→<base>_transcript.md(원본 타임라인 문단 + 타임코드). 이걸 읽고 분석한다. - 보조:
<base>_words.json(정밀 시각),<base>_cut.srt(컷 기준 문장) 전사가 없으면 컷편집을 먼저 돌리도록 요청한다 (추측 분석 금지).
분석 4종
1. 핵심 메시지 (1~3개)
영상이 진짜 전하려는 것. 제목/썸네일과 연결되는 주장. 각 메시지에 근거 발화 1줄 인용.
2. 하이라이트 / 훅 (타임코드 + 점수)
조회수·몰입을 끌 강한 구간. 용도 태그: [인트로후보] [숏폼후보] [강조].
hook 강도 5점 루브릭 (각 1점, 합산):
- 강한 단언/단정 ("~해야 합니다", "안 됩니다")
- 숫자·구체적 결과 (5%, 10만, 65만)
- 반전·역설 (통념 뒤집기)
- 호기심 격차 (답을 뒤로 미루는 떡밥)
- 감정 고조 (확신·분노·공감)
4점 이상만 [숏폼후보]/[인트로후보]로. 각 하이라이트에 점수와 근거 발화를 적는다.
숏폼 추출: 점수 높은 구간의 in-out 타임코드를 정하면 바로 9:16 클립을 만들 수 있다:
python3 engine/make_shorts.py "영상.mp4" "10:12-10:35" "13:40-14:05"
→ output/shorts/에 1080x1920 세로 + -14 LUFS 숏츠 생성. (자막은 별도)
3. 삭제 추천 구간 (타임코드 + 이유 + 인용)
엔진이 잡는 무음/추임새와 별개로, 내용상 군더더기:
- 같은 설명 반복, 삼천포(주제 이탈), 늘어지는 사족, 끊긴 미완성 발화
- 보수적으로 — 확실한 것만. 의미 손상 위험 높으면 추천하지 않고 '검토 요망'으로만 표기.
4. 챕터 지점 (타임코드 + 챕터명)
주제 전환점. 유튜브 챕터 타임스탬프로 바로 쓸 수 있게.
5. 제목 / 썸네일 문구 후보
영상에서 가장 강한 문장·반전·숫자·결과를 뽑아 제목 후보 35개 + 썸네일 문구 후보 35개. 비블 채널 톤(단호·구체·궁금증). 근거 발화 타임코드 병기. (비블 기획 워크플로우와 연결되는 지점)
타임코드 규칙
전사(_words.json)는 원본 시각, 자막(SRT)은 컷 타임라인 시각. 어느 기준인지 항상 표기한다. 컷 후 영상에 쓸 거면 SRT 기준으로 매칭.
출력
output/_workspace/10_research.md:
## 핵심 메시지
1. ... (근거: "...")
## 하이라이트
- 00:03:12 [인트로후보] "..."
## 삭제 추천
- 00:21:05~00:21:40 | 같은 설명 반복 | "..."
## 챕터
- 00:00:00 인트로 / 00:05:30 ...
왜 이렇게 하는가
- 컷 엔진은 '소리'(무음·추임새)를 자르지만, '내용'(삼천포·중복)은 못 판단한다. 리서처가 그 빈자리를 메운다.
- 추천마다 인용을 붙이는 건, 디렉터가 듣지 않고도 검증하고 비블이 신뢰하게 하기 위함.
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 · 68 lines · 98 tokens per session scan A 42e9a80d40d8
content-research is a skill published in the GitHub repository biblcontentofficial-art/Premiere-Pro-edit-bibl (21 stars, last pushed 1mo ago), licensed MIT. It adds 98 tokens to every session and 1,119 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
pr-to-video
Turn a GitHub pull request (a PR URL, owner/repo#N, or 'this PR' in a checked-out repo) into a code-change explainer video — changelog, feature reveal, fix, or refactor walkthrough built from the diff, commits, and files: the input is a code change, not a website. Not a product promo (/product-launch-video) or a no-PR…
general-video
Author or edit a custom HyperFrames composition when no specialized workflow fits, or when BRIEF.md sets flow: companion. Use for longer or multi-scene pieces, brand and sizzle reels, montages, static loops, static title cards, footage remixes, and freeform builds. Use motion-graphics instead for a short unnarrated…
motion-graphics
A short, design-led motion graphic where motion is the message — kinetic typography, stat count-up, chart/data-viz hit, logo sting / brand lockup, lower-third / callout / social overlay, animated map (highlight regions, connect places, zoom to a location), animated tweet / news-article / headline, webpage / UI…
gguf-quantization
GGUF format and llama.cpp quantization for efficient CPU/GPU inference. Use when deploying models on consumer hardware, Apple Silicon, or when needing flexible quantization from 2-8 bit without GPU requirements.
remake-reel
Analyze a reference video (a reel, montage, or ad the user likes) into an edit blueprint — shot boundaries, music beats, BPM, energy curve, the drop — and rebuild the same structure with the user's own footage. Use when someone says "make it like this video", "remake this reel with my clips", or asks what makes an…
stage-edit
Intelligent editing of real user-supplied footage—understand it with transcript/OCR/scene/silence/quality/vision evidence, then choose deterministic timeline operations or a constrained semantic AI edit. Trigger for repurpose, montage, cleanup, localization, narration, or local content changes.