[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M70MMBC4NMNP6DC3KCG9GDFA":3,"skill-same:01M70MMBC4NMNP6DC3KCG9GDFA":73},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":51,"forksCount":52,"repositoryHtmlUrl":53,"repositoryHomepageUrl":54,"ownerHtmlUrl":55,"ownerBlogUrl":56,"ownerDescription":57,"updatedAt":58,"install":59,"resources":72},"01M70MMBC4NMNP6DC3KCG9GDFA","hyperframes-media",null,"Asset preprocessing for HyperFrames compositions — text-to-speech narration (Kokoro), audio\u002Fvideo transcription (Whisper), and background removal for transparent overlays (u2net). Use when generating voiceover from text, transcribing speech for captions, removing the background from a video or image to use as a transparent overlay, choosing a TTS voice or whisper model, or chaining these (TTS → transcribe → captions). Each command downloads its own model on first run.","shengdabai\u002FTony-Claude-Code-Skills","shengdabai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F116417419?v=4","01M70F6D2WY5PQV5A92BKYJQEW","Tony Sheng","01M70F6D3GAAAYZ5H4CC1G8VXM",3,"MIT","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fhyperframes-hyperframes-media\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"hyperframes media","hyperframes","Media","asset","video transcription","background removal","run","command","voiceover","whisper","text","image","preprocessing","when","overlays","model","overlay","Captions","Audio","Video","2026-11-10 10:12:30",false,"---\nname: hyperframes-media\ndescription: Asset preprocessing for HyperFrames compositions — text-to-speech narration (Kokoro), audio\u002Fvideo transcription (Whisper), and background removal for transparent overlays (u2net). Use when generating voiceover from text, transcribing speech for captions, removing the background from a video or image to use as a transparent overlay, choosing a TTS voice or whisper model, or chaining these (TTS → transcribe → captions). Each command downloads its own model on first run.\n---\n\n# HyperFrames Media Preprocessing\n\nThree CLI commands that produce assets for compositions: `tts` (speech), `transcribe` (timestamps), and `remove-background` (transparent video). Each downloads a model on first run and caches it under `~\u002F.cache\u002Fhyperframes\u002F`. Drop the output into the project, then reference it from the composition HTML — see the `hyperframes` skill for the audio\u002Fvideo element conventions.\n\n## Text-to-Speech (`tts`)\n\nGenerate speech audio locally with Kokoro-82M. No API key.\n\n```bash\nnpx hyperframes tts \"Text here\" --voice af_nova --output narration.wav\nnpx hyperframes tts script.txt --voice bf_emma --output narration.wav\nnpx hyperframes tts --list                       # all 54 voices\n```\n\n### Voice Selection\n\nMatch voice to content. Default is `af_heart`.\n\n| Content type      | Voice                 | Why                           |\n| ----------------- | --------------------- | ----------------------------- |\n| Product demo      | `af_heart`\u002F`af_nova`  | Warm, professional            |\n| Tutorial \u002F how-to | `am_adam`\u002F`bf_emma`   | Neutral, easy to follow       |\n| Marketing \u002F promo | `af_sky`\u002F`am_michael` | Energetic or authoritative    |\n| Documentation     | `bf_emma`\u002F`bm_george` | Clear British English, formal |\n| Casual \u002F social   | `af_heart`\u002F`af_sky`   | Approachable, natural         |\n\n### Multilingual\n\nVoice IDs encode language in the first letter: `a`=American English, `b`=British English, `e`=Spanish, `f`=French, `h`=Hindi, `i`=Italian, `j`=Japanese, `p`=Brazilian Portuguese, `z`=Mandarin. The CLI auto-detects the phonemizer locale from the prefix — no `--lang` needed when the voice matches the text.\n\n```bash\nnpx hyperframes tts \"La reunión empieza a las nueve\" --voice ef_dora --output es.wav\nnpx hyperframes tts \"今日はいい天気ですね\" --voice jf_alpha --output ja.wav\n```\n\nUse `--lang` only to override auto-detection (stylized accents). Valid codes: `en-us`, `en-gb`, `es`, `fr-fr`, `hi`, `it`, `pt-br`, `ja`, `zh`. Non-English phonemization requires `espeak-ng` system-wide (`brew install espeak-ng` \u002F `apt-get install espeak-ng`).\n\n### Speed\n\n- `0.7-0.8` — tutorial, complex content, accessibility\n- `1.0` — natural pace (default)\n- `1.1-1.2` — intros, transitions, upbeat content\n- `1.5+` — rarely appropriate; test carefully\n\n### Long Scripts\n\nFor more than a few paragraphs, write to a `.txt` file and pass the path. Inputs over ~5 minutes of speech may benefit from splitting into segments.\n\n### Requirements\n\nPython 3.8+ with `kokoro-onnx` and `soundfile` (`pip install kokoro-onnx soundfile`). Model downloads on first use (~311 MB + ~27 MB voices, cached in `~\u002F.cache\u002Fhyperframes\u002Ftts\u002F`).\n\n## Transcription (`transcribe`)\n\nProduce a normalized `transcript.json` with word-level timestamps.\n\n```bash\nnpx hyperframes transcribe audio.mp3\nnpx hyperframes transcribe video.mp4 --model small --language es\nnpx hyperframes transcribe subtitles.srt          # import existing\nnpx hyperframes transcribe subtitles.vtt\nnpx hyperframes transcribe openai-response.json\n```\n\n### Language Rule (Non-Negotiable)\n\n**Never use `.en` models unless the user explicitly states the audio is English.** `.en` models (`small.en`, `medium.en`) **translate** non-English audio into English instead of transcribing it. This silently destroys the original language.\n\n1. Language known and non-English → `--model small --language \u003Ccode>` (no `.en` suffix)\n2. Language known and English → `--model small.en`\n3. Language unknown → `--model small` (no `.en`, no `--language`) — whisper auto-detects\n\n**Default model is `small`, not `small.en`.**\n\n### Model Sizes\n\n| Model      | Size   | Speed    | When to use                           |\n| ---------- | ------ | -------- | ------------------------------------- |\n| `tiny`     | 75 MB  | Fastest  | Quick previews, testing pipeline      |\n| `base`     | 142 MB | Fast     | Short clips, clear audio              |\n| `small`    | 466 MB | Moderate | **Default** — most content            |\n| `medium`   | 1.5 GB | Slow     | Important content, noisy audio, music |\n| `large-v3` | 3.1 GB | Slowest  | Production quality                    |\n\nMusic with vocals: start at `medium` minimum; produced tracks often need manual SRT\u002FVTT import. For caption-quality checks (mandatory after every transcription), the cleaning JS, retry rules, and the OpenAI\u002FGroq API import path, see [hyperframes\u002Freferences\u002Ftranscript-guide.md](..\u002Fhyperframes\u002Freferences\u002Ftranscript-guide.md).\n\n### Output Shape\n\nCompositions consume a flat array of word objects. The `id` field (`w0`, `w1`, ...) is added during normalization for stable references in caption overrides; it's optional for backwards compatibility.\n\n```json\n[\n  { \"id\": \"w0\", \"text\": \"Hello\", \"start\": 0.0, \"end\": 0.5 },\n  { \"id\": \"w1\", \"text\": \"world.\", \"start\": 0.6, \"end\": 1.2 }\n]\n```\n\n## Background Removal (`remove-background`)\n\nRemove the background from a video or image so the subject (typically a person — avatar, presenter, talking head) sits as a transparent overlay in a composition.\n\n```bash\nnpx hyperframes remove-background subject.mp4 -o transparent.webm  # default: VP9 alpha WebM\nnpx hyperframes remove-background subject.mp4 -o transparent.mov   # ProRes 4444 (editing)\nnpx hyperframes remove-background portrait.jpg -o cutout.png       # single-image cutout\nnpx hyperframes remove-background subject.mp4 -o subject.webm \\\n  --background-output plate.webm                                   # both layers in one pass\nnpx hyperframes remove-background subject.mp4 -o transparent.webm --device cpu\nnpx hyperframes remove-background --info                           # detected providers\n```\n\nUses `u2net_human_seg` (MIT). First run downloads ~168 MB of weights to `~\u002F.cache\u002Fhyperframes\u002Fbackground-removal\u002Fmodels\u002F`.\n\n### Layer separation (`--background-output`)\n\nPass `--background-output` (or `-b`) to emit a **second** transparent video alongside the cutout: same source RGB, alpha is `255 − mask` instead of `mask`. The cutout is the subject with a transparent background; the plate is the original surroundings with a transparent hole where the subject was.\n\n| File                             | Alpha is…                                                 | Use it for                                                      |\n| -------------------------------- | --------------------------------------------------------- | --------------------------------------------------------------- |\n| `-o subject.webm`                | The mask — subject opaque, background transparent         | Foreground layer, place on top                                  |\n| `--background-output plate.webm` | Inverse — surroundings opaque, subject region transparent | Bottom layer; put text or graphics between this and the subject |\n\nBoth outputs share the same `--quality` preset and run from a single inference pass — encode cost roughly doubles, segmentation cost stays the same. Only valid for video inputs and `.webm`\u002F`.mov` outputs.\n\n**Hole-cut plate, not an inpainted clean plate.** The subject region in `plate.webm` is fully transparent — composite something opaque under it to fill the hole. The single test for whether `--background-output` is the right tool: _will anything ever be visible through the subject's silhouette where the subject used to be?_\n\n| Use case                                                                            | Right tool                                                                         |\n| ----------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------- |\n| Text\u002Fgraphics between the cutout and the plate (this command's reason for existing) | **Hole-cut** (`--background-output`)                                               |\n| Subject onto an unrelated scene                                                     | Just `subject.webm`; ignore the plate                                              |\n| Show the room _without_ the person, alone over no other content                     | **Clean plate** — needs an inpainter (LaMa, ProPainter, E2FGVI). Not this command. |\n| Replace the subject with a different subject                                        | **Clean plate** — same as above                                                    |\n\nIf a user asks for \"the room with the person removed\" and intends to display it standalone, do **not** reach for `--background-output`. Tell them they need an inpainter.\n\nTypical layered composition (the canonical hole-cut use case):\n\n```html\n\u003C!-- z=1 the inverse-alpha plate fills everything except the subject region -->\n\u003Cvideo\n  src=\"plate.webm\"\n  data-start=\"0\"\n  data-duration=\"6\"\n  data-track-index=\"0\"\n  muted\n  playsinline\n>\u003C\u002Fvideo>\n\n\u003C!-- z=2 graphics \u002F text live between the two layers -->\n\u003Ch1 id=\"headline\" style=\"z-index:2; ...\">MAKE IT IN HYPERFRAMES\u003C\u002Fh1>\n\n\u003C!-- z=3 the cutout floats the subject back over the headline -->\n\u003Cdiv class=\"cutout-wrap\" style=\"position:absolute;inset:0;z-index:3\">\n  \u003Cvideo\n    src=\"subject.webm\"\n    data-start=\"0\"\n    data-duration=\"6\"\n    data-track-index=\"1\"\n    muted\n    playsinline\n  >\u003C\u002Fvideo>\n\u003C\u002Fdiv>\n```\n\nThis is functionally equivalent to the text-behind-subject pattern below, but you don't need the original `presenter.mp4` in the project — the plate replaces it. Useful when you want to ship just the two transparent layers and let the user drop arbitrary content between them.\n\n### Output Format\n\n| Format                | When                                                          |\n| --------------------- | ------------------------------------------------------------- |\n| `.webm` (VP9 + alpha) | Default. Compositions play this directly via `\u003Cvideo>`.       |\n| `.mov` (ProRes 4444)  | Editing in DaVinci\u002FPremiere\u002FFCP. Large files.                 |\n| `.png`                | Single-image cutout (still subject, layered over a backdrop). |\n\nChrome decodes VP9 alpha natively, so the `.webm` plugs into a composition like any other muted-autoplay video — see the `hyperframes` skill for the `\u003Cvideo>` track conventions.\n\n### Quality presets\n\n`--quality fast|balanced|best` controls only the VP9 encoder's CRF — segmentation quality is fixed.\n\n| Preset     | CRF | When                                                  |\n| ---------- | --- | ----------------------------------------------------- |\n| `fast`     | 30  | Iterating, smaller file, looser color match           |\n| `balanced` | 18  | Default. Visually identical for most uses             |\n| `best`     | 12  | Master \u002F final delivery. Largest file, tightest match |\n\n### Compositing patterns — pick the right one\n\nThe cutout webm is a **re-encoded copy** of the source mp4's RGB. That choice has consequences depending on what you put behind it:\n\n| Pattern                                                  | What's behind the cutout                   | Result                                                                                                                                                                                                                            |\n| -------------------------------------------------------- | ------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| **Cutout over a different scene** (most common)          | Static image, gradient, or unrelated video | Looks great. The cutout's RGB is the only source of the subject — no doubling, no edge halo. This is what `remove-background` is built for.                                                                                       |\n| **Cutout over its own source mp4** (text-behind-subject) | Same mp4 the cutout was generated from     | Two RGB sources for the same person. At default `--quality balanced` (crf 18) the doubling is barely visible; at `--quality fast` (crf 30) you'll see a faint color shift \u002F edge halo. Use `--quality best` (crf 12) for masters. |\n| **Cutout over a _different_ take of the same person**    | Footage of the same subject                | Will look like two separate people overlapping. Don't do this.                                                                                                                                                                    |\n\n**Text-behind-subject** (headline behind a presenter):\n\n```html\n\u003Cvideo\n  src=\"presenter.mp4\"\n  id=\"bg\"\n  data-start=\"0\"\n  data-duration=\"6\"\n  data-track-index=\"0\"\n  muted\n  playsinline\n>\u003C\u002Fvideo>\n\u003Ch1 id=\"headline\" style=\"z-index:2; ...\">MAKE IT IN HYPERFRAMES\u003C\u002Fh1>\n\u003Cdiv class=\"cutout-wrap\" style=\"position:absolute;inset:0;z-index:3;opacity:0\">\n  \u003Cvideo\n    src=\"presenter.webm\"\n    data-start=\"0\"\n    data-duration=\"6\"\n    data-track-index=\"1\"\n    muted\n    playsinline\n  >\u003C\u002Fvideo>\n\u003C\u002Fdiv>\n```\n\nTwo key rules:\n\n1. **Wrap the cutout video in a non-timed `\u003Cdiv>`** and animate the wrapper's opacity, not the video element's. The framework forces opacity:1 on active clips (any element with `data-start`\u002F`data-duration`), so animating the video's opacity directly is silently overridden. The wrapper has no `data-*` attributes, so it's owned by your CSS\u002FGSAP.\n2. **Both videos use `data-start=\"0\"` and `data-media-start=\"0\"`** so the framework decodes them in sync from t=0. Late-mounting the cutout (`data-start=3.3`) introduces a seek + warm-up that lands a frame off the base mp4 — visible as one frame of misalignment at the cut.\n\nThen GSAP-flip the wrapper opacity at the cut: `tl.set(cutoutWrap, { opacity: 1 }, 3.3)`.\n\n## TTS → Transcribe → Captions\n\nWhen there's no pre-recorded voiceover, generate one and transcribe it back to get word-level timestamps for captions:\n\n```bash\nnpx hyperframes tts script.txt --voice af_heart --output narration.wav\nnpx hyperframes transcribe narration.wav   # → transcript.json\n```\n\nWhisper extracts precise word boundaries from the generated audio, so caption timing matches delivery without hand-tuning.\n","# HyperFrames Media Preprocessing\n\nThree CLI commands that produce assets for compositions: `tts` (speech), `transcribe` (timestamps), and `remove-background` (transparent video). Each downloads a model on first run and caches it under `~\u002F.cache\u002Fhyperframes\u002F`. Drop the output into the project, then reference it from the composition HTML — see the `hyperframes` skill for the audio\u002Fvideo element conventions.\n\n## Text-to-Speech (`tts`)\n\nGenerate speech audio locally with Kokoro-82M. No API key.\n\n```bash\nnpx hyperframes tts \"Text here\" --voice af_nova --output narration.wav\nnpx hyperframes tts script.txt --voice bf_emma --output narration.wav\nnpx hyperframes tts --list                       # all 54 voices\n```\n\n### Voice Selection\n\nMatch voice to content. Default is `af_heart`.\n\n| Content type      | Voice                 | Why                           |\n| ----------------- | --------------------- | ----------------------------- |\n| Product demo      | `af_heart`\u002F`af_nova`  | Warm, professional            |\n| Tutorial \u002F how-to | `am_adam`\u002F`bf_emma`   | Neutral, easy to follow       |\n| Marketing \u002F promo | `af_sky`\u002F`am_michael` | Energetic or authoritative    |\n| Documentation     | `bf_emma`\u002F`bm_george` | Clear British English, formal |\n| Casual \u002F social   | `af_heart`\u002F`af_sky`   | Approachable, natural         |\n\n### Multilingual\n\nVoice IDs encode language in the first letter: `a`=American English, `b`=British English, `e`=Spanish, `f`=French, `h`=Hindi, `i`=Italian, `j`=Japanese, `p`=Brazilian Portuguese, `z`=Mandarin. The CLI auto-detects the phonemizer locale from the prefix — no `--lang` needed when the voice matches the text.\n\n```bash\nnpx hyperframes tts \"La reunión empieza a las nueve\" --voice ef_dora --output es.wav\nnpx hyperframes tts \"今日はいい天気ですね\" --voice jf_alpha --output ja.wav\n```\n\nUse `--lang` only to override auto-detection (stylized accents). Valid codes: `en-us`, `en-gb`, `es`, `fr-fr`, `hi`, `it`, `pt-br`, `ja`, `zh`. Non-English phonemization requires `espeak-ng` system-wide (`brew install espeak-ng` \u002F `apt-get install espeak-ng`).\n\n### Speed\n\n- `0.7-0.8` — tutorial, complex content, accessibility\n- `1.0` — natural pace (default)\n- `1.1-1.2` — intros, transitions, upbeat content\n- `1.5+` — rarely appropriate; test carefully\n\n### Long Scripts\n\nFor more than a few paragraphs, write to a `.txt` file and pass the path. Inputs over ~5 minutes of speech may benefit from splitting into segments.\n\n### Requirements\n\nPython 3.8+ with `kokoro-onnx` and `soundfile` (`pip install kokoro-onnx soundfile`). Model downloads on first use (~311 MB + ~27 MB voices, cached in `~\u002F.cache\u002Fhyperframes\u002Ftts\u002F`).\n\n## Transcription (`transcribe`)\n\nProduce a normalized `transcript.json` with word-level timestamps.\n\n```bash\nnpx hyperframes transcribe audio.mp3\nnpx hyperframes transcribe video.mp4 --model small --language es\nnpx hyperframes transcribe subtitles.srt          # import existing\nnpx hyperframes transcribe subtitles.vtt\nnpx hyperframes transcribe openai-response.json\n```\n\n### Language Rule (Non-Negotiable)\n\n**Never use `.en` models unless the user explicitly states the audio is English.** `.en` models (`small.en`, `medium.en`) **translate** non-English audio into English instead of transcribing it. This silently destroys the original language.\n\n1. Language known and non-English → `--model small --language \u003Ccode>` (no `.en` suffix)\n2. Language known and English → `--model small.en`\n3. Language unknown → `--model small` (no `.en`, no `--language`) — whisper auto-detects\n\n**Default model is `small`, not `small.en`.**\n\n### Model Sizes\n\n| Model      | Size   | Speed    | When to use                           |\n| ---------- | ------ | -------- | ------------------------------------- |\n| `tiny`     | 75 MB  | Fastest  | Quick previews, testing pipeline      |\n| `base`     | 142 MB | Fast     | Short clips, clear audio              |\n| `small`    | 466 MB | Moderate | **Default** — most content            |\n| `medium`   | 1.5 GB | Slow     | Important content, noisy audio, music |\n| `large-v3` | 3.1 GB | Slowest  | Production quality                    |\n\nMusic with vocals: start at `medium` minimum; produced tracks often need manual SRT\u002FVTT import. For caption-quality checks (mandatory after every transcription), the cleaning JS, retry rules, and the OpenAI\u002FGroq API import path, see [hyperframes\u002Freferences\u002Ftranscript-guide.md](..\u002Fhyperframes\u002Freferences\u002Ftranscript-guide.md).\n\n### Output Shape\n\nCompositions consume a flat array of word objects. The `id` field (`w0`, `w1`, ...) is added during normalization for stable references in caption overrides; it's optional for backwards compatibility.\n\n```json\n[\n  { \"id\": \"w0\", \"text\": \"Hello\", \"start\": 0.0, \"end\": 0.5 },\n  { \"id\": \"w1\", \"text\": \"world.\", \"start\": 0.6, \"end\": 1.2 }\n]\n```\n\n## Background Removal (`remove-background`)\n\nRemove the background from a video or image so the subject (typically a person — avatar, presenter, talking head) sits as a transparent overlay in a composition.\n\n```bash\nnpx hyperframes remove-background subject.mp4 -o transparent.webm  # default: VP9 alpha WebM\nnpx hyperframes remove-background subject.mp4 -o transparent.mov   # ProRes 4444 (editing)\nnpx hyperframes remove-background portrait.jpg -o cutout.png       # single-image cutout\nnpx hyperframes remove-background subject.mp4 -o subject.webm \\\n  --background-output plate.webm                                   # both layers in one pass\nnpx hyperframes remove-background subject.mp4 -o transparent.webm --device cpu\nnpx hyperframes remove-background --info                           # detected providers\n```\n\nUses `u2net_human_seg` (MIT). First run downloads ~168 MB of weights to `~\u002F.cache\u002Fhyperframes\u002Fbackground-removal\u002Fmodels\u002F`.\n\n### Layer separation (`--background-output`)\n\nPass `--background-output` (or `-b`) to emit a **second** transparent video alongside the cutout: same source RGB, alpha is `255 − mask` instead of `mask`. The cutout is the subject with a transparent background; the plate is the original surroundings with a transparent hole where the subject was.\n\n| File                             | Alpha is…                                                 | Use it for                                                      |\n| -------------------------------- | --------------------------------------------------------- | --------------------------------------------------------------- |\n| `-o subject.webm`                | The mask — subject opaque, background transparent         | Foreground layer, place on top                                  |\n| `--background-output plate.webm` | Inverse — surroundings opaque, subject region transparent | Bottom layer; put text or graphics between this and the subject |\n\nBoth outputs share the same `--quality` preset and run from a single inference pass — encode cost roughly doubles, segmentation cost stays the same. Only valid for video inputs and `.webm`\u002F`.mov` outputs.\n\n**Hole-cut plate, not an inpainted clean plate.** The subject region in `plate.webm` is fully transparent — composite something opaque under it to fill the hole. The single test for whether `--background-output` is the right tool: _will anything ever be visible through the subject's silhouette where the subject used to be?_\n\n| Use case                                                                            | Right tool                                                                         |\n| ----------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------- |\n| Text\u002Fgraphics between the cutout and the plate (this command's reason for existing) | **Hole-cut** (`--background-output`)                                               |\n| Subject onto an unrelated scene                                                     | Just `subject.webm`; ignore the plate                                              |\n| Show the room _without_ the person, alone over no other content                     | **Clean plate** — needs an inpainter (LaMa, ProPainter, E2FGVI). Not this command. |\n| Replace the subject with a different subject                                        | **Clean plate** — same as above                                                    |\n\nIf a user asks for \"the room with the person removed\" and intends to display it standalone, do **not** reach for `--background-output`. Tell them they need an inpainter.\n\nTypical layered composition (the canonical hole-cut use case):\n\n```html\n\u003C!-- z=1 the inverse-alpha plate fills everything except the subject region -->\n\u003Cvideo\n  src=\"plate.webm\"\n  data-start=\"0\"\n  data-duration=\"6\"\n  data-track-index=\"0\"\n  muted\n  playsinline\n>\u003C\u002Fvideo>\n\n\u003C!-- z=2 graphics \u002F text live between the two layers -->\n\u003Ch1 id=\"headline\" style=\"z-index:2; ...\">MAKE IT IN HYPERFRAMES\u003C\u002Fh1>\n\n\u003C!-- z=3 the cutout floats the subject back over the headline -->\n\u003Cdiv class=\"cutout-wrap\" style=\"position:absolute;inset:0;z-index:3\">\n  \u003Cvideo\n    src=\"subject.webm\"\n    data-start=\"0\"\n    data-duration=\"6\"\n    data-track-index=\"1\"\n    muted\n    playsinline\n  >\u003C\u002Fvideo>\n\u003C\u002Fdiv>\n```\n\nThis is functionally equivalent to the text-behind-subject pattern below, but you don't need the original `presenter.mp4` in the project — the plate replaces it. Useful when you want to ship just the two transparent layers and let the user drop arbitrary content between them.\n\n### Output Format\n\n| Format                | When                                                          |\n| --------------------- | ------------------------------------------------------------- |\n| `.webm` (VP9 + alpha) | Default. Compositions play this directly via `\u003Cvideo>`.       |\n| `.mov` (ProRes 4444)  | Editing in DaVinci\u002FPremiere\u002FFCP. Large files.                 |\n| `.png`                | Single-image cutout (still subject, layered over a backdrop). |\n\nChrome decodes VP9 alpha natively, so the `.webm` plugs into a composition like any other muted-autoplay video — see the `hyperframes` skill for the `\u003Cvideo>` track conventions.\n\n### Quality presets\n\n`--quality fast|balanced|best` controls only the VP9 encoder's CRF — segmentation quality is fixed.\n\n| Preset     | CRF | When                                                  |\n| ---------- | --- | ----------------------------------------------------- |\n| `fast`     | 30  | Iterating, smaller file, looser color match           |\n| `balanced` | 18  | Default. Visually identical for most uses             |\n| `best`     | 12  | Master \u002F final delivery. Largest file, tightest match |\n\n### Compositing patterns — pick the right one\n\nThe cutout webm is a **re-encoded copy** of the source mp4's RGB. That choice has consequences depending on what you put behind it:\n\n| Pattern                                                  | What's behind the cutout                   | Result                                                                                                                                                                                                                            |\n| -------------------------------------------------------- | ------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |\n| **Cutout over a different scene** (most common)          | Static image, gradient, or unrelated video | Looks great. The cutout's RGB is the only source of the subject — no doubling, no edge halo. This is what `remove-background` is built for.                                                                                       |\n| **Cutout over its own source mp4** (text-behind-subject) | Same mp4 the cutout was generated from     | Two RGB sources for the same person. At default `--quality balanced` (crf 18) the doubling is barely visible; at `--quality fast` (crf 30) you'll see a faint color shift \u002F edge halo. Use `--quality best` (crf 12) for masters. |\n| **Cutout over a _different_ take of the same person**    | Footage of the same subject                | Will look like two separate people overlapping. Don't do this.                                                                                                                                                                    |\n\n**Text-behind-subject** (headline behind a presenter):\n\n```html\n\u003Cvideo\n  src=\"presenter.mp4\"\n  id=\"bg\"\n  data-start=\"0\"\n  data-duration=\"6\"\n  data-track-index=\"0\"\n  muted\n  playsinline\n>\u003C\u002Fvideo>\n\u003Ch1 id=\"headline\" style=\"z-index:2; ...\">MAKE IT IN HYPERFRAMES\u003C\u002Fh1>\n\u003Cdiv class=\"cutout-wrap\" style=\"position:absolute;inset:0;z-index:3;opacity:0\">\n  \u003Cvideo\n    src=\"presenter.webm\"\n    data-start=\"0\"\n    data-duration=\"6\"\n    data-track-index=\"1\"\n    muted\n    playsinline\n  >\u003C\u002Fvideo>\n\u003C\u002Fdiv>\n```\n\nTwo key rules:\n\n1. **Wrap the cutout video in a non-timed `\u003Cdiv>`** and animate the wrapper's opacity, not the video element's. The framework forces opacity:1 on active clips (any element with `data-start`\u002F`data-duration`), so animating the video's opacity directly is silently overridden. The wrapper has no `data-*` attributes, so it's owned by your CSS\u002FGSAP.\n2. **Both videos use `data-start=\"0\"` and `data-media-start=\"0\"`** so the framework decodes them in sync from t=0. Late-mounting the cutout (`data-start=3.3`) introduces a seek + warm-up that lands a frame off the base mp4 — visible as one frame of misalignment at the cut.\n\nThen GSAP-flip the wrapper opacity at the cut: `tl.set(cutoutWrap, { opacity: 1 }, 3.3)`.\n\n## TTS → Transcribe → Captions\n\nWhen there's no pre-recorded voiceover, generate one and transcribe it back to get word-level timestamps for captions:\n\n```bash\nnpx hyperframes tts script.txt --voice af_heart --output narration.wav\nnpx hyperframes transcribe narration.wav   # → transcript.json\n```\n\nWhisper extracts precise word boundaries from the generated audio, so caption timing matches delivery without hand-tuning.\n",{"name":5,"description":7},"main","7850740bd837adb32f0a109ab2af82482196c6f2","skills\u002Fhyperframes-hyperframes-media",1,"https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills","https:\u002F\u002Fzturnsgo.com","https:\u002F\u002Fgithub.com\u002Fshengdabai","https:\u002F\u002Fzturnsgo.com\u002F","AI × education creator — producer, not consumer. Taught 6,000+ learners; shipping learning tools & daily essays with Claude Code, practicing @xiaolai's way 🥋","2026-11-01 21:28:23",{"source":8,"skillName":5,"commands":60,"prompt":70,"zipAvailable":71},[61,64,67],{"tool":62,"command":63},"npx","npx skills add shengdabai\u002FTony-Claude-Code-Skills --skill hyperframes-media",{"tool":65,"command":66},"bunx","bunx skills add shengdabai\u002FTony-Claude-Code-Skills --skill hyperframes-media",{"tool":68,"command":69},"pnpm","pnpm dlx skills add shengdabai\u002FTony-Claude-Code-Skills --skill hyperframes-media","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fhyperframes-hyperframes-media\u002FSKILL.md",true,[],[74,83,91,99,107,115],{"skillNo":75,"name":76,"nameZh":6,"description":77,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":78,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":82},"01M73H7XKBHE1ETQDZT94MF9HH","seedance-prompt-en","Write effective prompts for Jimeng Seedance 2.0 multimodal AI video generation. Use when users want to create video prompts using text, images, videos, and audio inputs with the @ reference system. Covers camera movements, effects replication, video extension, editing, music beat-matching, e-commerce ads, short dramas, and educational content.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fseedance-prompt-en\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fseedance-prompt-en\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":87,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":82},"01M73GRE9998Q1DZWC71N41JWG","onepage-pdf","Convert an HTML page into a single continuous-page PDF (one tall page, no pagination breaks), preserving desktop layout, backgrounds and selectable text. Use when the user asks to turn an HTML file\u002Freport\u002Fproposal into a \"single-page PDF\", \"long PDF\", \"不分页 PDF\", \"单页 PDF\", \"长图式 PDF\", or complains that an HTML-to-PDF export breaks into pages or cuts content. Supports optional string redaction with leak verification.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fonepage-pdf\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fonepage-pdf\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":95,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":82},"01M73GQG1KZDADVPV3MHAT1J5A","native-feel-cross-platform-desktop","Use when the user is designing, prototyping, or rewriting a desktop app that must run on multiple OSes (macOS + Windows, optionally Linux) AND feel indistinguishable from a native app to its users — fast launch, native windowing, native input handling, native materials. Trigger words include \"cross-platform desktop\", \"Electron alternative\", \"Tauri vs native\", \"WebView wrapper\", \"near-native performance\", \"Raycast architecture\", \"WebKit\u002FWebView2 quirks\", \"WKWebView\", \"system tray app\", \"global hotkey app\", \"launcher app\". Do NOT trigger this skill for pure web apps, pure mobile apps, or for greenfield projects that have no native-feel requirement.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fnative-feel-cross-platform-desktop\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fnative-feel-cross-platform-desktop\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":103,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":82},"01M73G87PDNSMKG2FHWCYER6N8","gstack","Router for the gstack skill suite. (gstack)","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fgstack\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fgstack\u002FSKILL.md",[],[],{"skillNo":108,"name":109,"nameZh":6,"description":110,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":111,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":82},"01M73G3EAQWKCM8DB8D0N0QSVG","fireworks-tech-graph","Use when the user wants to create any technical diagram - architecture, data flow, flowchart, sequence, agent\u002Fmemory, or concept map - and export as SVG+PNG. Trigger on: \"画图\" \"帮我画\" \"生成图\" \"做个图\" \"架构图\" \"流程图\" \"可视化一下\" \"出图\" \"generate diagram\" \"draw diagram\" \"visualize\" or any system\u002Fflow description the user wants illustrated.","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Ffireworks-tech-graph\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Ffireworks-tech-graph\u002FSKILL.md",[],[],{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":82},"01M73G086AXAH9DW1H5G5N3CPB","dbs-video-extract","接收抖音、小红书、微信视频号等短视频链接或分享文案，查询作品／账号数据并提取语音文字稿，按作者和标题归档为 Markdown。用户希望完整解析短视频、取得数据或文字稿时使用；运行前检查 TikHub 与轻抖使用凭证，缺少时用新手能理解的方式引导购买和安全配置。","https:\u002F\u002Fgithub.com\u002Fshengdabai\u002FTony-Claude-Code-Skills\u002Fblob\u002Fmain\u002Fskills\u002Fdbs-video-extract\u002FSKILL.md",[],[]]