[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7473DX213Z1AFXMYQYCJWVF":3,"skill-same:01M7473DX213Z1AFXMYQYCJWVF":58},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":35,"ownerDescription":36,"updatedAt":37,"install":38,"resources":51},"01M7473DX213Z1AFXMYQYCJWVF","writing-video-prompts",null,"Writes the motion prompt for a video clip — what moves, how the camera behaves, the physics and the audio, in the shape the chosen model wants. Works from a brief alone, from a start frame, from a source video, or from a storyboard sheet where each panel becomes a cut. Handles one clip or a whole set, returning one prompt per clip. Nothing is generated here. Use when a clip's model, duration and content are settled and it needs the prompt written.","bg-szy\u002FTOP-SKILLS","bg-szy","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F106576190?v=4","01M73A4E8KPV67SFXRY8XZGNBX","ShuZhiyong","01M73A4E96MQ294MHM74TZBVQD",6,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fwriting-video-prompts__skills-supercmohq-writing-video-prompts\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: writing-video-prompts\ndescription: Writes the motion prompt for a video clip — what moves, how the camera behaves, the physics and the audio, in the shape the chosen model wants. Works from a brief alone, from a start frame, from a source video, or from a storyboard sheet where each panel becomes a cut. Handles one clip or a whole set, returning one prompt per clip. Nothing is generated here. Use when a clip's model, duration and content are settled and it needs the prompt written.\nlicense: Apache-2.0\nmetadata:\n  version: \"0.5.0\"\n  category: creative\n  summary: \"Translates your scenes into the highly specific, detailed instructions that different video models require. It times the camera motion and steers around known AI rendering limits to match your exact visual brief.\"\n---\n\n# Clip prompts\n\nTurn a clip into a prompt its video model will honour. **Text only — nothing is generated here.**\n\n**A video may be one clip or several.** Write a prompt for every clip in it, then hand them all back\ntogether. Clips in one video share a model, an aspect ratio and a look.\n\nWrite only what the clip actually has. No person on camera means no performance direction, no product\nmeans no product rules, no speech means no dialogue.\n\n## Workflow\n\n### Step 1: Read what the clip has\n\nAsk only for what would change the prompt and can't be defaulted.\n\n- **The model and the duration.** Where no model was named, use `seedance-2.0-fast` and say so. The\n  aspect ratio is a call parameter rather than prompt text — carry it if it was given, don't ask.\n- **What happens in the clip** — as beats, or as a single action.\n- **The media, and what each one is for** — a storyboard sheet, a start frame, an end frame,\n  reference images, a source video, an audio track. **Any of them may be absent**, and they differ from\n  clip to clip.\n- **The dialog segment**, where anyone speaks, and whether they are on camera or heard over the\n  picture.\n\n### Step 2: Pick the clip's form\n\n**With a storyboard sheet, one cut per panel** — panel 1 is the first cut, panel 2 the second, hard\ncut between them. \n\nWithout a sheet, cut only where the brief asks for one.\n\n**Where a beat's opening frame is fixed** — by a panel or by a start frame — the hands, the grip, the\nproduct, the props and the positions are already there when it opens. Write what happens next, not\nhow it starts.\nWrite \"a hand reaches in\" or \"she brings it into frame\" only when the frame genuinely has no such\nelement. Introduce something that is already there and the model renders a second one — an extra\nhand, a duplicate arm, a doubled person.\n\n### Step 3: Read the chosen model's guide\n\n| Model | Guide |\n| --- | --- |\n| `seedance-2.0` | `references\u002Fprompt-seedance.md` |\n| `veo-3.1` | `references\u002Fprompt-veo.md` |\n| `gemini-omni` | `references\u002Fprompt-gemini-omni.md` |\n| `kling-3.0-pro` | `references\u002Fprompt-kling.md` |\n| `grok-imagine-video` | `references\u002Fprompt-grok.md` |\n| `wan-2.7` | `references\u002Fprompt-wan.md` |\n\nThe `-fast` \u002F `-lite` \u002F `-turbo` \u002F `-standard` \u002F `-mini` variants share their family's guide.\n\n**The guide owns how the prompt is laid out** — the order the elements go in, how references are\nlabelled, how speech and sound are marked, and how to keep something out of frame. Follow it on all\nof those, over anything the steps below imply about layout.\n\n**It does not decide whether the clip cuts.** Step 2 already did, from the media. A guide showing how\nto write a sequence of shots is giving you the layout to use *if* the clip has one — not a reason to\nadd cuts to a clip that doesn't.\n\n### Step 4: Label the media\n\nWhere more than one piece of media goes on the call, name each one in the prompt, in the order it\nwill be passed, and say once what each is. Unlabelled, the model picks its own subject out of the\nset. **The labelling convention comes from the model's guide** — a numbered slot, a tag, or plain\nwords. A clip carrying a single start frame and nothing else has nothing to label — just say what it\nopens on.\n\n- **A storyboard sheet** — labelled once at the top as the storyboard for this clip. Don't contradict\n  it and don't caption it: a sentence that would work as a caption for the panel is a description of\n  a still, so write that panel's beat in motion instead.\n- **A start or end frame** — named as the frame the clip opens or closes on.\n- **The person** — labelled on first mention in a beat, then the plain noun. Repeat the label later\n  in the same beat and the render snaps back to the reference, breaking continuity.\n- **The product** — label it where it is shown intact: first introduction, a hero close-up, a label\n  point-out. Drop the label during any verb that changes it — opening, tearing, pouring, scooping,\n  applying, drinking, mixing, putting on — because carrying it through forces the product back to its\n  reference state mid-beat, so an opened jar snaps shut. Pick the label up again once it is shown\n  intact.\n- **Several product images** — point at whichever shows the angle that beat needs.\n\n**Never put a label inside quoted speech.** Quoted text is read aloud as written, so the label gets\nspoken.\n\n### Step 5: Read the storyboard\n\nSkip this where the clip has no sheet.\n\nRun `image_analysis` on the sheet before writing anything, and take these from each panel as it\nactually rendered:\n\n- the camera position — held at arm's length, locked off, over the shoulder\n- the framing distance\n- the action\n- which hand is doing what\n- where the product is — held, set down, worn, hidden, absent\n- the expression\n\n**Where the sheet and the plan disagree, the sheet is what the model will see, so the sheet wins.**\nWrite the cut to the panel, and say which panel disagreed and how.\n\n### Step 6: Write the clip\n\nWrite 6a to 6d, laid out in the order the model's guide sets out.\n\n**6a — The look.** One line: the light, and how the camera behaves across the clip. Add the delivery\nregister where anyone speaks.\n\n**6b — The beats**, in order.\n\n- **Each cut is one or two sentences, carrying one verb that moves something.** A second moving\n  verb is a second cut.\n- **With cuts**, head each with its number and its span — `Cut 2 (3.5-7s):` — and end every cut but\n  the last with a hard-cut instruction on its own line. Both go in whatever the guide says about\n  timings: the spans are what make the cuts add up to the clip's length, and without the hard-cut\n  line the model runs the beats together as one continuous move.\n- **Without cuts, time the beats only where the model's guide uses timings.** Some models read a\n  written span; others are made unstable by one and want the beats in order with the pacing following\n  the action.\n- **How long each beat is.** Each one runs two to four seconds of the clip, and they add up to its\n  length exactly.\n- **The first beat opens already underway** rather than on a held pose. Where a frame is fixed, it\n  opens exactly on what that shows — nothing enters and nothing changes state before the motion\n  starts. Where a person is on camera, that means hands already moving or a reaction already on the\n  face, and the first words landing at the top of the beat rather than after a pause.\n\n**6c — The camera, in every beat.** Where a sheet or a start frame already fixes what is in frame and\nwhere the camera sits, only how it moves is yours — a push in, a pull back, a tilt, a pan, a rack\nfocus, a handheld sway. Without one, write the framing and the position too. Locked off means the\nframe doesn't move and everything moves inside it, so keep any word suggesting an unsteady grip or a\ntravelling lens out of those beats. At arm's length the camera is the phone and carries a small\nnatural unsteadiness; nothing showing a handset, a screen, or either reflected belongs in shot.\n\n**6d — The person, where one is on camera.**\n\n- **Hands, counted at every moment.** A phone held at arm's length takes up one hand, so a single hand\n  is free to act; a locked-off shot leaves both. Describe a third job — hold one thing, work a second,\n  point at a third — and a third hand appears. Where hands don't matter to the beat, leave them out.\n  When an arm's-length beat calls for two working hands, it can't be shot as framed — send it back to\n  the storyboard to re-frame that beat rather than working around it in the prompt.\n- **Performance.** Several small movements within each beat, and at least one thing that changes —\n  weight moving forward, a breath landing, a grin breaking after a pause. A beat with no change inside\n  it reads as a frozen frame. Pitch each one higher than the level you want on screen, since the model\n  plays performance back flatter than it is written. Vary the expression from beat to beat. Where they\n  speak, let the delivery slip and recover once somewhere in the clip.\n\n**Every beat is bounded by these:**\n\n- **One action per beat** — one continuous motion at one object. Chained verbs get dropped or mangled.\n- **Nothing below limb scale.** Moving, holding, tilting, lifting, gesturing and walking render.\n  Unscrewing a cap, pulling a drawstring, pumping a pump head, working a zip or a clasp do not. Where\n  a beat needs the thing in a new state, open it with the thing already in that state.\n- **Interactions respect the physical chain.** Nobody drinks from a sealed bottle. Where the whole\n  chain won't fit, open the clip further along it.\n- **Positions hold through a beat.** Where something is, how it's held, how it's worn stays put unless\n  a visible movement changes it. An unstated change renders as a teleport, and once changed it stays\n  changed.\n- **Describe only what the media show.** Texture, back panels, interiors and hidden mechanisms\n  outside the frame come back invented.\n- **Nothing can be written as absent.** A thing named is a thing rendered, whether object, action or\n  sound.\n\nEvery piece of media also goes to the model on the call, so don't restate what it shows — the same\ninstruction twice renders a slideshow. But write the prompt full: written thin, the image wins and\nthe clip comes back as held poses with pauses between them.\n\n### Step 7: Write the audio\n\nSkip this where the clip carries no sound.\n\nWhere anyone speaks, split the clip's script segment across the beats at natural phrase breaks,\nroughly in proportion to their lengths. The words themselves are exact — no rewording, no insertions,\nno deletions — and the pieces must reassemble into the whole segment with nothing heard twice.\n**Mark the speech the way the model's guide marks it**; that marking is what drives lip movement.\n\nThe audio line says how the speech is produced and what else is heard. On camera: the person speaking\nin their own accent, lip-synced to the segments above. As voiceover: spoken off-screen, no mouth\nmoving. Say what the recording sounds like, following the look the plan named — a phone-shot piece\nmeans phone-microphone audio and room tone.\n\nWhere anyone speaks on camera, give each beat a moment with the mouth closed while the hands work. Lip\nmovement is the weakest thing the model renders, and continuous speech across a clip smears it.\n\nAmbience is room tone plus what the beat's action actually makes. Name the source and the surface it\nhappens on, keep it to two or three sounds at once, and never write a sound as absent — a sound named\nin a beat whose action doesn't produce it makes the model add the action to match. Don't repeat spoken\nwords in the ambience line, and don't name or imply music unless the user asked for it.\n\n### Step 8: Hand the prompts back\n\nWrite every clip's prompt before any of them is generated. They don't depend on each other, and a\nprompt is far cheaper to fix on paper than after it has rendered.\n\nHand back one prompt per clip, in order, each tied to its clip, along with the media that go with it\nand the order they were labelled in.\n\n## Edge cases\n\n- **No beats were supplied, only a brief** → write the clip as one continuous shot and split it into\n  beats yourself, to the seconds in Step 6.\n- **The beats need more time than the clip has** → say so and hand back the beats that fit. Don't\n  compress them into fewer seconds; the extra ones get dropped or mangled in the render.\n- **A start frame arrives alongside reference images** → they cannot be sent on one call. Say which\n  the clip should keep, and write the prompt for that form.\n- **The model is one with no guide here** → call `list_video_models` for its accepted inputs, and\n  write to the guide of the family it most resembles.\n\n## Reference\n\n**Model prompt guides — read the one for the chosen model:**\n\n- `references\u002Fprompt-seedance.md` — native audio markers, one-move-per-shot, storyboards.\n- `references\u002Fprompt-veo.md` — cinematography-first order, `SFX:` \u002F `Ambient noise:` markers, timecodes.\n- `references\u002Fprompt-gemini-omni.md` — conversational editing of an existing clip; edits, not extends.\n- `references\u002Fprompt-kling.md` — director-script style, `@Element` references, per-speaker dialogue labels.\n- `references\u002Fprompt-grok.md` — tight front-loaded briefs, positive-only exclusions, one action per clip.\n- `references\u002Fprompt-wan.md` — front-loaded prose, multi-shot timestamps, driving-audio sync.\n","# Clip prompts\n\nTurn a clip into a prompt its video model will honour. **Text only — nothing is generated here.**\n\n**A video may be one clip or several.** Write a prompt for every clip in it, then hand them all back\ntogether. Clips in one video share a model, an aspect ratio and a look.\n\nWrite only what the clip actually has. No person on camera means no performance direction, no product\nmeans no product rules, no speech means no dialogue.\n\n## Workflow\n\n### Step 1: Read what the clip has\n\nAsk only for what would change the prompt and can't be defaulted.\n\n- **The model and the duration.** Where no model was named, use `seedance-2.0-fast` and say so. The\n  aspect ratio is a call parameter rather than prompt text — carry it if it was given, don't ask.\n- **What happens in the clip** — as beats, or as a single action.\n- **The media, and what each one is for** — a storyboard sheet, a start frame, an end frame,\n  reference images, a source video, an audio track. **Any of them may be absent**, and they differ from\n  clip to clip.\n- **The dialog segment**, where anyone speaks, and whether they are on camera or heard over the\n  picture.\n\n### Step 2: Pick the clip's form\n\n**With a storyboard sheet, one cut per panel** — panel 1 is the first cut, panel 2 the second, hard\ncut between them. \n\nWithout a sheet, cut only where the brief asks for one.\n\n**Where a beat's opening frame is fixed** — by a panel or by a start frame — the hands, the grip, the\nproduct, the props and the positions are already there when it opens. Write what happens next, not\nhow it starts.\nWrite \"a hand reaches in\" or \"she brings it into frame\" only when the frame genuinely has no such\nelement. Introduce something that is already there and the model renders a second one — an extra\nhand, a duplicate arm, a doubled person.\n\n### Step 3: Read the chosen model's guide\n\n| Model | Guide |\n| --- | --- |\n| `seedance-2.0` | `references\u002Fprompt-seedance.md` |\n| `veo-3.1` | `references\u002Fprompt-veo.md` |\n| `gemini-omni` | `references\u002Fprompt-gemini-omni.md` |\n| `kling-3.0-pro` | `references\u002Fprompt-kling.md` |\n| `grok-imagine-video` | `references\u002Fprompt-grok.md` |\n| `wan-2.7` | `references\u002Fprompt-wan.md` |\n\nThe `-fast` \u002F `-lite` \u002F `-turbo` \u002F `-standard` \u002F `-mini` variants share their family's guide.\n\n**The guide owns how the prompt is laid out** — the order the elements go in, how references are\nlabelled, how speech and sound are marked, and how to keep something out of frame. Follow it on all\nof those, over anything the steps below imply about layout.\n\n**It does not decide whether the clip cuts.** Step 2 already did, from the media. A guide showing how\nto write a sequence of shots is giving you the layout to use *if* the clip has one — not a reason to\nadd cuts to a clip that doesn't.\n\n### Step 4: Label the media\n\nWhere more than one piece of media goes on the call, name each one in the prompt, in the order it\nwill be passed, and say once what each is. Unlabelled, the model picks its own subject out of the\nset. **The labelling convention comes from the model's guide** — a numbered slot, a tag, or plain\nwords. A clip carrying a single start frame and nothing else has nothing to label — just say what it\nopens on.\n\n- **A storyboard sheet** — labelled once at the top as the storyboard for this clip. Don't contradict\n  it and don't caption it: a sentence that would work as a caption for the panel is a description of\n  a still, so write that panel's beat in motion instead.\n- **A start or end frame** — named as the frame the clip opens or closes on.\n- **The person** — labelled on first mention in a beat, then the plain noun. Repeat the label later\n  in the same beat and the render snaps back to the reference, breaking continuity.\n- **The product** — label it where it is shown intact: first introduction, a hero close-up, a label\n  point-out. Drop the label during any verb that changes it — opening, tearing, pouring, scooping,\n  applying, drinking, mixing, putting on — because carrying it through forces the product back to its\n  reference state mid-beat, so an opened jar snaps shut. Pick the label up again once it is shown\n  intact.\n- **Several product images** — point at whichever shows the angle that beat needs.\n\n**Never put a label inside quoted speech.** Quoted text is read aloud as written, so the label gets\nspoken.\n\n### Step 5: Read the storyboard\n\nSkip this where the clip has no sheet.\n\nRun `image_analysis` on the sheet before writing anything, and take these from each panel as it\nactually rendered:\n\n- the camera position — held at arm's length, locked off, over the shoulder\n- the framing distance\n- the action\n- which hand is doing what\n- where the product is — held, set down, worn, hidden, absent\n- the expression\n\n**Where the sheet and the plan disagree, the sheet is what the model will see, so the sheet wins.**\nWrite the cut to the panel, and say which panel disagreed and how.\n\n### Step 6: Write the clip\n\nWrite 6a to 6d, laid out in the order the model's guide sets out.\n\n**6a — The look.** One line: the light, and how the camera behaves across the clip. Add the delivery\nregister where anyone speaks.\n\n**6b — The beats**, in order.\n\n- **Each cut is one or two sentences, carrying one verb that moves something.** A second moving\n  verb is a second cut.\n- **With cuts**, head each with its number and its span — `Cut 2 (3.5-7s):` — and end every cut but\n  the last with a hard-cut instruction on its own line. Both go in whatever the guide says about\n  timings: the spans are what make the cuts add up to the clip's length, and without the hard-cut\n  line the model runs the beats together as one continuous move.\n- **Without cuts, time the beats only where the model's guide uses timings.** Some models read a\n  written span; others are made unstable by one and want the beats in order with the pacing following\n  the action.\n- **How long each beat is.** Each one runs two to four seconds of the clip, and they add up to its\n  length exactly.\n- **The first beat opens already underway** rather than on a held pose. Where a frame is fixed, it\n  opens exactly on what that shows — nothing enters and nothing changes state before the motion\n  starts. Where a person is on camera, that means hands already moving or a reaction already on the\n  face, and the first words landing at the top of the beat rather than after a pause.\n\n**6c — The camera, in every beat.** Where a sheet or a start frame already fixes what is in frame and\nwhere the camera sits, only how it moves is yours — a push in, a pull back, a tilt, a pan, a rack\nfocus, a handheld sway. Without one, write the framing and the position too. Locked off means the\nframe doesn't move and everything moves inside it, so keep any word suggesting an unsteady grip or a\ntravelling lens out of those beats. At arm's length the camera is the phone and carries a small\nnatural unsteadiness; nothing showing a handset, a screen, or either reflected belongs in shot.\n\n**6d — The person, where one is on camera.**\n\n- **Hands, counted at every moment.** A phone held at arm's length takes up one hand, so a single hand\n  is free to act; a locked-off shot leaves both. Describe a third job — hold one thing, work a second,\n  point at a third — and a third hand appears. Where hands don't matter to the beat, leave them out.\n  When an arm's-length beat calls for two working hands, it can't be shot as framed — send it back to\n  the storyboard to re-frame that beat rather than working around it in the prompt.\n- **Performance.** Several small movements within each beat, and at least one thing that changes —\n  weight moving forward, a breath landing, a grin breaking after a pause. A beat with no change inside\n  it reads as a frozen frame. Pitch each one higher than the level you want on screen, since the model\n  plays performance back flatter than it is written. Vary the expression from beat to beat. Where they\n  speak, let the delivery slip and recover once somewhere in the clip.\n\n**Every beat is bounded by these:**\n\n- **One action per beat** — one continuous motion at one object. Chained verbs get dropped or mangled.\n- **Nothing below limb scale.** Moving, holding, tilting, lifting, gesturing and walking render.\n  Unscrewing a cap, pulling a drawstring, pumping a pump head, working a zip or a clasp do not. Where\n  a beat needs the thing in a new state, open it with the thing already in that state.\n- **Interactions respect the physical chain.** Nobody drinks from a sealed bottle. Where the whole\n  chain won't fit, open the clip further along it.\n- **Positions hold through a beat.** Where something is, how it's held, how it's worn stays put unless\n  a visible movement changes it. An unstated change renders as a teleport, and once changed it stays\n  changed.\n- **Describe only what the media show.** Texture, back panels, interiors and hidden mechanisms\n  outside the frame come back invented.\n- **Nothing can be written as absent.** A thing named is a thing rendered, whether object, action or\n  sound.\n\nEvery piece of media also goes to the model on the call, so don't restate what it shows — the same\ninstruction twice renders a slideshow. But write the prompt full: written thin, the image wins and\nthe clip comes back as held poses with pauses between them.\n\n### Step 7: Write the audio\n\nSkip this where the clip carries no sound.\n\nWhere anyone speaks, split the clip's script segment across the beats at natural phrase breaks,\nroughly in proportion to their lengths. The words themselves are exact — no rewording, no insertions,\nno deletions — and the pieces must reassemble into the whole segment with nothing heard twice.\n**Mark the speech the way the model's guide marks it**; that marking is what drives lip movement.\n\nThe audio line says how the speech is produced and what else is heard. On camera: the person speaking\nin their own accent, lip-synced to the segments above. As voiceover: spoken off-screen, no mouth\nmoving. Say what the recording sounds like, following the look the plan named — a phone-shot piece\nmeans phone-microphone audio and room tone.\n\nWhere anyone speaks on camera, give each beat a moment with the mouth closed while the hands work. Lip\nmovement is the weakest thing the model renders, and continuous speech across a clip smears it.\n\nAmbience is room tone plus what the beat's action actually makes. Name the source and the surface it\nhappens on, keep it to two or three sounds at once, and never write a sound as absent — a sound named\nin a beat whose action doesn't produce it makes the model add the action to match. Don't repeat spoken\nwords in the ambience line, and don't name or imply music unless the user asked for it.\n\n### Step 8: Hand the prompts back\n\nWrite every clip's prompt before any of them is generated. They don't depend on each other, and a\nprompt is far cheaper to fix on paper than after it has rendered.\n\nHand back one prompt per clip, in order, each tied to its clip, along with the media that go with it\nand the order they were labelled in.\n\n## Edge cases\n\n- **No beats were supplied, only a brief** → write the clip as one continuous shot and split it into\n  beats yourself, to the seconds in Step 6.\n- **The beats need more time than the clip has** → say so and hand back the beats that fit. Don't\n  compress them into fewer seconds; the extra ones get dropped or mangled in the render.\n- **A start frame arrives alongside reference images** → they cannot be sent on one call. Say which\n  the clip should keep, and write the prompt for that form.\n- **The model is one with no guide here** → call `list_video_models` for its accepted inputs, and\n  write to the guide of the family it most resembles.\n\n## Reference\n\n**Model prompt guides — read the one for the chosen model:**\n\n- `references\u002Fprompt-seedance.md` — native audio markers, one-move-per-shot, storyboards.\n- `references\u002Fprompt-veo.md` — cinematography-first order, `SFX:` \u002F `Ambient noise:` markers, timecodes.\n- `references\u002Fprompt-gemini-omni.md` — conversational editing of an existing clip; edits, not extends.\n- `references\u002Fprompt-kling.md` — director-script style, `@Element` references, per-speaker dialogue labels.\n- `references\u002Fprompt-grok.md` — tight front-loaded briefs, positive-only exclusions, one action per clip.\n- `references\u002Fprompt-wan.md` — front-loaded prose, multi-shot timestamps, driving-audio sync.\n",{"name":5,"license":15,"metadata":24,"description":7},{"summary":25,"version":26,"category":27},"Translates your scenes into the highly specific, detailed instructions that different video models require. It times the camera motion and steers around known AI rendering limits to match your exact visual brief.","0.5.0","creative","master","63f17866001c5ca854bf2fa0b678ce277133de5b","skills\u002Fmarketplace\u002Fwriting-video-prompts__skills-supercmohq-writing-video-prompts",0,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS","https:\u002F\u002Fbg-szy.github.io\u002FTOP-SKILLS\u002F","https:\u002F\u002Fgithub.com\u002Fbg-szy","https:\u002F\u002Fwww.ecust.edu.cn\u002F","a student of East China University of Science and Technology","2026-11-15 03:38:02",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add bg-szy\u002FTOP-SKILLS --skill writing-video-prompts",{"tool":44,"command":45},"bunx","bunx skills add bg-szy\u002FTOP-SKILLS --skill writing-video-prompts",{"tool":47,"command":48},"pnpm","pnpm dlx skills add bg-szy\u002FTOP-SKILLS --skill writing-video-prompts","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fwriting-video-prompts__skills-supercmohq-writing-video-prompts\u002FSKILL.md",true,[52],{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"skill-report.json","REFERENCE","参考文档",28664,"https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fwriting-video-prompts__skills-supercmohq-writing-video-prompts\u002Fskill-report.json",[59,68,75,82,89,96],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":67},"01M7DXW9N826PKXK174PVW3038","serpmaven","Scrape Google search results (SERP) to JSON with SERP Maven (serpmaven) - a real Chrome browser on a residential proxy with sticky sessions. Run it as a CLI, a Python library, a local HTTP search API (`serve`), an MCP server or a Docker image; set the proxy login, workers, pages and cost tracking. Use when the user wants Google results, rank tracking, a self-hosted SERP API, or live Google search for an AI agent.","MIT","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fserpmaven\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":67},"01M7DXVHSTWDV65475QEQ915M7","revnu-landing-page-audit","Find out why a landing page isn't turning visitors into signups, demos or sales. Audits one page against what the ad, post or search that sends people there promised, then checks the first screen, proof, call to action, friction and mobile. Returns at most 7 prioritized fixes, plus what is fine and what could not be checked; rewrites specific lines and fixes the page in the repo when asked. Never promises a conversion lift. Works without a Revnu account. Use when the user says or asks: \"audit my landing page\", \"why isn't this page converting\", \"review my homepage\", \"my ads get clicks but no signups\", \"people bounce from the pricing page\", \"CRO review\", \"improve conversion rate\", \"is my hero section clear\", \"roast my landing page\", \"rewrite my headline\", \"what's wrong with this page\", \"landing page feedback\", \"make the signup page convert better\". Not for SEO or AI-search audits, page speed engineering on its own, or building a new page from scratch.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-landing-page-audit\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":67},"01M7DXV8X2D7B3VTD4HHYYHTGR","revnu-growth-context","Write down once who buys, what they're stuck on, what it costs, what proof you have and what you've already tried, so every growth task starts from facts instead of a questionnaire. Creates or updates .agents\u002Frevnu-growth.md from the repo, the live site and the user, and imports an existing product-marketing.md. Use when the user says \"set up our marketing context\", \"who is our ideal customer\", \"write our ICP\", \"describe the product for marketing\", \"positioning doc\", \"what's our brand voice\", \"remember this about our customers\", \"log what we tried\", \"what have we learned from marketing\", \"we changed pricing, update the context\", \"update our growth notes\", \"summarize the business for the agent\", \"import my product-marketing.md\", \"start a growth doc\", or before a first marketing, SEO, outreach or launch task when no context file exists. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-growth-context\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":67},"01M7DXV0DGJ1G0A60PAHC6J3C5","revnu-get-listed","Get onto the third-party pages AI answers and search cite when buyers ask for the best tool in your category: \"best X\" roundups, directories, review sites, GitHub awesome lists, community threads and comparison pages. Finds which ones are cited, ranks them by fit and effort, and drafts each submission, PR or pitch by the target's own rules. Use when the user says \"get us listed\", \"which directories should we submit to\", \"how do we get into best-of lists\", \"why does ChatGPT recommend competitors and not us\", \"get mentioned in AI answers\", \"submit to awesome lists\", \"get on G2\", \"where are our competitors listed\", \"pitch listicle authors\", \"find backlink opportunities\", \"launch directories\", \"get into roundups\", \"Product Hunt alternatives\", \"where should we be listed\", \"get cited by Perplexity\". Never submits, pays or posts without a yes. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-get-listed\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":67},"01M7DXTSPMGT9D6DSS0HP9BSTA","revnu-ai-readiness-audit","Audit whether your website gives ChatGPT, Perplexity, Claude, Gemini and Google's AI Overviews something worth quoting, and fix the pages that don't. Use when the user says or asks \"is my site ready for AI search\", \"audit my site for AI answers\", \"AEO audit\", \"GEO audit\", \"why doesn't ChatGPT quote our pricing\", \"make our pages citable by AI\", \"do I need llms.txt\", \"does schema help with ChatGPT\", \"check our structured data\", \"is our content visible without JavaScript\", \"AI SEO audit\", \"optimize our homepage for AI answers\", \"LLM SEO check\", \"why does AI get our product wrong\", \"fix our site for AI Overviews\". Checks fetchability and indexability, answers in the raw HTML, entity facts, schema (including JS-injected markup), headings, dates and llms.txt, then gives at most 10 prioritised fixes and edits the repo when asked. Works without a Revnu account.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Frevnu-ai-readiness-audit\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":63,"licenseFileUrl":6,"sourceHtmlUrl":100,"categories":101,"keywords":102,"publishedAt":67},"01M7DXSZHCVTRS620GC3S97JZV","proxy-troubleshooting","Diagnose residential proxy failures, with NodeMaven's gateway as the reference - 407 with correct credentials, 406, 410, 500, a connection that hangs, ERR_TUNNEL_CONNECTION_FAILED in a browser, Playwright or Puppeteer ignoring proxy credentials, a sticky session that changes IP, geo-targeting that is silently ignored, passwords with special characters, and unexpected traffic from headless Chrome. Use when a proxy connection fails, returns the wrong location, or costs more traffic than expected.","https:\u002F\u002Fgithub.com\u002Fbg-szy\u002FTOP-SKILLS\u002Fblob\u002Fmaster\u002Fskills\u002Fmarketplace\u002Fproxy-troubleshooting\u002FSKILL.md",[],[]]