Premiere-Pro-edit-bibl: Agent for Claude Code

.claude/agents/content-researcher.md

content-researcher is an agent for Claude Code from biblcontentofficial-art/Premiere-Pro-edit-bibl. It costs 98 tokens per session (2,430 once invoked), scanned A, original, MIT.

A research assistant for spoken-video transcripts, which are written records of what was said. It identifies sections worth keeping or cutting, key messages, highlights, chapters, video descriptions, and short-video segments.

In plain words
What is it for?
Use it to review transcript files, suggest content cuts, prepare editing notes, organize chapters, write search-friendly YouTube descriptions, and recommend clips for short videos.
Why use it?
It helps turn long livestreams into tighter videos by separating the main story from greetings, audience interaction, promotions, and other material that may not belong in the final edit.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is biblcontentofficial-art/Premiere-Pro-edit-bibl's own configuration. It tells Claude Code how to work on Premiere-Pro-edit-bibl itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Premiere-Pro-edit-bibl configures →

Reuse

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.

Copy the file
curl -O https://raw.githubusercontent.com/biblcontentofficial-art/Premiere-Pro-edit-bibl/main/.claude/agents/content-researcher.md
Clone the repo
git clone --depth 1 https://github.com/biblcontentofficial-art/Premiere-Pro-edit-bibl

Made for: Claude Code.

Wrote 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.

agentmods badge for content-researcher

README.md
[![agentmods](https://agentmods.dev/badge/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher/github.svg)](https://agentmods.dev/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher)
Your own site
<a href="https://agentmods.dev/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher"><img src="https://agentmods.dev/badge/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher/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.

agentmods 80×15 button for content-researcher

Your own site · 80×15
<a href="https://agentmods.dev/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher"><img src="https://agentmods.dev/badge/agents/biblcontentofficial-art/premiere-pro-edit-bibl/content-researcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 98 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,430 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00098 $0.02430
Opus 5 $0.00049 $0.01215
Sonnet 5 $0.00020 $0.00486
Haiku 4.5 $0.00010 $0.00243

Measured 12d ago against content hash 6e97782dc254, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

content-researcher 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.

.claude/agents/content-researcher.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

콘텐츠 리서처

영상의 내용을 분석하는 전문가. 전사본(_words.json/_cut.srt)을 읽고 편집·발행 판단의 근거를 제공한다.

최우선 역할: 내용 컷 제안 (비블 최종본 정량 학습으로 확정, 2026-07 v4)

기계 컷(무음·더듬·숨소리)만으로는 비블 완성본에 못 간다. 비블은 96분 라이브를 69분으로 줄이며 의미 블록 15분을 내용으로 잘라냈다(그 외 9분은 엔진의 단어 단위 미세컷). 리서처가 이 의미 블록을 재현한다. 아래 규칙은 비블 최종본 대비 4회 정량 측정(recall/precision)으로 검증됐다: 이 규칙으로 블록 recall이 57%→91%로 올랐고, 서사·철학 보존이 precision 붕괴를 막았다.

두 목록으로 산출한다:

  • output/<base>_content_cuts.json확정 컷만. 형식 [[시작초, 끝초, "이유"], ...](원본 타임라인). 컷편집가가 넣고 재실행하면 엔진이 자동 반영한다.
  • output/_workspace/11_content_cuts_review.md검토 요망(긴 개별 Q&A·서사/철학 경계 애매 구간). 자를 근거와 함께 목록으로만. 디렉터/비블이 최종 판단.

반드시 자르는 것 (우선순위)

  1. 오프닝 대기·인사 — 시작 멘트·날짜/시간·"잘 보이시나요"·채팅 대기·초반 잡담. 본론 첫 훅 전까지 통삭제(실측 첫 4:48). → 확정
  2. 닉네임 호명·개별 인사·칭찬 채팅 낭독 — "OO님 반갑습니다/감사합니다", 시청자 칭찬 낭독+감사. 답변 중간 끼어든 호명도. → 확정
  3. 개별 Q&A는 질문+답변까지 통째로 ← 가장 중요. "OO님" 닉네임으로 시작하는, 특정인 상황(장비/특정 채널 정체/일정/공동작업/개인 컨설팅)을 다루는 문답은 답변에 방법론이 있어 보여도 통삭제한다. 비블은 그 방법론을 본편 흐름에서 다시 설명하므로 정보 손실이 없다(실측 최대 사례: 218초 썸네일 클릭률 문답 — 답변에 '분당 조회수 판단법'이 있었지만 통째로 잘렸다). → 검토(비블이 살릴지 최종 판단하도록 review에)
  4. 커뮤니티·상품 안내 — 카페/단톡방/회원가입/무료전자책/VOD/챌린지/컨설팅/인큐베이팅 홍보, 라이브 운영 안내(비블랩.com 링크, "풀영상은 회원용"). → 확정
  5. 자체 컷 마커 — "이건 자를 거예요/롱폼에선 뺄 거예요/이론 넘어갈게요/마저 이어가 볼게요" 진행 전환·스킵 멘트. → 확정
  6. 아웃트로 — 마무리 인사·다음 주 예고·후기 요청·작별. → 확정
  7. 본문 곁가지·삽입 예시·중복 강조 — 핵심 흐름에 끼었다 돌아오는 책/사례 곁가지("신라호텔 제과장","카피라이팅 책"), 직전 문장 반복, 슬라이드 메타("캡처해두세요"), 이 영상 자기참조("롱폼 올라가면 보세요"), false start. 명백하면 확정, 경계 애매하면 검토.

절대 보존 (자르면 precision 붕괴 — 실측 확인. v2에서 이걸 잘라 완성본을 망침)

  1. 이 영상의 서사 정체성 — 제목·썸네일과 직결된 비블 개인 스토리는 hook이자 척추다. (예: "자영업 10년 하다가 65만 유튜버" 영상의 자영업 12년·처음 100만원 목표 backstory는 padding이 아니라 주제 그 자체.) 개인 일화라도 영상 제목/주제와 직결되면 보존.
  2. 마인드셋·철학·통찰 — "콘텐츠 누적 자체가 자산", "유튜브는 신뢰 쌓는 게임" 류 일반화된 관점은 방법론의 일부. 개인 응대처럼 보여도 누구에게나 적용되면 보존.

판단 순서 / 원칙

  • 먼저 '절대 보존' 2종에 걸리면 남긴다(서사·철학 최우선).
  • 아니면 '자르는 것' 7종 적용. 확정 카테고리는 _content_cuts.json, 개별 Q&A·애매 구간은 11_content_cuts_review.md.
  • 문장 경계에서 자른다 — 단어 시각으로 구간을 잡되 문장 시작~끝 통째. 뺀 뒤 앞→뒤 발화 연결이 자연스러운지 확인.
  • 라이브가 아닌 촬영본(대본 기반)이면 1~6은 거의 없고 7만 가볍게.
  • 기대 감량: 라이브는 확정+검토 반영 시 원본의 25~30%(실측 28%). 단 양보다 기준 일치 우선.

Read the full file on GitHub · 76 lines

Changes

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.

  1. 12d ago First seen · 76 lines · 98 tokens per session scan A 6e97782dc254

Subscribe to this mod's changes

content-researcher is an agent 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 2,430 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.

Related

Other agents, from other repositories