[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6BXZJC7DDEGJRAZ74GT4NWQ":3,"skill-same:01M6BXZJC7DDEGJRAZ74GT4NWQ":82},{"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":48,"linkOnly":49,"markdownContent":50,"markdownBody":51,"frontMatter":52,"refName":53,"commitSha":54,"skillDirectory":55,"forksCount":56,"repositoryHtmlUrl":57,"repositoryHomepageUrl":58,"ownerHtmlUrl":59,"ownerBlogUrl":12,"ownerDescription":60,"updatedAt":61,"install":62,"resources":75},"01M6BXZJC7DDEGJRAZ74GT4NWQ","flux-2-klein-prompting",null,"Prompt Black Forest Labs' FLUX.2 [klein] — the subject\u002Fenvironment\u002Fstyle\u002Ftechnical hierarchy it processes in order, guidance-scale and step choices, negative prompts aimed at real failure modes, and the seed-locked one-variable-at-a-time loop its distilled checkpoints make fast enough to run. Which dials exist depends on base versus distilled. Use whenever the model id contains flux-2\u002Fklein or flux-2-klein (fal-ai\u002Fflux-2\u002Fklein\u002F4b, fal-ai\u002Fflux-2\u002Fklein\u002F9b, their \u002Fbase, \u002Fedit and \u002Flora variants, fal-ai\u002Fflux-2\u002Fklein\u002Frealtime, black-forest-labs\u002Fflux-2-klein-4b, black-forest-labs\u002Fflux-2-klein-9b) on generate_image, edit_image or a TextToImage node. Not for FLUX.2 pro, flex, max or dev.","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\u002Fflux-2-klein-prompting\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47],"prompting","flux","flux 2 klein prompting","edit","distilled","variants","hierarchy","at","which","black","time","seed","max","image","node","fal","model","loop","LoRA","fast","dev","run","generate","realtime","2026-11-10 10:12:30",false,"---\nname: flux-2-klein-prompting\ndescription: Prompt Black Forest Labs' FLUX.2 [klein] — the subject\u002Fenvironment\u002Fstyle\u002Ftechnical hierarchy it processes in order, guidance-scale and step choices, negative prompts aimed at real failure modes, and the seed-locked one-variable-at-a-time loop its distilled checkpoints make fast enough to run. Which dials exist depends on base versus distilled. Use whenever the model id contains flux-2\u002Fklein or flux-2-klein (fal-ai\u002Fflux-2\u002Fklein\u002F4b, fal-ai\u002Fflux-2\u002Fklein\u002F9b, their \u002Fbase, \u002Fedit and \u002Flora variants, fal-ai\u002Fflux-2\u002Fklein\u002Frealtime, black-forest-labs\u002Fflux-2-klein-4b, black-forest-labs\u002Fflux-2-klein-9b) on generate_image, edit_image or a TextToImage node. Not for FLUX.2 pro, flex, max or dev.\n---\n\n# FLUX.2 [klein] → structure first, then iterate fast\n\nklein is the small, fast member of the FLUX.2 family, at 4B or 9B. It ships\ntwo kinds of checkpoint. The distilled ones (every route without `\u002Fbase`) run\nin 4 to 8 steps, which fal's guide puts under a second. The `\u002Fbase` ones run\n28 steps by default and carry the guidance, negative-prompt and acceleration\ndials. The speed is the point of the distilled routes. When a generation takes\n30 seconds, prompt refinement stalls; here you can change one clause and look\nagain, so the working method below is worth more than any single prompt.\n\nReach it with `find_model` for `text_to_image` or `image_to_image`, then\n`generate_image` \u002F `edit_image`. `generate_image` forwards only the prompt, a\nnegative prompt, the size and a quality string, and `edit_image` only the\nimages, the prompt, a negative prompt, a target size and a strength. Every other\nfield below exists only on the route's own node: find it with `search_nodes`,\nread its fields with `get_node_info`, and run it with `run_node`. Each run is a\npaid generation, so confirm the scope with the user before a long loop.\n\n## Prompt hierarchy\n\nklein reads a prompt in this order, and content words (nouns) move the output\nmore than modifiers do. Write it in the same order:\n\n1. **Subject** — concrete beats abstract. \"A woman in her mid-30s with\n   shoulder-length auburn hair\" over \"a person\".\n2. **Environment** — lighting, atmosphere, spatial relationships. \"First light\n   filtering through morning mist\" gives rendering cues that \"morning scene\"\n   does not.\n3. **Style and mood** — direction that guides without overwhelming:\n   \"documentary photography\", \"studio lighting\", \"cinematic\".\n4. **Technical** — composition (\"rule of thirds\"), depth of field (\"shallow\n   focus on the foreground\"), palette (\"warm earth tones\").\n\n> Professional headshot of a male architect in his 40s, salt-and-pepper beard,\n> black-rimmed glasses, charcoal blazer. Modern office background with\n> architectural models visible but softly blurred. Natural window light from\n> the left creating gentle shadows. Corporate photography style, sharp focus on\n> the eyes, neutral grey backdrop.\n\n## Parameters\n\n| Parameter | Where it exists | What it trades | Where to sit |\n| :--- | :--- | :--- | :--- |\n| `guidance_scale` | fal `\u002Fbase` routes (default 5); Replicate `-base` calls it `guidance` (default 4) | Prompt adherence against creative freedom | 2–4 for artistic interpretation, 5–8 for product photography and technical illustration |\n| `num_inference_steps` | fal only: 4–50 on `\u002Fbase` (default 28), 4–8 on distilled (default 4), 1–8 on `realtime` | Quality against time | Low while developing a prompt, high for print-ready output |\n| `acceleration` | fal `\u002Fbase` routes | Detail against throughput | `regular` for production, `high` for maximum speed |\n| `negative_prompt` | fal `\u002Fbase` routes | See below | Failure modes only |\n| `seed` | every route | Nothing — it removes noise from the comparison | Pin it while testing prompt variations |\n\nOn a distilled route the prompt and the seed are the whole instrument, and a\nnegative prompt sent through `generate_image` is dropped.\n\nAspect ratio follows the deployment: landscape for presentations and web,\nportrait for social and mobile, square for profiles and balanced compositions.\n\n## Technique\n\n- **Emphasis is natural language, not weights.** fal's guide says there is no\n  weight syntax. \"Prominently featuring\", \"with particular attention to\" and\n  \"especially detailed\" are how you raise an element's priority.\n- **Negative prompts target failure modes, not quality** (`\u002Fbase` only). For\n  portraits: \"distorted features, unnatural proportions, extra limbs\". For\n  landscapes:\n  \"oversaturated colours, artificial lighting, lens distortion\". A generic\n  quality list buys nothing.\n- **Text renders well when specified.** Give the copy, the font class, the\n  colour, the placement and the capitalisation: \"a white coffee mug with the\n  text 'GOOD MORNING' in bold sans-serif black letters, centred on the mug\".\n  Copy that belongs to the design, in other words — a title or a super on a\n  still headed for a timeline goes on as a text clip afterwards\n  (`caption-titles`), not into the render.\n- **Multi-reference conditioning is the family's edge.** \"The subject from the\n  first image wearing the jacket from the second image, photographed in the\n  environment from the third image.\"\n\n## The five mistakes\n\n- **Overloaded prompts.** fal's guide says that past roughly 100 words the\n  model starts dropping clauses. Every word should be doing something.\n- **Vague style references.** \"Make it look good\" is not a direction. \"Shot on\n  Hasselblad medium format\" or \"impressionist painting technique\" is.\n- **Missing composition.** Prompts describe the subject thoroughly and never\n  say where it sits. Add \"centred composition\", \"rule of thirds with the\n  subject on the left vertical\", or \"bird's eye view\".\n- **Conflicting instructions.** \"Photorealistic portrait\" and \"watercolour\n  painting style\" in one prompt commits to neither.\n- **Neglected lighting.** Lighting is what makes an image dimensional. Always\n  name it: golden hour, studio softbox, rim lighting creating a silhouette.\n\n## The loop klein is for\n\nStart with subject, environment and style. Generate. Pin the seed. Change one\nelement — add lighting detail, refine the subject, move the composition — and\ngenerate again. One variable at a time is what tells you which clause drove\nwhich change; a distilled route's turnaround is what makes that affordable.\n\nKeep the prompts that worked. A library of formulas per use case, with the\nguidance and step values that suited each, is the output of this loop as much\nas the images are.\n\nAdapted from fal's FLUX.2 [klein] prompt guide:\nhttps:\u002F\u002Ffal.ai\u002Flearn\u002Fdevs\u002Fflux-2-klein-prompt-guide\n","# FLUX.2 [klein] → structure first, then iterate fast\n\nklein is the small, fast member of the FLUX.2 family, at 4B or 9B. It ships\ntwo kinds of checkpoint. The distilled ones (every route without `\u002Fbase`) run\nin 4 to 8 steps, which fal's guide puts under a second. The `\u002Fbase` ones run\n28 steps by default and carry the guidance, negative-prompt and acceleration\ndials. The speed is the point of the distilled routes. When a generation takes\n30 seconds, prompt refinement stalls; here you can change one clause and look\nagain, so the working method below is worth more than any single prompt.\n\nReach it with `find_model` for `text_to_image` or `image_to_image`, then\n`generate_image` \u002F `edit_image`. `generate_image` forwards only the prompt, a\nnegative prompt, the size and a quality string, and `edit_image` only the\nimages, the prompt, a negative prompt, a target size and a strength. Every other\nfield below exists only on the route's own node: find it with `search_nodes`,\nread its fields with `get_node_info`, and run it with `run_node`. Each run is a\npaid generation, so confirm the scope with the user before a long loop.\n\n## Prompt hierarchy\n\nklein reads a prompt in this order, and content words (nouns) move the output\nmore than modifiers do. Write it in the same order:\n\n1. **Subject** — concrete beats abstract. \"A woman in her mid-30s with\n   shoulder-length auburn hair\" over \"a person\".\n2. **Environment** — lighting, atmosphere, spatial relationships. \"First light\n   filtering through morning mist\" gives rendering cues that \"morning scene\"\n   does not.\n3. **Style and mood** — direction that guides without overwhelming:\n   \"documentary photography\", \"studio lighting\", \"cinematic\".\n4. **Technical** — composition (\"rule of thirds\"), depth of field (\"shallow\n   focus on the foreground\"), palette (\"warm earth tones\").\n\n> Professional headshot of a male architect in his 40s, salt-and-pepper beard,\n> black-rimmed glasses, charcoal blazer. Modern office background with\n> architectural models visible but softly blurred. Natural window light from\n> the left creating gentle shadows. Corporate photography style, sharp focus on\n> the eyes, neutral grey backdrop.\n\n## Parameters\n\n| Parameter | Where it exists | What it trades | Where to sit |\n| :--- | :--- | :--- | :--- |\n| `guidance_scale` | fal `\u002Fbase` routes (default 5); Replicate `-base` calls it `guidance` (default 4) | Prompt adherence against creative freedom | 2–4 for artistic interpretation, 5–8 for product photography and technical illustration |\n| `num_inference_steps` | fal only: 4–50 on `\u002Fbase` (default 28), 4–8 on distilled (default 4), 1–8 on `realtime` | Quality against time | Low while developing a prompt, high for print-ready output |\n| `acceleration` | fal `\u002Fbase` routes | Detail against throughput | `regular` for production, `high` for maximum speed |\n| `negative_prompt` | fal `\u002Fbase` routes | See below | Failure modes only |\n| `seed` | every route | Nothing — it removes noise from the comparison | Pin it while testing prompt variations |\n\nOn a distilled route the prompt and the seed are the whole instrument, and a\nnegative prompt sent through `generate_image` is dropped.\n\nAspect ratio follows the deployment: landscape for presentations and web,\nportrait for social and mobile, square for profiles and balanced compositions.\n\n## Technique\n\n- **Emphasis is natural language, not weights.** fal's guide says there is no\n  weight syntax. \"Prominently featuring\", \"with particular attention to\" and\n  \"especially detailed\" are how you raise an element's priority.\n- **Negative prompts target failure modes, not quality** (`\u002Fbase` only). For\n  portraits: \"distorted features, unnatural proportions, extra limbs\". For\n  landscapes:\n  \"oversaturated colours, artificial lighting, lens distortion\". A generic\n  quality list buys nothing.\n- **Text renders well when specified.** Give the copy, the font class, the\n  colour, the placement and the capitalisation: \"a white coffee mug with the\n  text 'GOOD MORNING' in bold sans-serif black letters, centred on the mug\".\n  Copy that belongs to the design, in other words — a title or a super on a\n  still headed for a timeline goes on as a text clip afterwards\n  (`caption-titles`), not into the render.\n- **Multi-reference conditioning is the family's edge.** \"The subject from the\n  first image wearing the jacket from the second image, photographed in the\n  environment from the third image.\"\n\n## The five mistakes\n\n- **Overloaded prompts.** fal's guide says that past roughly 100 words the\n  model starts dropping clauses. Every word should be doing something.\n- **Vague style references.** \"Make it look good\" is not a direction. \"Shot on\n  Hasselblad medium format\" or \"impressionist painting technique\" is.\n- **Missing composition.** Prompts describe the subject thoroughly and never\n  say where it sits. Add \"centred composition\", \"rule of thirds with the\n  subject on the left vertical\", or \"bird's eye view\".\n- **Conflicting instructions.** \"Photorealistic portrait\" and \"watercolour\n  painting style\" in one prompt commits to neither.\n- **Neglected lighting.** Lighting is what makes an image dimensional. Always\n  name it: golden hour, studio softbox, rim lighting creating a silhouette.\n\n## The loop klein is for\n\nStart with subject, environment and style. Generate. Pin the seed. Change one\nelement — add lighting detail, refine the subject, move the composition — and\ngenerate again. One variable at a time is what tells you which clause drove\nwhich change; a distilled route's turnaround is what makes that affordable.\n\nKeep the prompts that worked. A library of formulas per use case, with the\nguidance and step values that suited each, is the output of this loop as much\nas the images are.\n\nAdapted from fal's FLUX.2 [klein] prompt guide:\nhttps:\u002F\u002Ffal.ai\u002Flearn\u002Fdevs\u002Fflux-2-klein-prompt-guide\n",{"name":5,"description":7},"main","c7226baa82daca42a19a5855c116544f97aa8bac","packages\u002Fsystem-skills\u002Fflux-2-klein-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":63,"prompt":73,"zipAvailable":74},[64,67,70],{"tool":65,"command":66},"npx","npx skills add nodetool-ai\u002Fnodetool --skill flux-2-klein-prompting",{"tool":68,"command":69},"bunx","bunx skills add nodetool-ai\u002Fnodetool --skill flux-2-klein-prompting",{"tool":71,"command":72},"pnpm","pnpm dlx skills add nodetool-ai\u002Fnodetool --skill flux-2-klein-prompting","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002Fpackages\u002Fsystem-skills\u002Fflux-2-klein-prompting\u002FSKILL.md",true,[76],{"path":77,"resourceType":78,"resourceTypeDesc":79,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fsubmission-contract.md","REFERENCE","参考文档",3505,"https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002Fportable\u002Fskills\u002Fcmux-steer\u002Freferences\u002Fsubmission-contract.md",[83,92,126,149,164,180],{"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":87,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":91},"01M742MX3Y2ZYK884ZXTRFKB9P","humanizer","Rewrite AI-sounding text so it reads like the writer without changing what it says.\nUse when editing or reviewing prose for AI tells: not-X-but-Y contrasts, one-line\nclosers, staged openers, forced triads, dashes everywhere, inflated claims, sales\nlanguage, stock AI words, bold labels, or filler.","MIT","https:\u002F\u002Fgithub.com\u002Fnodetool-ai\u002Fnodetool\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fhumanizer\u002FSKILL.md",[],[],"2026-11-11 15:11:14",{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":106,"publishedAt":48},"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",[98,101],{"categoryNo":99,"categoryName":100,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",{"categoryNo":102,"categoryName":103,"taxonomyType":104,"taxonomyTypeDesc":105},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[94,107,108,109,110,111,112,113,114,115,116,117,38,118,119,120,121,122,45,123,124,125],"MCP","Media","workflows","when","generation","media generation","with","visual","use","collections","for","work","debug","graph","validate","assets","server","mcp server","workflow",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":133,"publishedAt":48},"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",[132],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[24,134,135,136,137,138,139,38,140,141,142,34,143,144,46,145,37,146,40,147,148],"wan 2 6 prompting","line","expansion","input","image-to-video","text-to-video","route","only","mode","flash","Budget","text","animate","Audio","Video",{"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":157,"publishedAt":48},"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",[155,156],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":99,"categoryName":100,"taxonomyType":21,"taxonomyTypeDesc":22},[125,158,148,159,160,161,162,115,163,94,117],"video workflow","reusable","storyboard","build","batches","single",{"skillNo":165,"name":166,"nameZh":6,"description":167,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":168,"categories":169,"keywords":171,"publishedAt":48},"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",[170],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[172,173,148,113,174,175,94,176,177,178,179],"clone","video clone","cast","new","analyze","product","structure","reference",{"skillNo":181,"name":182,"nameZh":6,"description":183,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":184,"categories":185,"keywords":187,"publishedAt":48},"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",[186],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[188,148,147,189,190,191,192,193,31,194,195,196,197,110,198,199,40,200,201,202,203],"video audio continuity","continuity","write","why","back","veo 3","cut","later","across","filter","generated","how to","mix","design","boundary","how"]