[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6BXZYF73D4K383YAFPH4K76":3,"skill-same:01M6BXZYF73D4K383YAFPH4K76":81},{"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":47,"linkOnly":48,"markdownContent":49,"markdownBody":50,"frontMatter":51,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":57,"ownerHtmlUrl":58,"ownerBlogUrl":12,"ownerDescription":59,"updatedAt":60,"install":61,"resources":74},"01M6BXZYF73D4K383YAFPH4K76","hailuo-prompting",null,"Prompt MiniMax's Hailuo video line — the chronological beat order it animates from, the fifteen bracketed camera commands the Director endpoints accept and the three-per-bracket rule, when to switch prompt_optimizer off, why there is no negative prompt to fall back on, and prompting the delta on image-to-video. Use whenever the model id names Hailuo or MiniMax video-01 (fal-ai\u002Fminimax\u002Fhailuo-2.3\u002Fpro\u002Ftext-to-video, fal-ai\u002Fminimax\u002Fhailuo-02\u002Fpro\u002Fimage-to-video, fal-ai\u002Fminimax\u002Fhailuo-02-fast\u002Fimage-to-video, fal-ai\u002Fminimax\u002Fvideo-01-director, fal-ai\u002Fminimax\u002Fvideo-01-subject-reference, hailuo\u002F2-3-image-to-video-pro, minimax\u002Fhailuo-2.3, minimax\u002Fhailuo-02-fast, and the direct MiniMax provider's MiniMax-Hailuo-2.3, MiniMax-Hailuo-02, T2V-01-Director, I2V-01-Director and S2V-01) on generate_video, animate_image, generate_video_from_references or a TextToVideo node. Not for MiniMax H3, which has its own guide.","nodetool-ai\u002Fnodetool","nodetool-ai","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F164873569?v=4","01M6BXPFDW36K9K1MR8ZH63HG2","nodetool.ai","01M6BXPFJ893XM219GQN60B0WK",635,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fhailuo-prompting\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46],"prompting","hailuo prompting","node","when","which","image-to-video","text-to-video","line","commands","director","optimizer","back","references","image","fal","direct","model","animate","text","why","Video","fast","generate","2026-11-10 10:12:30",false,"---\nname: hailuo-prompting\ndescription: Prompt MiniMax's Hailuo video line — the chronological beat order it animates from, the fifteen bracketed camera commands the Director endpoints accept and the three-per-bracket rule, when to switch prompt_optimizer off, why there is no negative prompt to fall back on, and prompting the delta on image-to-video. Use whenever the model id names Hailuo or MiniMax video-01 (fal-ai\u002Fminimax\u002Fhailuo-2.3\u002Fpro\u002Ftext-to-video, fal-ai\u002Fminimax\u002Fhailuo-02\u002Fpro\u002Fimage-to-video, fal-ai\u002Fminimax\u002Fhailuo-02-fast\u002Fimage-to-video, fal-ai\u002Fminimax\u002Fvideo-01-director, fal-ai\u002Fminimax\u002Fvideo-01-subject-reference, hailuo\u002F2-3-image-to-video-pro, minimax\u002Fhailuo-2.3, minimax\u002Fhailuo-02-fast, and the direct MiniMax provider's MiniMax-Hailuo-2.3, MiniMax-Hailuo-02, T2V-01-Director, I2V-01-Director and S2V-01) on generate_video, animate_image, generate_video_from_references or a TextToVideo node. Not for MiniMax H3, which has its own guide.\n---\n\n# Hailuo → a sequence of beats, in order\n\nHailuo animates the order it reads. It has no negative prompt, no shot list and\nno CFG dial, so every control you get is in the wording: what happens, in what\norder, and how the camera behaves while it happens. Clips run 6 or 10 seconds\n(10 s is unavailable at 1080p on Hailuo 02).\n\nReach it with `find_model` for `text_to_video` or `image_to_video`, then\n`generate_video` \u002F `animate_image`. Those forward only the prompt (and the\nimage), the duration, the aspect ratio and the resolution. `prompt_optimizer`\nexists only on the route's own node: find it with `search_nodes`, read its\nfields with `get_node_info`, and run it with `run_node`.\n\n## Write the action as beats\n\nBreak one motion into the beats it is made of and connect them with sequencing\nwords — first, then, as, finally. The model animates that order.\n\n> A skateboarder rolls toward a handrail in an empty underground car park.\n> First he crouches low over the deck, then he pops the tail and the board\n> snaps up under him, and as he lands on the rail the trucks grind along the\n> steel, sparks trailing behind. Finally he drops off the end and rolls out of\n> frame past a pillar.\n\nCompare \"a skateboarder does a trick on a rail\", which gives the model the\nwhole choreography to invent and no reason to resolve it the way you meant.\n\nEmotional beats work the same way. \"She reads the letter with growing concern,\ntears well up, then she crumples the paper and throws it\" is three animatable\nstates; \"she is sad about the letter\" is one adjective.\n\n## Camera\n\nOn the Director endpoints (`video-01-director` and its image-to-video form,\nand MiniMax's own `T2V-01-Director` and `I2V-01-Director`), camera moves are\ncommands in square brackets. Fifteen are supported:\n\n`[Truck left]` `[Truck right]` `[Pan left]` `[Pan right]` `[Push in]`\n`[Pull out]` `[Pedestal up]` `[Pedestal down]` `[Tilt up]` `[Tilt down]`\n`[Zoom in]` `[Zoom out]` `[Shake]` `[Tracking shot]` `[Static shot]`\n\nTwo placement rules carry all of it. Commands inside one bracket happen at the\nsame time, capped at three: `[Truck left, Pan right, Zoom in]`. Commands at\ndifferent points in the prompt happen in that order:\n\n```\n[Static shot] The chef sets the plate down under the pass light. [Push in]\nHe wipes the rim with a cloth, then looks up. [Tilt up, Pan right] The dining\nroom comes into view behind him.\n```\n\nThe bracket contract is documented for the Director endpoints only, so on\nevery other Hailuo endpoint write camera direction as prose — \"a slow tracking\nshot follows her\", \"the camera racks focus from the leaves to her face\".\n\n## prompt_optimizer\n\nIt rewrites your prompt with an LLM before generation. It defaults to true on\nevery fal and Replicate Hailuo and video-01 route and to false on kie's Hailuo\n02 routes. kie's Hailuo 2.3, the AtlasCloud routes and the direct MiniMax video\nmodels have no such field. Leave it on for a short prompt: it fills in the\ndetail you did not write. Switch it off once your prompt is a deliberate brief\n— it will otherwise rewrite away the specific wardrobe, the specific light, and\nthe beat order you just spent effort on.\n\n## Image to video\n\nThe frame is already decided. Prompt only what changes: motion, camera, light\nover time, environment animation. Naming the subject briefly to anchor the\nreference is fine — \"the red-haired woman turns slowly to face camera\" — but a\nparagraph re-describing the still is a paragraph not spent on the delta.\n\nKeep the framing consistent with the source. Asking a close-up input for a wide\nestablishing move is how you get a hallucinated body.\n\n`video-01-subject-reference` (and MiniMax's own `S2V-01`, reached through\n`generate_video_from_references`) holds a face across the clip from one\nreference image; describe the person's action, not their appearance.\n\n## No negative prompt\n\nThere is no field, and \"no text, no watermark, not blurry\" in the prompt is\nread as content. Say what should be there instead. Empty sky, not \"no birds\".\nA bare concrete wall, not \"no graffiti\".\n\n## Symptoms\n\n| What went wrong | What to change |\n| :--- | :--- |\n| The action resolves in the wrong order | Number the beats with first \u002F then \u002F as \u002F finally |\n| The camera drifts | Use a bracketed command on a Director endpoint, or name the move in prose |\n| Your specifics disappeared | Set `prompt_optimizer` false on the route's node |\n| A body appears where the input image was cropped | Match the prompt's framing to the input |\n| The thing you excluded shows up anyway | Rewrite the exclusion as a positive description |\n\nGrade the render with `analyze_video` and `detect_video_scenes`, and\n`understand_video` when you need a written read of what landed.\n\nAdapted from MiniMax's camera-movement contract as documented on the fal\nDirector endpoints, and Akool's Hailuo prompt guide:\nhttps:\u002F\u002Ffal.ai\u002Fmodels\u002Ffal-ai\u002Fminimax\u002Fvideo-01-director\u002Fapi\nhttps:\u002F\u002Fakool.com\u002Fblog-posts\u002Fminimax-hailuo-video-prompt-guide\n","# Hailuo → a sequence of beats, in order\n\nHailuo animates the order it reads. It has no negative prompt, no shot list and\nno CFG dial, so every control you get is in the wording: what happens, in what\norder, and how the camera behaves while it happens. Clips run 6 or 10 seconds\n(10 s is unavailable at 1080p on Hailuo 02).\n\nReach it with `find_model` for `text_to_video` or `image_to_video`, then\n`generate_video` \u002F `animate_image`. Those forward only the prompt (and the\nimage), the duration, the aspect ratio and the resolution. `prompt_optimizer`\nexists only on the route's own node: find it with `search_nodes`, read its\nfields with `get_node_info`, and run it with `run_node`.\n\n## Write the action as beats\n\nBreak one motion into the beats it is made of and connect them with sequencing\nwords — first, then, as, finally. The model animates that order.\n\n> A skateboarder rolls toward a handrail in an empty underground car park.\n> First he crouches low over the deck, then he pops the tail and the board\n> snaps up under him, and as he lands on the rail the trucks grind along the\n> steel, sparks trailing behind. Finally he drops off the end and rolls out of\n> frame past a pillar.\n\nCompare \"a skateboarder does a trick on a rail\", which gives the model the\nwhole choreography to invent and no reason to resolve it the way you meant.\n\nEmotional beats work the same way. \"She reads the letter with growing concern,\ntears well up, then she crumples the paper and throws it\" is three animatable\nstates; \"she is sad about the letter\" is one adjective.\n\n## Camera\n\nOn the Director endpoints (`video-01-director` and its image-to-video form,\nand MiniMax's own `T2V-01-Director` and `I2V-01-Director`), camera moves are\ncommands in square brackets. Fifteen are supported:\n\n`[Truck left]` `[Truck right]` `[Pan left]` `[Pan right]` `[Push in]`\n`[Pull out]` `[Pedestal up]` `[Pedestal down]` `[Tilt up]` `[Tilt down]`\n`[Zoom in]` `[Zoom out]` `[Shake]` `[Tracking shot]` `[Static shot]`\n\nTwo placement rules carry all of it. Commands inside one bracket happen at the\nsame time, capped at three: `[Truck left, Pan right, Zoom in]`. Commands at\ndifferent points in the prompt happen in that order:\n\n```\n[Static shot] The chef sets the plate down under the pass light. [Push in]\nHe wipes the rim with a cloth, then looks up. [Tilt up, Pan right] The dining\nroom comes into view behind him.\n```\n\nThe bracket contract is documented for the Director endpoints only, so on\nevery other Hailuo endpoint write camera direction as prose — \"a slow tracking\nshot follows her\", \"the camera racks focus from the leaves to her face\".\n\n## prompt_optimizer\n\nIt rewrites your prompt with an LLM before generation. It defaults to true on\nevery fal and Replicate Hailuo and video-01 route and to false on kie's Hailuo\n02 routes. kie's Hailuo 2.3, the AtlasCloud routes and the direct MiniMax video\nmodels have no such field. Leave it on for a short prompt: it fills in the\ndetail you did not write. Switch it off once your prompt is a deliberate brief\n— it will otherwise rewrite away the specific wardrobe, the specific light, and\nthe beat order you just spent effort on.\n\n## Image to video\n\nThe frame is already decided. Prompt only what changes: motion, camera, light\nover time, environment animation. Naming the subject briefly to anchor the\nreference is fine — \"the red-haired woman turns slowly to face camera\" — but a\nparagraph re-describing the still is a paragraph not spent on the delta.\n\nKeep the framing consistent with the source. Asking a close-up input for a wide\nestablishing move is how you get a hallucinated body.\n\n`video-01-subject-reference` (and MiniMax's own `S2V-01`, reached through\n`generate_video_from_references`) holds a face across the clip from one\nreference image; describe the person's action, not their appearance.\n\n## No negative prompt\n\nThere is no field, and \"no text, no watermark, not blurry\" in the prompt is\nread as content. Say what should be there instead. Empty sky, not \"no birds\".\nA bare concrete wall, not \"no graffiti\".\n\n## Symptoms\n\n| What went wrong | What to change |\n| :--- | :--- |\n| The action resolves in the wrong order | Number the beats with first \u002F then \u002F as \u002F finally |\n| The camera drifts | Use a bracketed command on a Director endpoint, or name the move in prose |\n| Your specifics disappeared | Set `prompt_optimizer` false on the route's node |\n| A body appears where the input image was cropped | Match the prompt's framing to the input |\n| The thing you excluded shows up anyway | Rewrite the exclusion as a positive description |\n\nGrade the render with `analyze_video` and `detect_video_scenes`, and\n`understand_video` when you need a written read of what landed.\n\nAdapted from MiniMax's camera-movement contract as documented on the fal\nDirector endpoints, and Akool's Hailuo prompt guide:\nhttps:\u002F\u002Ffal.ai\u002Fmodels\u002Ffal-ai\u002Fminimax\u002Fvideo-01-director\u002Fapi\nhttps:\u002F\u002Fakool.com\u002Fblog-posts\u002Fminimax-hailuo-video-prompt-guide\n",{"name":5,"description":7},"main","c7226baa82daca42a19a5855c116544f97aa8bac","packages\u002Fsystem-skills\u002Fhailuo-prompting",91,"https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool","https:\u002F\u002Fnodetool.ai\u002F","https:\u002F\u002Fgithub.com\u002Fnodetool-ai","No-code AI development","2026-11-15 03:27:49",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":73},[63,66,69],{"tool":64,"command":65},"npx","npx skills add nodetool-ai\u002Fnodetool --skill hailuo-prompting",{"tool":67,"command":68},"bunx","bunx skills add nodetool-ai\u002Fnodetool --skill hailuo-prompting",{"tool":70,"command":71},"pnpm","pnpm dlx skills add nodetool-ai\u002Fnodetool --skill hailuo-prompting","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fhailuo-prompting\u002FSKILL.md",true,[75],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"README.md","REFERENCE","参考文档",2271,"https:\u002F\u002Fgithub.com\u002FTrollz1004\u002FANTIGRAVITY\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fdevrel-content\u002FREADME.md",[82,116,134,149,165,188],{"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":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":96,"publishedAt":47},"01M6BY4D93R0Q9400ZHJ04DXP1","nodetool","Use the NodeTool MCP server to build, validate, run, and debug visual AI workflows, and to work with assets, collections, and jobs. Use when the user mentions NodeTool or asks for a workflow graph, a node, or a media generation run on their NodeTool server.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fplugins\u002Fchatgpt\u002Fnodetool\u002Fskills\u002Fnodetool\u002FSKILL.md",[88,91],{"categoryNo":89,"categoryName":90,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":92,"categoryName":93,"taxonomyType":94,"taxonomyTypeDesc":95},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[84,97,98,99,100,101,102,103,104,105,106,107,108,26,109,110,111,112,113,114,115,27],"MCP","mcp server","workflow","workflows","validate","debug","run","graph","server","Media","assets","work","use","visual","generation","collections","media generation","with","for",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":123,"publishedAt":47},"01M6BY4AEB05ZPBRVD078ATR2H","wan-2-6-prompting","Prompt Alibaba's Wan 2.6 video — text-to-video with a global style line plus timing-bracketed shots, and image-to-video that describes only what changes over time. Covers the shot-mode switch per provider, the prompt budget, negative prompts, motion intensity wording, audio input and prompt expansion. Use whenever the model id names a Wan 2.6 video route (alibaba\u002Fwan-2.6\u002Ftext-to-video, alibaba\u002Fwan-2.6\u002Fimage-to-video, wan-video\u002Fwan-2.6-t2v, wan-video\u002Fwan-2.6-i2v, wan-video\u002Fwan2.6-i2v-flash) on generate_video, animate_image or a TextToVideo node. Not for Wan 2.5 or 2.7, or the Wan 2.6 image routes.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fwan-2-6-prompting\u002FSKILL.md",[122],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[124,24,37,46,125,44,126,42,41,40,127,26,128,129,130,31,131,30,29,132,133],"wan 2 6 prompting","Budget","Audio","flash","only","route","time","input","mode","expansion",{"skillNo":135,"name":136,"nameZh":6,"description":137,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":138,"categories":139,"keywords":142,"publishedAt":47},"01M6BY47T4748ZG5ZMTAR97ZWZ","video-workflow","Build a reusable NodeTool video workflow for batches or repeated inputs. For a single finished piece, use a storyboard skill.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-workflow\u002FSKILL.md",[140,141],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":89,"categoryName":90,"taxonomyType":21,"taxonomyTypeDesc":22},[99,44,143,84,109,144,145,146,147,115,148],"video workflow","reusable","storyboard","build","batches","single",{"skillNo":150,"name":151,"nameZh":6,"description":152,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":153,"categories":154,"keywords":156,"publishedAt":47},"01M6BY4587N2SYN315SXGEMZC3","video-clone","Analyze a reference video and recreate its shot structure with a new product or cast in NodeTool.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-clone\u002FSKILL.md",[155],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[44,157,158,159,160,84,161,162,163,114,164],"video clone","clone","product","analyze","reference","cast","structure","new",{"skillNo":166,"name":167,"nameZh":6,"description":168,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":169,"categories":170,"keywords":172,"publishedAt":47},"01M6BY4257C2KQYYMD9EK6DK8E","video-audio-continuity","Keep sound continuous across a multi-scene piece cut from generated video — why one clip per scene hard-cuts the audio at every boundary, when to write all the scenes into a single generation instead, the sound-design vocabulary that survives a provider's copyright filter, and how to check the length, the aspect ratio and the mix of what came back. Use when a piece has more than one scene and the video model writes its own audio (Seedance 2, Veo 3, MiniMax H3, Kling 2.6 and later), or when a cut's sound drops out at the shot changes.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fvideo-audio-continuity\u002FSKILL.md",[171],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[44,126,173,174,175,176,177,43,40,178,179,180,35,181,182,183,184,185,186,187,27],"video audio continuity","continuity","design","boundary","how","write","generated","later","across","cut","how to","filter","mix","at","veo 3",{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":192,"categories":193,"keywords":195,"publishedAt":47},"01M6BY3ZANBSQMZRH4MG55R8Z7","veo-3-prompting","Prompt Google's Veo 3 video line — the five-element shot\u002Fsetting\u002Fsubject\u002Faction\u002Fdialogue structure, the cinematography vocabulary it reads directly, duration and aspect-ratio tradeoffs, the audio and auto-fix switches per provider, and the fixes for visual drift, temporal inconsistency and audio mismatch. Use whenever the model id contains veo3 or veo-3 (fal-ai\u002Fveo3.1 and its fast, lite, image-to-video, first-last-frame and reference-to-video routes, fal-ai\u002Fveo3\u002Fimage-to-video, google\u002Fveo3.1 on atlascloud, google\u002Fveo-3 and google\u002Fveo-3.1 on replicate, kie's veo-3-1, and Gemini's own veo-3.1-generate-preview) on generate_video, animate_image, generate_video_from_references or a TextToVideo node.","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fveo-3-prompting\u002FSKILL.md",[194],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[196,197,24,187,37,198,199,46,45,200,44,126,41,40,38,201,202,203,204,36,205,163,206,207,208],"veo 3 prompting","veo","google","gemini","first","kie","replicate","fix","auto","drift","preview","element","auto-fix"]