claude-skills 是做什麼的？
Orchestrator for end-to-end video localization: routes to transcribe → translate → dub → burn when the user asks for the full pipeline in one go.
Orchestrator for end-to-end video localization: routes to transcribe → translate → dub → burn when the user asks for the full pipeline in one go.
claude-skills is a Claude Code agent skill that orchestrator for end-to-end video localization: routes to transcribe → translate → dub → burn when the user asks for the full pipeline in one go.
npx skills add https://github.com/jianshuo/claude-skills/tree/main/wjs-localizing-videoOrchestrator for end-to-end video localization: routes to transcribe → translate → dub → burn when the user asks for the full pipeline in one go.
This repo contains 5 individual skills — each has its own dedicated page.
One-encode final localize cut: burn hardsubs (libass) and/or soft-mux, mix TTS dub over original audio bed. Auto-falls back to static ffmpeg with libass. Upload-ready MP4.
Time-aligned TTS dub from a target-language SRT: Volcano (豆包) for Chinese, edge-tts for any language; optional multi-speaker visual diarization. Outputs *_lang_dub.mp4 for burn/mix.
Agent skill at wjs-overlaying-video/SKILL.md
Source-language SRT from audio/video: Volcano ASR for Chinese, OpenAI Whisper word-level for other languages; punctuation-bounded cues capped for on-screen reading.
Translate source SRT to target language with punctuation-bounded re-segmentation so no cue ends mid-sentence; bilingual SRT option for review.
Extract frames or short clips from videos using ffmpeg.
Summarize or transcribe URLs, YouTube/videos, podcasts, articles, transcripts, PDFs, and local files.
Use when testing, fixing, or extending the OpenClaw Control UI GUI with Vitest + Playwright end-to-end checks, mocked Gateway WebSocket flows, mocked dashboard runs, screenshots/videos, or agent-verifiable browser proof.
Obra superpowers agent skills.
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work
Use when you have a spec or requirements for a multi-step task, before touching code