菜单
Translate video subtitles to any language with native-quality refinement. Full pipeline: transcribe → translate → refine → embed RTL-safe subtitles. Use for: translate video, תרגם סרטון, video translation, foreign subtitles, Hebrew subtitles, translated captions.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
WhatsApp automation (Green API primary, WAHA fallback). Use for: send/receive messages, send voice/images/files, retrieve incoming messages (text, voice, images), transcribe voice messages. WAME = 'WhatsApp ME' = send me by WhatsApp.
Spin up an instant browser voice session (OpenAI Realtime gpt-realtime-2) to close a topic in a short conversation instead of working through documents. Generic & white-label - works for any process. Supports live data work (read/update files, JSON, run commands), and distill mode (no tools, ends with a structured deliverable). Has a generic canvas that can display images, markdown, code, html, json, video, audio - perfect for "let's go over X" flows where the agent shows you items one by one and you react in real time. Use when user says "let's close this in a voice call", "run a quick voice session about X", "תפעיל שיחה קולית", "let's go over the [images/leads/PRs/files/notes]", or when a task is faster as a 3-minute conversation than as a document edit.
The 10x10 method — generate breadth, then converge with human judgment. Use whenever a single AI output won't nail it and quality matters (design, copy, naming, posters, messaging, strategy options, code approaches), OR when the user says '10x10', 'ten by ten', 'give me 10 options', 'show me variations', or asks to refine/tighten an output instead of round-after-round corrections.
Never teach a principle without an immediate rep. When designing ANY teaching content — a workshop, a lesson plan, a deck, a course module, a webinar, an explainer video, an educational post — pair every theoretical unit with an embodiment the learner does right now. Use whenever building or reviewing teaching material, lesson plans, workshop flows, course outlines, or training content, OR when the user says 'teach-by-doing', 'add an exercise', 'make it hands-on', 'don't leave it abstract', 'תרגיל לכל עיקרון', 'הטמעה מיידית'.
Inject creative absurdity to break AI pattern-thinking before creative tasks. Randomly samples surreal micro-stories to spark unconventional thinking. Use before brainstorming, naming, writing, or any task needing fresh perspective.
Deploy static or interactive frontend content to GitHub Pages using gh CLI. Use when the user wants to publish, share, or make accessible any HTML/CSS/JS content - including demos, prototypes, visualizations, landing pages, portfolios, documentation, interactive tools, games, or any browser-based project. Activate whenever content needs to be publicly viewable via URL, not just when "website" is explicitly mentioned.
| name | translate-video |
| description | Translate video subtitles to any language with native-quality refinement. Full pipeline: transcribe → translate → refine → embed RTL-safe subtitles. Use for: translate video, תרגם סרטון, video translation, foreign subtitles, Hebrew subtitles, translated captions. |
| argument-hint | [video-path] [target-language] [--shorts|--regular] |
End-to-end video translation pipeline: transcribe → translate → refine subtitles → embed.
/translate-video /path/to/video.mp4 he --regular
/translate-video /path/to/video.mp4 he --shorts
$1 — video file path (required)$2 — target language code (default: he). See references/languages.md$3 — --shorts (TikTok/Reels) or --regular (YouTube/tutorials). If omitted, ask the user.If audio > 25MB, extract first:
ffmpeg -i "$VIDEO" -vn -acodec libmp3lame -ab 128k "$AUDIO.mp3" -y
Transcribe with word-level JSON (always include --json):
cd ~/.claude/skills/transcribe/scripts && [ -d node_modules ] || npm install --silent
npx ts-node transcribe.ts -i "$INPUT" -o "$BASENAME.srt" --json
Produces: {basename}.srt, {basename}.md, {basename}_transcript.json
Read .md for full context. Translate the .srt — preserve all timestamps and index numbers exactly. See translation rules in references/modes.md.
Read references/modes.md for full rules.
--shorts: Fix text only, preserve all timestamps. No merging.
--regular: Merge into full sentences using word-level timestamps.
.md (word counts per group)GROUP_SIZES in scripts/build-timestamps.py and run itEnforces MAX 2 lines, MAX chars/line (38 for --shorts, 42 for --regular):
python3 ~/.claude/skills/translate-video/scripts/postprocess.py "$SRT" 42
python3 ~/.claude/skills/translate-video/scripts/rtl-fix.py "$SRT"
~/.local/bin/ffmpeg-ass -i "$VIDEO" \
-vf "subtitles=$SRT:force_style='FontSize=24,PrimaryColour=&H00FFFFFF,OutlineColour=&H00000000,Outline=2,Shadow=1,Alignment=2,MarginV=30'" \
-c:v libx264 -preset fast -crf 23 -c:a copy "$OUTPUT" -y
open "$OUTPUT"
⚠️ Do NOT use Docker ffmpeg for long videos on ARM Mac — x86 emulation is ~100x slower.
| File | Description |
|---|---|
{name}.srt | Original language SRT |
{name}.md | Readable transcript |
{name}_transcript.json | Word-level timestamps |
{name}_{lang}.srt | Translated + refined SRT |
{name}_{lang}_subtitled.mp4 | Final video |
| File | Purpose |
|---|---|
| references/modes.md | Detailed --shorts and --regular rules |
| references/languages.md | Language codes + RTL flags |
| scripts/build-timestamps.py | Word-index cursor for --regular timestamps |
| scripts/postprocess.py | Enforce line limits on any SRT |
| scripts/rtl-fix.py | Apply RTL Unicode markers |