SkillCDN 팩
이미 쓰는 AI 안의 마케터, 디자이너, 개발자. 숏폼 광고, 카드뉴스, 발표 덱, 웹 앱, 회의록을 직접 만들고, 보여 드리기 전에 검토합니다. 일에 맞는 스킬을 찾는 법은 docs/pack/finding-a-skill.md에 있습니다.
https://skillcdn.ai/gh/skillcdn/skills내 AI에 이 주소를 연결하면 이 스킬들을 쓸 수 있어요. 연결 방법 보기
- 저장소
skillcdn/skills - 기본 브랜치
main - 커밋
9e3fd2c - 라이선스
MIT - 검색단어와 의미
저장소 둘러보기
스킬과 문서의 원문에서 단어와 의미로 찾습니다. 화면용 번역은 검색되지 않아요.
The skills that drive Higgsfield (higgsfield-…, in any area) share how the tools behave, how credits are quoted, how models and images are found, how a take is decoded, and what the video models do w…
문서
- Higgsfield: what its skills shareThe skills that drive Higgsfield (higgsfield-…, in any area) share how the tools behave, how credits are quoted, how models and images are found, how a take is decoded, and what the video models do w…docs/higgsfield/README.md
- Portraits and first framesEvery person in a Higgsfield video has an approved portrait, and every generated shot an approved still first frame. The portrait keeps a face the same from shot to shot and from run to run; the fram…docs/higgsfield/cast.md
- Decoding a takeWhat a take says is settled by machine before anyone asks whether it matters: two plain speech-to-text decodes, a third to break a tie, word probabilities, a loudness envelope and a contact sheet. Th…docs/higgsfield/decodes.md
- Models, tiers and creditsHow a Higgsfield skill finds the models it needs in the catalog at run time, locks the lowest tier, quotes every credit before spending it, and handles what the server answers instead of a job. A ski…docs/higgsfield/models.md
- Pronunciation: what the video models do with soundsVideo models with native dialogue have no pronunciation dictionary, and a transcript often cannot show the fault. This file is the knowledge, per language, gathered from real runs of every Higgsfield…docs/higgsfield/pronunciation.md
- The sandboxsandbox_exec is a Linux box with ffmpeg, sox, ImageMagick, Pillow, faster-whisper, Playwright with headless Chromium, curl and two Latin caption fonts, where a Higgsfield skill reads frames, decodes…docs/higgsfield/sandbox.md