Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/tobyilee/course-builder/script-writingnpx skills add tobyilee/course-builder --skill script-writinggit clone --depth 1 https://github.com/tobyilee/course-builderWhat 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 | $0.00060 | $0.02394 |
| Opus 5 | $0.00030 | $0.01197 |
| Sonnet 5 | $0.00012 | $0.00479 |
| Haiku 4.5 | $0.00006 | $0.00239 |
Grade A, and why
script-writing 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Script Writing — Speakable Transcript
transcript는 귀로 듣는 콘텐츠다. note를 그대로 낭독하면 부자연스럽다. 문장 길이, 리듬, 호흡, 강조를 음성 기준으로 재설계.
구조 (plain text)
[slide 1]
첫 문장. 리듬을 위한 짧은 문장.
[pause:400]
두 번째 문장은 조금 더 길어도 괜찮지만 25 단어를 넘지 않도록 합니다.
[pause:600]
핵심 강조 전에는 조금 더 쉬어줍니다.
[slide 2]
새 슬라이드로 넘어갑니다.
...
Line layout (subtitle-sync 필수)
한 슬라이드의 내레이션을 절대 한 줄에 몰아넣지 말 것. generate-player.py의 parse_transcript_by_slide는 newline으로 분할해 각 줄을 자막 추적 단위로 사용하고, char_proportional_line_times가 각 줄의 글자 수 비율로 슬라이드 MP3 duration을 쪼개 {start,end}를 부여한다. 한 줄짜리 슬라이드는 자막 한 덩어리 + 하이라이트 불가.
규칙:
- 발화 단위 한 개 = 한 줄. 문장 또는
[pause:…]사이의 한 호흡 단위. [pause:NNN]마커는 자기만의 줄에 둔다 (인라인 금지). 플레이어가 pause 줄은 strip → 빈 줄로 필터링.[slide N]역시 자기만의 줄.- 빈 줄은 슬라이드 사이 구분용으로만 허용.
❌ 금지 (한 줄에 몰아쓰기 — 추적 불가):
[slide 1]
첫 문장. [pause:400] 두 번째 문장. [pause:600] 세 번째 문장.
✅ 올바름 (줄 단위 분할 — 추적 가능):
[slide 1]
첫 문장.
[pause:400]
두 번째 문장.
[pause:600]
세 번째 문장.
규칙
문장 길이
- 영어 ≤25 단어, 한국어 ≤20 어절
- 한 호흡에 읽을 수 있는 길이
축약·구어체 허용
- "그래서 이게 뭐냐면", "이게 꽤 중요합니다"
- 너무 문어적 표현 지양 ("기인한다" → "때문입니다")
Cue와 Marker
| 태그 | 용도 | 예 |
|---|---|---|
[slide N] |
TTS 후처리 시 오디오 구간 분할 | [slide 3] |
[pause:ms] |
문장 사이 호흡 | [pause:400] |
[emph]...[/emph] |
강조 (SSML 변환 시 emphasis) | [emph]중요[/emph] |
pause 권장값:
- 문장 사이: 300~400ms
- 문단 전환: 600~800ms
- 강조/질문 후: 800~1200ms
- 슬라이드 전환 후 첫 문장 앞: 500ms
Unspeakable 변환 (핵심)
Raw code literal → 설명형
- ❌
const x = () => { return 1; } - ✅ "const 키워드로 x를 선언하고, 화살표 함수로 1을 반환합니다"
긴 숫자 → 구어
- ❌ "3.14159265358979"
- ✅ "파이, 약 3.14"
- ❌ "403,825개"
- ✅ "약 40만 개"
URL → 이름
- ❌ "https://react.dev/learn/server-components"
- ✅ "React 공식 문서의 서버 컴포넌트 페이지"
심볼·이모지 → 설명 또는 제거
- ❌ "Use 👉 this"
- ✅ "이것을 사용하세요"
길이 계산
문자/분 속도는 voice와 TTS 엔진, speed 설정에 따라 달라진다. 아래 실측 표 기반:
Voice별 발화 속도 (char/sec, speed=1.0 기준)
| 엔진 | Voice | ko 실측 | en 실측 |
|---|---|---|---|
OpenAI gpt-4o-mini-tts |
nova |
≈4.3 | ≈3.8 (est.) |
OpenAI tts-1 / tts-1-hd |
기본값 | ≈4.2 | ≈3.7 (est.) |
| edge-tts | ko-KR-SunHiNeural |
≈4.3 | — |
| edge-tts | en-US-*Neural |
— | ≈3.5 |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 2d ago First seen · 202 lines · 60 tokens per session scan A 9a0092f36ba1
script-writing is a skill published in the GitHub repository tobyilee/course-builder (22 stars, last pushed 4mo ago), licensed MIT. It adds 60 tokens to every session and 2,394 once invoked, about $0.0003 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
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…
media-ingest
Ingest video, audio, PDF, book, screenshot, and GitHub repo content into the brain. Multi-format handling with entity extraction and backlink propagation. Covers video-ingest, youtube-ingest, and book-ingest subtypes.
diagnostic-stem-delivery
Audio production with diagnostic analysis, timecode parsing from documents, and verified export workflow.
vox-director
Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end on the Atlas Cloud API + local ffmpeg — script, collage keyframes, motion, voice-over, music, captions, all automated. Use this whenever the user wants a "Vox style" video, a paper/torn-paper collage animation, a "motion collage"…
seedance-vocab-ja
This skill should be used when the user asks for Japanese Seedance 2.0 prompt wording, Japanese cinematic vocabulary, or translation of camera, lighting, action, VFX, audio, and production terms into Japanese.
model-compatibility
Model family compatibility matrix covering loaders, resolutions, samplers, CFG, VAE, ControlNet, and LoRA compatibility for SD 1.5, SDXL, Flux, SD3, and video models.