使用Gemini AI和Tavily Research生成短视频和长视频YouTube脚本
高级
这是一个Content Creation, Multimodal AI领域的自动化工作流，包含 22 个节点。主要使用 If, Set, GoogleDocs, FormTrigger, HttpRequest 等节点。 使用Gemini AI和Tavily Research生成短视频和长视频YouTube脚本
前置要求
- •可能需要目标 API 的认证凭证
- •Google Gemini API Key
使用的节点 (22)
工作流预览
可视化展示节点连接关系，支持缩放和平移
导出工作流
复制以下 JSON 配置到 n8n 导入，即可使用此工作流
{
"meta": {
"instanceId": "31645fe974a636f0f842088a1a2b47fcb85d774386add7f76f469cfce7c26489"
},
"nodes": [
{
"id": "b57b0c96-d084-4b99-9f2a-df8892654219",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"notes": "You are a senior YouTube script writer for the Website Learners channel.\nWrite a complete script with:\n ① a catchy hook (max 20 words)\n ② a friendly intro\n ③ 3‑5 body sections with smooth transitions\n ④ a final CTA to like & subscribe.\nKeep the tone casual and engaging.",
"position": [
368,
160
],
"parameters": {
"text": "=Here is the video title: {{ $('On form submission').item.json['Provide topic'] }}\n\nHere is the summary: {{ $('Websearch').item.json.message.content }}",
"options": {
"systemMessage": "=You are a senior YouTube short script writer. you have to create a complete youtube short video script with the provided summary which is collected from top google results for the given title.\n---\n\n✦ OUTPUT FORMAT ✦\n\n• Return the entire script as plain text exactly like this example\n\n• Style Reference (tone, pacing, formatting):\n{{ $json.short_style_ref }}\n\nuse it for a reference script\n\n(Keep one blank line between every sentence or logical beat, no markdown, no JSON.)\n",
"returnIntermediateSteps": false
},
"promptType": "define"
},
"typeVersion": 1.9
},
{
"id": "edfa00aa-dc19-4e4e-8e65-e2a6192d5a19",
"name": "On form submission",
"type": "n8n-nodes-base.formTrigger",
"position": [
-880,
480
],
"webhookId": "6ab6f2cf-862e-4ed4-a3f7-6cdee0929c79",
"parameters": {
"options": {},
"formTitle": "Provide topic",
"formFields": {
"values": [
{
"fieldLabel": "Provide topic",
"requiredField": true
},
{
"fieldType": "dropdown",
"fieldLabel": "Choose video type",
"fieldOptions": {
"values": [
{
"option": "Short video"
},
{
"option": "Long video"
}
]
},
"requiredField": true
}
]
}
},
"typeVersion": 2.2
},
{
"id": "554ca2ba-2091-41ce-b656-31fb35df36a6",
"name": "Create Doc",
"type": "n8n-nodes-base.googleDocs",
"position": [
688,
160
],
"parameters": {
"title": "={{ $('On form submission').item.json['Provide topic'] }}_Shorts",
"folderId": "default"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "OZWcSlohPeNaPD9Q",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "ba1d37b8-bec1-4a0e-ba06-a5ca71659284",
"name": "Update Doc",
"type": "n8n-nodes-base.googleDocs",
"position": [
880,
160
],
"parameters": {
"actionsUi": {
"actionFields": [
{
"text": "={{ $('AI Agent').item.json.output }}",
"action": "insert"
}
]
},
"operation": "update",
"documentURL": "={{ $json.id }}"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "OZWcSlohPeNaPD9Q",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "4e2aa706-c4eb-4da2-811c-4dff3116b12f",
"name": "条件判断",
"type": "n8n-nodes-base.if",
"position": [
-656,
480
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "93dde7a7-9eb8-4c16-beaf-6f66a9c08734",
"operator": {
"name": "filter.operator.equals",
"type": "string",
"operation": "equals"
},
"leftValue": "={{ $json['Choose video type'] }}",
"rightValue": "Short video"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "b34e4035-dd43-4d0b-8f89-c9cb310d4f06",
"name": "Create Outline",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
176,
624
],
"parameters": {
"text": "=Summary: {{ $json.message.content }}\n\nNow based on this summary, try to create a youtube video outline for the topic \"{{ $('Tavily1').item.json.query }}\". With proper headings (include sub headings under each main heading). Each item in the outline needs to be an important and essential and crucial information that is needed to solve the user's intent. Do not include any fluff or nice to have content. Focus on the must haves.\nAvoid the words \"section 1, 2, etc \" in the outline. Just stick to the actual headings.",
"promptType": "define"
},
"typeVersion": 1.6
},
{
"id": "fdf6dddb-44d1-411c-a807-c4bf1c8e5de7",
"name": "AI Agent1",
"type": "@n8n/n8n-nodes-langchain.agent",
"notes": "You are a senior YouTube script writer for the Website Learners channel.\nWrite a complete script with:\n ① a catchy hook (max 20 words)\n ② a friendly intro\n ③ 3‑5 body sections with smooth transitions\n ④ a final CTA to like & subscribe.\nKeep the tone casual and engaging.",
"position": [
656,
624
],
"parameters": {
"text": "=### CONTEXT\n• Video Title: {{ $('Tavily1').item.json.query }}\n• Style Reference (tone, pacing, formatting): \n{{ $json.style_ref }}\n\n• Outline with sub‑points (exact order to follow):\n{{ $('Create Outline').item.json.text }}\n\n• Requested Video Type: {{ $('On form submission').item.json['Choose video type'] }}\n\n### TASK\nWrite a YouTube script **2 500 – 3 000 words** long that:\n\n1. **Matches the requested video type automatically**\n • If `videoType` = “listicle” → produce a numbered list style (Tool #1, Tool #2 …) with snappy transitions. \n • If `videoType` = “experimental” → use a story‑driven, curiosity‑building narrative (e.g., “We pitted four tools head‑to‑head—here’s what happened…”). \n • If `videoType` = “how‑to” → write step‑by‑step instructions with clear sub‑steps and demonstration cues (screen actions, clicks, etc.).\n\n2. Intro (≤ 30 words)\n • Capture attention immediately.\n • Do **NOT** mention any host or channel name—keep it neutral.\n\n3. **Body**\n • Follow the outline’s order exactly.\n • For each tool/section:\n – Start with a heading in the form **“Tool #X – <Name>”** (or “Step #X” for how‑to videos). \n – Expand into 2 – 3 short paragraphs (~150 words) covering key features **plus one practical real‑world example**. \n – End with a 1‑sentence segue that smoothly leads into the next tool/step (e.g., “Speaking of collaboration… let’s look at Notion AI.”).\n\n4. Outro / Recap\n • Summarize main takeaways in 3‑4 sentences.\n • End with a neutral CTA: invite viewers to like, comment, and subscribe—no host or channel names.\n\n✳️ Language rule: Write at a 6th‑to‑8th‑grade reading level. \n• Use short sentences (≤ 18 words). \n• Prefer simple, common words (say “use” instead of “utilize”). \n• No jargon unless you explain it in one plain‑English sentence. \n\n### OUTPUT FORMAT\nPlain text, one blank line between paragraphs, no JSON fences.\n",
"options": {
"returnIntermediateSteps": false
},
"promptType": "define"
},
"typeVersion": 1.9
},
{
"id": "73512944-a516-4dc7-9863-1b91ab0c5c6e",
"name": "Create Doc1",
"type": "n8n-nodes-base.googleDocs",
"position": [
976,
624
],
"parameters": {
"title": "={{ $('On form submission').item.json['Provide topic'] }}",
"folderId": "default"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "OZWcSlohPeNaPD9Q",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "4d126194-61d1-4e3c-a3a2-6477db9ed0c6",
"name": "Update Doc1",
"type": "n8n-nodes-base.googleDocs",
"position": [
1152,
624
],
"parameters": {
"actionsUi": {
"actionFields": [
{
"text": "={{ $('AI Agent1').item.json.output }}",
"action": "insert"
}
]
},
"operation": "update",
"documentURL": "={{ $json.id }}"
},
"credentials": {
"googleDocsOAuth2Api": {
"id": "OZWcSlohPeNaPD9Q",
"name": "Google Docs account"
}
},
"typeVersion": 2
},
{
"id": "3215df61-ef6a-41b0-b108-bad0b8197ae1",
"name": "Tavily",
"type": "n8n-nodes-base.httpRequest",
"position": [
-352,
160
],
"parameters": {
"url": "https://api.tavily.com/search",
"method": "POST",
"options": {},
"jsonBody": "={\n \"query\": \"{{ $('On form submission').item.json['Provide topic'] }}\",\n \"search_depth\": \"basic\",\n \"max_results\": 3,\n \"include_answer\": true,\n \"include_raw_content\": true\n}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth",
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "Bearer YOUR-TAVILY-API-KEY"
}
]
}
},
"credentials": {
"httpHeaderAuth": {
"id": "3P1mixIwWxeFh266",
"name": "Header Auth account"
}
},
"typeVersion": 4.2
},
{
"id": "e0185da0-7b61-497f-bb26-c0dc7707f6c2",
"name": "Tavily1",
"type": "n8n-nodes-base.httpRequest",
"position": [
-304,
624
],
"parameters": {
"url": "https://api.tavily.com/search",
"method": "POST",
"options": {},
"jsonBody": "={\n \"query\": \"{{ $('On form submission').item.json['Provide topic'] }}\",\n \"search_depth\": \"advanced\",\n \"max_results\": 10,\n \"include_answer\": true,\n \"include_raw_content\": true\n}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth",
"headerParameters": {
"parameters": [
{
"name": "Authorization",
"value": "Bearer YOUR-TAVILY-API-KEY"
}
]
}
},
"credentials": {
"httpHeaderAuth": {
"id": "3P1mixIwWxeFh266",
"name": "Header Auth account"
}
},
"typeVersion": 4.2
},
{
"id": "d7838788-fe6a-4ea1-bfc2-dec45401012b",
"name": "Reference Script",
"type": "n8n-nodes-base.set",
"position": [
192,
160
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "c0fdf687-0dff-4e9f-9400-9c58247ce61f",
"name": "short_style_ref",
"type": "string",
"value": "SCRIPT 1: Experimental type script\n\nIntro:\n\nEver felt stuck on a task and wished someone could just guide you through it—step by step/right on your screen?\nWell, guess what? There is an AI tool which can do that. Hey google! Can you see my screen\nYes i can see your screen\nCan you help separate the audio from this video\nYes i can help you with that, just right click on the file,.....\nWow, that was pretty cool, right?\n\nIn this video, I’ll show you how to use this powerful tool step by step. Plus, I’ll put it to the test with real tasks—like working in Microsoft Excel, editing in Photoshop, or even help that I want with my PC.\n\nSo Stick around till the end, I’m A from my channel, and let’s get started!\n\n\n\nEver been stuck on a task and wished someone could just guide you through it step by step?\nWell, that’s exactly what Google can do now.\nIt can see what’s on your screen and walk you through tasks in real time. And the best part? It’s completely free.\nIn this video, I’ll show you the step-by-step process of using it and test it with real tasks like Microsoft Excel, Photoshop, and even setting up a new user on a PC.\nWatch this video till the end to see what it really delivers.\nSo this is A from my channel and let’s get started.\n\nVersion 2:\nImagine learning complex software as easily as chatting with a mentor who knows exactly what you need. Sounds unreal? Well, it’s not—because that future is here.\nGoogle has quietly launched an AI tool that’s changing the game. This isn’t just another chatbot or tutorial—it’s an adaptive, intelligent assistant that slashes your learning curve and skyrockets your productivity. And the best part? It’s completely free.\nIn this video, we’re diving deep into Google AI Studio to show you how it’s already making an impact. We’ll walk you through how it works, what you can do with it, and how it can make learning faster and easier than ever.\nStay with me, because in the next 10 minutes, you’re about to discover a learning hack so powerful, you’ll wonder how you ever lived without it.\nSo this is A from my channel and let’s get started.\n----------------------------------------------------------------------------------------------------------\nFirst, let’s go to Google AI studio,\nJust click the link in the description\nAnd it’ll take you to this page\nClick ‘sign in to Google Studio’ and sign in using your google account.\nNow go to “stream real-time”\n\nHere, you have three ways to interact with Google AI studio. You can talk to Google using your microphone, or show what you’re looking at through your webcam, or even share your screen so it can guide you step by step.\nBut before you start, remember that Google AI records your screen and microphone while you use it. Google uses this data to train their AI model, so make sure you don’t have any confidential information on your screen.\nI’m going to share my screen and see how it guides me.\nYou can click here and select the screen you want to share.\n—--------------------------------------------------------------------------------------------------------------------\nA: Hey Google, I need some help with Excel, and I just shared my screen.\nGoogle: OK, I see. You need some help with Excel. Is there anything specific you need help with?\nA: Yes! Can you see my screen right now?\nGoogle: Yes, I can see your screen. I see a spreadsheet with sales data. What would you like to do?\nA: I want to create a pivot table to analyze the sales performance of each country. Can you guide me through it step by step?\nGoogle: Sure! First, please select all of the data in your spreadsheet, including the headers. Have you done that?\nA: All right, I’ve selected all the data. What should I do next?\nGoogle: Great! Now that you’ve selected all the data, click on the Insert tab at the top of the screen, and then select the Pivot Table option.\nA: OK, done. What’s next?\nGoogle: I see you’ve opened the Create Pivot Table window. Now, select whether you want to insert the pivot table into a new sheet or an existing sheet, \nA: Let it be in new sheet\n\nGoogle: Then click Create.\nA: Yes, I did that.\nGoogle: Now, since you want to analyze sales performance by country, in the Pivot Table Editor, under Rows, click Add and choose the Country field. Then, under Values, click Add and select the Sales field.\nA: OK, done.\nGoogle: Great! You have now created a pivot table to find the sales per country. Is there anything else I can help you with?\nWow, that was crazy! Google just watched our screen, understood what we needed and helped us solve the problem step by step.\nHere’s a quick note: Each session lasts only 10 minutes and will automatically stop. If your problem isn’t solved, don’t worry, you can just start a new session and give a quick recap of what you’ve done. Since Google doesn’t remember past chats, this helps it pick up where you left off and continue helping you.\nOkay, Let’s start a new one.\nNow, let's see how Google can assist with another tool. This time, I’ll ask for help with Photoshop.\nSo let’s go back to Google Studio.\nClick here to start the session.\nI’m going to share my screen again.\nHey google, I need some help with Photoshop. Can you see my screen?\nGoogle: \nA: I want to remove this person from this picture. Can you tell me how to do that step by step?\nGoogle: Sure! To remove the person in the background, first, select the Object Selection Tool from the toolbar on the left side of the screen. It looks like an arrow pointing to a rectangle with a dotted outline. Do you see the tool?\nA: Yes. I’ve selected the “object selection tool”. Now what should I do next?\nGoogle: Now, click on the person you want to remove.\nA: OK.\nGoogle: Once you click on the person, a selection of dotted lines will appear around them. Did that happen?\nA: Yeah, I can see the dotted lines around the person.\nGoogle: Now press the Delete key on your keyboard. Photoshop should automatically fill in the space with Content-Aware Fill.\nA: Wow! The person is gone. That was super easy.\nGoogle: You're welcome! Is there anything else I can assist you with today?\nOkay, let’s try something different. I’ll ask Google to guide me through creating another user on this PC. \nLet me go back to Google Studio and share my screen.\nA: Hey Google, I need to add a new user to my windows 10 PC. Can you help me with that?\nGoogle: Yes! To add a new user on your Windows PC, first, open the Settings app. Click on the Start button at the bottom left of your screen, it’s the Windows logo,\nthen click the gear icon.\nA: OK, done. next?\nGoogle: Now, click on the Accounts icon. (It looks like two people.)\nA: Got it. I’m in the Accounts section. What’s next?\nGoogle: Now, on the left side of the screen, under Accounts, click on Family & Other Users.\nA: OK.\nGoogle: Now that you're on the Family & Other Users page, you should see a button that says Add someone else to this PC. Can you see that?\nA: Yes, I see it.\nGoogle: Great! Click on that button to continue creating a new user account.\nGoogle: Now, it’s asking for a username and a password. Type in the username you’d like and a password, then click Next.\nA: okay, Looks like the new user account has been created.\nGoogle: Great! Is there anything else I can help you with?\nA: Nope, that’s it. Thank you!\nAlright, that was pretty cool! \nSo, after using Google AI studio, I must say it’s super helpful in learning how to use a new software. It gives clear guidance and makes things easier. Of course it’s not flawless. Sometimes, it might provide incorrect steps, but if you notice an issue, just point it out, and it will correct itself. \nBut overall, it does a great job walking you through each step.\nSo that’s it guys.\n(This is how you can use Google AI to learn any new software.)\nHope you liked this video.\nIf you found this video helpful, do give it a thumbs up and subscribe to my channel. \nThanks for watching\nI’ll see you in the next video\nTake care \nBye - Bye…\n\n\nSCRIPT 2: “Listicle” type script\n\nIntro:\n\n\nAI video generation is evolving rapidly, and the competition among platforms is fiercer than ever. So, I decided to put them to the test. In this video, I’ll be using the same image and prompt to create videos with each tool, comparing their quality and performance. Along the way, I’ll showcase some practical real-world use cases, including incredible lip-syncing tools and more.\nStay tuned for some jaw-dropping results!\nThis is A from my channel and let’s get started.\n\nRunwayml\nFirst, we have runway ml\nTo use the tool\nClick the link in the description\nIt’ll take you to this page.\nNow Click ‘try runway.’\nAnd sign in using your Google account.\nOnce you’re signed in.\nHere, you'll find different features available on RunwayML.\nNow to create your video \nClick ’generate video.’\nHere we have three main models: you can choose the one you want.\nI’m going to select “Gen 3 Alpha Turbo” since it’s the fastest.\nSo let’s select that.\nNow to create your video. \nYou have two options from text to video and image to video.\nWe are going to create our video using “image to video”.\nSo We’re going to upload an image and then convert it into a video.\nlet’s drag and drop the image \nThen click “crop” to fit the image on the screen.\nHere, you can describe how you want your video to be\nThis helps the AI understand what you want and give you the best results.\n(So I’m going to ask it to create\n“A cinematic wide establishing shot of a man walking through battlefield, with soldiers moving in the background.”)\nOnce you’ve given the prompt,\nyou can choose the video duration here,\nthen click ‘generate.’\nAnd your video will be generated.\nAs you can see, the video has come out really well and the quality is also good.\nSo to download this video\nJust click here and it’ll be downloaded to your computer.\nSo now we have seen how to generate videos from Runway and what kind of video it can create.\nKling AI\nNext, let’s move to the second tool which is Kling AI\nTo use the tool\nClick the link in the description,\nand it’ll take you to this page,\nNow click “sign in for free credits”/ sign in\nFill in these details and create your account.\nNow go to AI videos,\nLike the previous tool, Kling AI also has two options for creating videos — text to video and image to video.\nLet’s select ‘image to video.’\nNow I’m going to upload the same image,\nEnter the same prompt \nAnd Click ‘generate.’\nLet’s see how this tool generates the video\nNow this process will take some time \njust wait for a while\nAnd it's done\nThe video looks pretty good, and compared to the previous tool, the quality is much better. You can also see the movements of the soldiers here.\nLet’s download it by clicking here.\nAnd your video will be downloaded.\nLuma AI\nOkay, let’s move on to the third tool, which is Luma AI.\nTo use the tool\nClick the link in the description \nAnd it’ll take you to this page\nNow click “try now.”\nAnd Sign in with your Google account.\nOnce you’re signed in\nLet’s click here and upload the same image and prompt we used earlier.\nthen click here\nAs you can see, the video has been generated and it looks pretty decent, actually.\nSo to download this video, just click here\nAnd it’ll be downloaded to your computer. \nAlright! \nMinimax AI\nLet’s move to the fourth tool, which is Minimax AI\nTo use the tool\nClick the link in the description \nAnd Sign in with your Google account.\nThen click here and upload the same image and prompt we used earlier.\nNow click here and your video will be ready.\nWell, the video looks good. \nLet’s download it.\nOkay \nNow that we've generated the videos using all four tools, let’s compare the results.\nComparison of results:\n\nRunway did a great job with good video quality, following the prompt of the man walking in a Warfield, but the soldiers in the background aren’t moving.\nKling AI is more realistic than RunwayML. You can see the soldiers walking, though the video angle could be better. Overall, it’s great.\nLuma AI didn’t get it right, actually. The characters at the back don’t look real and the video quality is also poor.\nFinally, Minimax AI is almost perfect. The visuals are realistic, and the movements are smooth.\n\n\nAccording to me When it comes to quality and relevance to the prompt, Kling AI and minimax are equally good and stand out as the best.\nExample 2:\nNext, let’s look at a few more examples and see how these tools perform in different scenarios.\nI have uploaded an image of a girl and asked these tools to create a video of a model filming a commercial with a beautiful smile and flowing hair.\nLet’s see the results.\n\nActually, Kling AI just nailed it here. It captures the smile perfectly, and the video quality is great.\nRunwayML’s video is also pretty good, but the smile is lacking. Also the hair movement feels a bit stiff compared to Kling AI.\nLuma AI’s video looks okay. The video has some movements, but the model’s face and lips look unnatural. \nMinimax’s output looks great, and it follows the prompts accurately.\nAgain, Kling AI and Minimax did well here…\nExample 3:\nAlright, we’ll see one more example\nThis time, I’ve uploaded an image of a man running on the seashore and asked the tools to make a video with details like waves touching his feet and water splashing around him.\nSo let’s see the results.\nThe video which Runway generated turned out pretty good, but the detailing could’ve been better. You can see the waves aren’t touching his feet, and his legs also have some disturbance.\nKlingAI did an awesome job. You can see the water splashing as he runs, and the details on the sand are also very good.\nMinimax’s video quality is great. Although it doesn't fully follow the prompt, the video still turned out really well.\nLuma AI also did a great job with the video, but the quality is quite low.\nOkay, now that we’ve compared the top 4 tools.\nSo when it comes to the overall ranking, I would say KlingAI is first, minimax second, then RunwayML and Luma AI… \nNow these AI tools are not just for creating fun videos,\nYou can also use it for your business.\nHere’s an ad I made using Kling AI and minimax.\nIt’s impressive, right!\nBut wait, there’s more.\nI’ve got a few other interesting AI tools that you should definitely check out.\nLet’s go through them one by one.\nHedra AI\nSo first, we have Hedra AI.\nThis tool can make any image talk by syncing it with audio, just like this.\nTo use the tool, \nClick the link in the description, and it will take you to Hedra AI.\nOnce you're there, \nClick on ‘Create’ and sign in with your Google account.\nNow go to import audio,\ndrag and drop the audio file here\nAnd you can adjust the length of the audio by dragging it like this. \nThen upload the image to which you want to sync the audio, \nand click ‘Generate.’ \nAs you can see, it has turned your image into a video with good lip sync.\nGW promo\nOkay guys, before we move on, let me show you a tool that’ll be a huge time-saver if you’re looking to write blogs.\n\nIt's called GravityWrite, \nJust enter your title, click 'create outline,' and you've got an outline.\n\nHit 'generate all,' and in seconds, you have a complete blog with stunning images.\n\nVisit gravitywrite.com, and get it for one full year at just $97!\n\nKrea AI\nNext, we have an interesting tool which can turn your images into a video with smooth transitions like this\nSo to check it out, just click the link in the description, \nand it’ll take you to Krea AI.\nOnce you're there, \ngo to the ‘Videos’ section \nand sign in with your Google account. \nThen click ‘Start’ and click ‘Open App.’\nHere, you’ll find some default images and their prompts\nSo first, let’s delete these,\nJust hover on the image and click on this delete symbol.\nOnce you’ve deleted it\nYou can set the duration of the video here\nI'm going to set it to 9 seconds\nNow to add the image\nClick on the timeline, \nthen click here and upload the image \nNow in the same way, you can add all the images you want.\nJust leave some space between the images to create the transitions in the video.\nOnce you’ve added the images, \nNext, you need to describe how the image has to be transformed.\nSo just click here and enter the prompt.\nIn the same way, you can add the prompt to the other images.\nOnce you’ve given the prompt,\nChoose a style for your video\nThen click here, select the aspect ratio,\nAnd click ‘generate.’\nAs you can see, your video has been generated.\nActually, the video has come out well and the transitions between the images are also good.\nNow to download the video\nJust click here\nAnd it’ll be downloaded to your computer. Alright! \nViggel AI\nLet’s check out the next tool, viggle AI.\nWith this tool, you can replace any character in a video. Imagine yourself acting in your favorite movie scene, just like this.\nAmazing right!\nTo know how to do that.\nJust click the link in the description\nAnd it’ll take you to viggle AI\n\nThis tool is available on multiple platforms. \nI’m going to try it on the web. \nSo let’s click here \nand Sign in with your Google account.\nHere you need to upload the video in which you want to change/replace the character.\n\nBefore you upload, just make sure your video meets these conditions.\nYour video should have only one character, \nand nothing blocking or hiding that character. Also, avoid using footage with dark scenes, and for best results keep the video under 30 seconds.\nThis is the video which I’m going to upload,\nSo Let’s drag and drop it here.\nNext, upload the image of the character you want to replace in the video.\n\nMake sure to Use a full-body photo with a clear background; the image size shouldn’t be more than 10MB.\nLet’s Drag and drop the image here, and Click ‘Create.’\nAnd your video will be generated.\nYou can see that the character has been replaced with my image. There are a few glitches here and there, but the result is still impressive.\nYou can download the video by clicking here and it’ll be saved to your computer.\nWsk promo\nBefore we move to the next tool, I wanted to quickly mention a great web hosting platform called WebSpaceKit that makes it super easy to build and launch WordPress sites. With WebSpaceKit, you get access to a free domain, free SSL, and have 1000s of free themes and plugins available to really customize your site. If you're looking to start a new WordPress project or want an easier way to manage your existing sites, definitely check out WebSpaceKit. You can even get 10% off by going through the link in the description or pinned comments.\nVidu AI\nAlright, now let’s dive into our final tool. \nThis amazing tool allows you to turn your images into those viral hugging videos that have been trending online.\nTo use this tool\nClick the link in the description, \nand it’ll take you to “Vidu AI”\nNow click ‘try vidu’\nAnd Sign in with your Google account.\nNow Go to create videos\nHere, you need to Upload an image FEATURING two people together like this. \nYou can create an image like this using Canva, where you add two different photos together and download it as a single image.\nThen, describe what actions they’ve to do in the video\nI’ve asked to make them hug each other.\nAnd then click ‘create.’\nAs you can see, it has generated a video of them hugging each other.\nOutro \nSo, that’s it, guys! \nThese are the best AI tools I’ve found for video creation. \nCheck them out, and let us know which one’s your favorite in the comments below.\nBefore we wrap up, I wanted to remind you about two incredible tools that can help you take your content creation to the next level. First, there's GravityWrite, an AI writing assistant that generates high-quality content based on easy-to-use templates. And second, WebSpaceKit a web hosting platform that simplifies the process of launching and building a WordPress site. Links to the tools are given below.\nIf you liked this video, do give it a thumbs-up.\nAnd don’t forget to subscribe to my channel for more videos like this one.\nThanks for watching.\nI’ll see you in the next video.\nTake care.\nBye-Bye...\n\n\nSCRIPT 3: “How to” type script\n\nOutline:\nTo clone a web app\nWe’re going to do 4 parts\nPart 1 - Generate the Web app structure.\nStep 1: Create an account on Lovable\nStep 2: Add a screenshot of the reference web app\nStep 3: Get the prompt to create the app.\nDemo of testing the app\nPart 2: Make the App Work\nStep 1: Connect the Web App to a Database\nSign up for Supabase using GitHub.\nCreate an organization in Supabase.\nStart a new project in Supabase.\nLink the project with Supabase.\nStep 2: Get your OpenAI API Key.\nStep 3: Add the API Key to the project.\nPart 3: Customize Your Web App\nStep 1: Generate a History Page \nStep 2: Generate an Authentication/Sign-up Page\nStep 3: Generate a Landing Page\nStep 4: Edit any element in the web app.\nPart 4: Publish your web app.\nDemo: Testing the app.\n\n\n\nIntro:\n\nWhat if I told you that you could clone any web app — design, features, and all — without writing a single line of code? Sounds wild, right? Well, watch this…\n\nYep, that just happened — and it’s all thanks to a crazy-powerful AI tool I just found. \nIn this video, I’m going to show you how to clone a fully functional web application in just minutes - the user login, database, custom design, and even publishing it online with a free domain — all with zero coding!\nI’ll walk you through how to customize everything to make it your own and deploy it live with just one click.\nSo if you’ve ever dreamed of launching your own app without the tech headache, you need to see this.\nI’m A from my channel, and let’s get started!\nTo clone a web app\nWe’re going to do 4 parts\nThe first part is to “Generate the structure of the Web App”\nSo in this part, we’ll generate the complete structure of the web app, including its design and layout.\nTo do that\nWe’re going to do 3 steps\nThe first step is to “Create an account on Lovable.”\nThis is the AI tool we’re going to use to generate a web app.\nSo to create an account,\nClick the link in the description,\nNow click ‘sign up.’\nAnd sign in with your Google account.\nOkay, so now we’ve successfully created our account on Lovable.\nOnce you’ve created your account.\nLet’s go to step 2, which is to “Add a screenshot of the reference app.”\nTo clone any web app, you need to provide the AI tool with a reference image. This helps the AI analyze the design and layout, so it can create an accurate replica of the app.\nFirst, you need to take a screenshot of the app you want to clone.\nIn this video, I’m going to clone 11 Labs, a popular text-to-speech AI tool.\nSo let’s go to Elevenlabs and take a screenshot of it.\nIf you’re using Windows, just press Windows + Print Screen. For Mac, you can press Command + Shift + 4\nOnce you’ve taken the screenshot,\nGo back to lovable\nClick here and upload the screenshot.\nOnce you’re done,\nNext we need to add the prompt to guide the AI to clone/generate the app for you.\nSo, let’s go to step 3, which is to “Get the prompt to create the app.”\nTo generate the app, we need to give clear instructions to Lovable AI about what it should do. The more details we provide, the better the AI will understand and build the app exactly how we want it.\nTo make this easier, I’ve created a ChatGPT prompt that will act as a chatbot and help us write the perfect description for our app. \nI’ve given the prompt in the description,\nJust copy the prompt\nThen go to chatgpt and paste it here\nAnd click send.\nNow ChatGPT will act as an AI assistant and ask you a few simple questions about the app you’re cloning, like the type of app and its primary function.\nJust answer them one by one.\nFirst, it will ask the type of application that we’re cloning,\nSince I’m creating a text-to-speech web application. I’ll enter that.\nThen, it will ask about the primary function of the app\nI’m going to enter, “It’ll convert text to speech with different voice models.”\nNext, ChatGPT will provide a list of the 10 best API providers and ask us to select one. This will vary depending on the type of app you’re trying to create. Since I’m cloning a text to speech tool, the AI cannot build the backend for the app. That’s why we need to select a text to speech API. \nIf you want to use another API that is not on the list, you can mention it/them here.\nI’m going to go with ‘open AI’, since it provides high-quality voice generation compared to the others.\nNow based on your answers, ChatGPT will generate the detailed prompt to generate your webapp.\nOnce you’ve got the prompt, we can generate the web app using Lovable\nSo to generate the web app, \njust copy the prompt, go back to Lovable AI, and paste it here.\nThen, click here to start generating the web app.\nAs you can see, the tool has started writing the code for our app.\nOnce it's done, \nYou can see our web application is ready, and it looks exactly like the screenshot we provided.\nSo now we’ve successfully generated the structure of our app.\nNext, let’s test the app and see if it works\nJust enter a text into the input field and click Generate.\nAs you can see, The app is not working and has not generated the voice. \nLet’s ask Lovable AI why the app is not working and how to solve this. You can ask your question in a chat over here.\nAs you can see, it’s saying that API integration is not done. \nIf you see here, it also suggests the solution.\nTo make the app work, we need to store the open AI API key securely in a database and integrate the open AI API Key. \nSo, why do we need to do this? If we add the API key directly to our app, it will be visible to the public in the source code. This means anyone accessing the site could misuse it. So to keep the API key secure, we need to set up a backend server for our web app.\nNext, let’s go to part 2 of this tutorial, which is to “Create the backend for our app.”\nA backend is essential to making any app fully functional. It handles storage, API integrations, and all the core functionalities. Without a backend, the app won’t work properly.\nSo, to create a backend for our app\nFirst we need to “Connect our web app to Supabase.”\nSupabase is a backend integration feature available in Lovable,\nwhich lets us manage the database, user login, and file storage easily.\nSo, to connect it\nGo to supabase and click ‘connect supabase’\n(Again click ‘connect supabase’)\nAnd sign in with your GitHub account\nNow click ‘authorize supabase’\nAnd it will ask you to create an organization.\nThis will be needed to connect supabase to your web app.\nSo let’s click ‘create organization.’\nClick ‘’create” again.\nThen click ‘authorize lovable’\nNow we’ve linked supabase to lovable.\nTo finish setting up the supabase connection,\nWe need to create a project in supabase.\nThis is where all your app’s data will be securely stored and managed.\nTo create it,\nGo to “connect to a project” \nAnd click “Create new project.”\nHere, give a name for the project\nAdd a password \nand click ‘Create a project.’\nNow, we’ve created the project \nNext, we need to connect the project/this to our app.\nTo connect it, go to supabase \nClick here and you’ll see the project which you’ve created.\nclick ‘connect’\nAnd click ‘connect again.’\nSo now we’ve successfully connected our web app to supabase.\nLet’s ask Lovable for the next steps to make our app work. \nAs you can see, it’s suggesting that we need the OpenAI API key to enable the app’s text-to-speech functionality.\nBut remember, these steps might be different depending on the type of app you’re creating.\nNext, let’s see how to get the open ai api key\nTo get the api key \nClick here\nIt’ll take you to open ai\nNow click ‘login’ \nAnd sign in with your Google account.\nHere, click ‘create new secret key’\nThe,n enter a name and click create key \nAnd the api key will be created.\nOnce you’ve got the API key\nTo add this key to your app,\nCopy it, and go back to “lovable.\nClick ‘Add API Key.’ \nPaste the key into the input field and click ‘Submit.’ \nOnce you’ve added the api key, \nThe tool will secure your/the API key in the database.\nAnd it’s done. \nSo now we’ve successfully created the backend for our app and it’s ready to use.\nTesting the app\nOkay, let’s test our app and see how it works.\nFirst, we’ll enter the text in this field and click ‘generate’\nThe voice has been generated. Let’s play it\nWell, that sounds great, actually! We can download the audio by clicking here.\nWe can also try out different voice models. Let’s select a new one and adjust the stability setting to make the voice sound more natural.\nAnd click generate.\nThere it is; the audio has been generated with the selected voice.\nNow that we’ve tested our app, it’s working perfectly.\nNext, let’s go to part 3 of this tutorial, which is to “Customize your web app”\nNow, if you see here, our app doesn’t have the option to store the generated audio files.\nHistory page \nNext, let’s see how to create a history page on our app where all the generated audio files are displayed, like this. \nSo to create the history page,\nLet’s go to the chat \nI’m going to instruct lovable to create a history page where I can see all the generated audio and ask it to store them in supabase (bucket).\nAs you can see, the tool has started writing SQL commands to create a Storage Bucket for the Audio files.\nOnce it’s done\nWe’ll get the “apply changes” button here\nClick on it \nAs you can see, the history page has been added to our app.\nTo check if the history page works, let’s generate a new voice. \nThen, if we refresh this page and go to history\nYou can see that, we have the generated audio file.\nNow that we’ve successfully created a history page for our app.\nLet’s confirm/test if the audio files are stored properly in the database.\nTo check that\nLet’s go to supabase dashboard\nGo here and click ‘storage’\nHere, there will be a folder called Audio Files, where all the generated audio files will be displayed.\nOkay\nRight now, anyone can access and use our web app. But what if we want only registered users to access it?\nSignup page\nNext, let’s see how to create the authentication page for our app, which is basically a sign-in page for the users.\nTo create it\nLike we did before, let’s instruct the tool to create the authentication page for our web.\nThen click here\nYou can see that, the tool has started working on it, and it also created a table to store user data.\nOnce that’s done\nClick ‘apply changes’\nAnd you can see that, the signup page has been added to our web app.\nLet’s sign up \nFirst, go to the signup page\nEnter the email and password\nThen click ‘sign up’\nAnd that’s it we’ve signed up.\nLet’s check if our user details are stored on the database\nTo do that\nLet’s click ‘users’\nHere you can see that, the user details are stored here\nSo now, any registered user can log in to the app and access it securely.\nNow, we have created the authentication page for our web app.\nAlright!\nLanding page\nNext, let’s create a landing page for our web app.\nThis is the first page users will see when they visit your app, so it’s important to make it visually appealing.\nSo to create the landing page\nLet’s instruct lovable to add a landing page with the features required and click here\nNow you can see that, our landing page has been created.\nIf we click this button, \nIt’ll take us to the sign up page.\nSo now we’ve successfully created the landing page for our web app.\nLet’s say you want to change any element on your app\nHow will you do that?\nCustomize the app\nNext, let’s see how you can customize the elements on your app.\nTo do that\nGo to ‘edit’ here\nNow click the element which you want to change \nI’m going to change this content \nYou’ll see an editing section like this \nJust make the changes here and it’ll be added to your app.\nNow, let’s say you want to change the color of this button. Just click on it, and you’ll get options to adjust the color and even the size.\nIn the same way, you can customize any element on your website.\nOkay, once you’re done with the customization\nPublish the app\nLet’s go to the final part, which is to “publish your web app.”\nOnce you publish your web app, it’ll be live on the internet, and anyone can access it from anywhere.\nTo publish the app\nJust click ‘publish’\nAnd you’ll be given a domain for your web app.\nClick ’publish’ again\nAnd your app will be published and live on the internet. \nIf you click here, it’ll take you to the app\nNow, let’s check if everything works fine\nClick ‘try for free’\nAnd then sign in\nEnter the text, click ‘generate’\nAnd the voice will be generated.\nSo we’ve successfully published our app and made it live on the internet.\nSo that’s it, guys\nThis is how you can clone any web app using AI.\nHope you liked this video\nMake sure you subscribe to my channel for more videos like this one\nThanks for watching \nI’ll see you in the next video\nTake care\nBye - Bye…"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "e90e2682-e20e-454e-bf49-3e715fcee61f",
"name": "Reference Script1",
"type": "n8n-nodes-base.set",
"position": [
496,
624
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "241186a1-6f63-419a-8eaa-519a51b2bda3",
"name": "style_ref",
"type": "string",
"value": "Intro: AI video generation is evolving rapidly, and the competition among platforms is fiercer than ever. So, I decided to put them to the test. In this video, I’ll be using the same image and prompt to create videos with each tool, comparing their quality and performance. Along the way, I’ll showcase some practical real-world use cases, including incredible lip-syncing tools and more. Stay tuned for some jaw-dropping results! This is A from my channel and let’s get started. Runwayml First, we have runway ml To use the tool Click the link in the description It’ll take you to this page. Now Click ‘try runway.’ And sign in using your Google account. Once you’re signed in. Here, you'll find different features available on RunwayML. Now to create your video Click ’generate video.’ Here we have three main models: you can choose the one you want. I’m going to select “Gen 3 Alpha Turbo” since it’s the fastest. So let’s select that. Now to create your video. You have two options from text to video and image to video. We are going to create our video using “image to video”. So We’re going to upload an image and then convert it into a video. let’s drag and drop the image Then click “crop” to fit the image on the screen. Here, you can describe how you want your video to be This helps the AI understand what you want and give you the best results. (So I’m going to ask it to create “A cinematic wide establishing shot of a man walking through battlefield, with soldiers moving in the background.”) Once you’ve given the prompt, you can choose the video duration here, then click ‘generate.’ And your video will be generated. As you can see, the video has come out really well and the quality is also good. So to download this video Just click here and it’ll be downloaded to your computer. So now we have seen how to generate videos from Runway and what kind of video it can create. Kling AI Next, let’s move to the second tool which is Kling AI To use the tool Click the link in the description, and it’ll take you to this page, Now click “sign in for free credits”/ sign in Fill in these details and create your account. Now go to AI videos, Like the previous tool, Kling AI also has two options for creating videos — text to video and image to video. Let’s select ‘image to video.’ Now I’m going to upload the same image, Enter the same prompt And Click ‘generate.’ Let’s see how this tool generates the video Now this process will take some time just wait for a while And it's done The video looks pretty good, and compared to the previous tool, the quality is much better. You can also see the movements of the soldiers here. Let’s download it by clicking here. And your video will be downloaded. Luma AI Okay, let’s move on to the third tool, which is Luma AI. To use the tool Click the link in the description And it’ll take you to this page Now click “try now.” And Sign in with your Google account. Once you’re signed in Let’s click here and upload the same image and prompt we used earlier. then click here As you can see, the video has been generated and it looks pretty decent, actually. So to download this video, just click here And it’ll be downloaded to your computer. Alright! Minimax AI Let’s move to the fourth tool, which is Minimax AI To use the tool Click the link in the description And Sign in with your Google account. Then click here and upload the same image and prompt we used earlier. Now click here and your video will be ready. Well, the video looks good. Let’s download it. Okay Now that we've generated the videos using all four tools, let’s compare the results. Comparison of results: Runway did a great job with good video quality, following the prompt of the man walking in a Warfield, but the soldiers in the background aren’t moving. Kling AI is more realistic than RunwayML. You can see the soldiers walking, though the video angle could be better. Overall, it’s great. Luma AI didn’t get it right, actually. The characters at the back don’t look real and the video quality is also poor. Finally, Minimax AI is almost perfect. The visuals are realistic, and the movements are smooth. According to me When it comes to quality and relevance to the prompt, Kling AI and minimax are equally good and stand out as the best. Example 2: Next, let’s look at a few more examples and see how these tools perform in different scenarios. I have uploaded an image of a girl and asked these tools to create a video of a model filming a commercial with a beautiful smile and flowing hair. Let’s see the results. Actually, Kling AI just nailed it here. It captures the smile perfectly, and the video quality is great. RunwayML’s video is also pretty good, but the smile is lacking. Also the hair movement feels a bit stiff compared to Kling AI. Luma AI’s video looks okay. The video has some movements, but the model’s face and lips look unnatural. Minimax’s output looks great, and it follows the prompts accurately. Again, Kling AI and Minimax did well here… Example 3: Alright, we’ll see one more example This time, I’ve uploaded an image of a man running on the seashore and asked the tools to make a video with details like waves touching his feet and water splashing around him. So let’s see the results. The video which Runway generated turned out pretty good, but the detailing could’ve been better. You can see the waves aren’t touching his feet, and his legs also have some disturbance. KlingAI did an awesome job. You can see the water splashing as he runs, and the details on the sand are also very good. Minimax’s video quality is great. Although it doesn't fully follow the prompt, the video still turned out really well. Luma AI also did a great job with the video, but the quality is quite low. Okay, now that we’ve compared the top 4 tools. So when it comes to the overall ranking, I would say KlingAI is first, minimax second, then RunwayML and Luma AI… Now these AI tools are not just for creating fun videos, You can also use it for your business. Here’s an ad I made using Kling AI and minimax. It’s impressive, right! But wait, there’s more. I’ve got a few other interesting AI tools that you should definitely check out. Let’s go through them one by one. Hedra AI So first, we have Hedra AI. This tool can make any image talk by syncing it with audio, just like this. To use the tool, Click the link in the description, and it will take you to Hedra AI. Once you're there, Click on ‘Create’ and sign in with your Google account. Now go to import audio, drag and drop the audio file here And you can adjust the length of the audio by dragging it like this. Then upload the image to which you want to sync the audio, and click ‘Generate.’ As you can see, it has turned your image into a video with good lip sync. GW promo Okay guys, before we move on, let me show you a tool that’ll be a huge time-saver if you’re looking to write blogs. It's called GravityWrite, Just enter your title, click 'create outline,' and you've got an outline. Hit 'generate all,' and in seconds, you have a complete blog with stunning images. Visit gravitywrite.com, and get it for one full year at just $97! Krea AI Next, we have an interesting tool which can turn your images into a video with smooth transitions like this So to check it out, just click the link in the description, and it’ll take you to Krea AI. Once you're there, go to the ‘Videos’ section and sign in with your Google account. Then click ‘Start’ and click ‘Open App.’ Here, you’ll find some default images and their prompts So first, let’s delete these, Just hover on the image and click on this delete symbol. Once you’ve deleted it You can set the duration of the video here I'm going to set it to 9 seconds Now to add the image Click on the timeline, then click here and upload the image Now in the same way, you can add all the images you want. Just leave some space between the images to create the transitions in the video. Once you’ve added the images, Next, you need to describe how the image has to be transformed. So just click here and enter the prompt. In the same way, you can add the prompt to the other images. Once you’ve given the prompt, Choose a style for your video Then click here, select the aspect ratio, And click ‘generate.’ As you can see, your video has been generated. Actually, the video has come out well and the transitions between the images are also good. Now to download the video Just click here And it’ll be downloaded to your computer. Alright! Viggel AI Let’s check out the next tool, viggle AI. With this tool, you can replace any character in a video. Imagine yourself acting in your favorite movie scene, just like this. Amazing right! To know how to do that. Just click the link in the description And it’ll take you to viggle AI This tool is available on multiple platforms. I’m going to try it on the web. So let’s click here and Sign in with your Google account. Here you need to upload the video in which you want to change/replace the character. Before you upload, just make sure your video meets these conditions. Your video should have only one character, and nothing blocking or hiding that character. Also, avoid using footage with dark scenes, and for best results keep the video under 30 seconds. This is the video which I’m going to upload, So Let’s drag and drop it here. Next, upload the image of the character you want to replace in the video. Make sure to Use a full-body photo with a clear background; the image size shouldn’t be more than 10MB. Let’s Drag and drop the image here, and Click ‘Create.’ And your video will be generated. You can see that the character has been replaced with my image. There are a few glitches here and there, but the result is still impressive. You can download the video by clicking here and it’ll be saved to your computer. Wsk promo Before we move to the next tool, I wanted to quickly mention a great web hosting platform called WebSpaceKit that makes it super easy to build and launch WordPress sites. With WebSpaceKit, you get access to a free domain, free SSL, and have 1000s of free themes and plugins available to really customize your site. If you're looking to start a new WordPress project or want an easier way to manage your existing sites, definitely check out WebSpaceKit. You can even get 10% off by going through the link in the description or pinned comments. Vidu AI Alright, now let’s dive into our final tool. This amazing tool allows you to turn your images into those viral hugging videos that have been trending online. To use this tool Click the link in the description, and it’ll take you to “Vidu AI” Now click ‘try vidu’ And Sign in with your Google account. Now Go to create videos Here, you need to Upload an image FEATURING two people together like this. You can create an image like this using Canva, where you add two different photos together and download it as a single image. Then, describe what actions they’ve to do in the video I’ve asked to make them hug each other. And then click ‘create.’ As you can see, it has generated a video of them hugging each other. Outro So, that’s it, guys! These are the best AI tools I’ve found for video creation. Check them out, and let us know which one’s your favorite in the comments below. Before we wrap up, I wanted to remind you about two incredible tools that can help you take your content creation to the next level. First, there's GravityWrite, an AI writing assistant that generates high-quality content based on easy-to-use templates. And second, WebSpaceKit a web hosting platform that simplifies the process of launching and building a WordPress site. Links to the tools are given below. If you liked this video, do give it a thumbs-up. And don’t forget to subscribe to my channel for more videos like this one. Thanks for watching. I’ll see you in the next video. Take care. Bye-Bye...\n\nSCRIPT 1: Experimental type script\nIntro:\n\n\nEver felt stuck on a task and wished someone could just guide you through it—step by step/right on your screen?\nWell, guess what? There is an AI tool which can do that. Hey google! Can you see my screen\nYes i can see your screen\nCan you help separate the audio from this video\nYes i can help you with that, just right click on the file,.....\nWow, that was pretty cool, right?\n\nIn this video, I’ll show you how to use this powerful tool step by step. Plus, I’ll put it to the test with real tasks—like working in Microsoft Excel, editing in Photoshop, or even help that I want with my PC.\n\nSo Stick around till the end, I’m A from my channel, and let’s get started!\n\n\n\nEver been stuck on a task and wished someone could just guide you through it step by step?\nWell, that’s exactly what Google can do now.\nIt can see what’s on your screen and walk you through tasks in real time. And the best part? It’s completely free.\nIn this video, I’ll show you the step-by-step process of using it and test it with real tasks like Microsoft Excel, Photoshop, and even setting up a new user on a PC.\nWatch this video till the end to see what it really delivers.\nSo this is A from my channel and let’s get started.\n\nVersion 2:\nImagine learning complex software as easily as chatting with a mentor who knows exactly what you need. Sounds unreal? Well, it’s not—because that future is here.\nGoogle has quietly launched an AI tool that’s changing the game. This isn’t just another chatbot or tutorial—it’s an adaptive, intelligent assistant that slashes your learning curve and skyrockets your productivity. And the best part? It’s completely free.\nIn this video, we’re diving deep into Google AI Studio to show you how it’s already making an impact. We’ll walk you through how it works, what you can do with it, and how it can make learning faster and easier than ever.\nStay with me, because in the next 10 minutes, you’re about to discover a learning hack so powerful, you’ll wonder how you ever lived without it.\nSo this is A from my channel and let’s get started.\n----------------------------------------------------------------------------------------------------------\nFirst, let’s go to Google AI studio,\nJust click the link in the description\nAnd it’ll take you to this page\nClick ‘sign in to Google Studio’ and sign in using your google account.\nNow go to “stream real-time”\n\nHere, you have three ways to interact with Google AI studio. You can talk to Google using your microphone, or show what you’re looking at through your webcam, or even share your screen so it can guide you step by step.\nBut before you start, remember that Google AI records your screen and microphone while you use it. Google uses this data to train their AI model, so make sure you don’t have any confidential information on your screen.\nI’m going to share my screen and see how it guides me.\nYou can click here and select the screen you want to share.\n—--------------------------------------------------------------------------------------------------------------------\nA: Hey Google, I need some help with Excel, and I just shared my screen.\nGoogle: OK, I see. You need some help with Excel. Is there anything specific you need help with?\nA: Yes! Can you see my screen right now?\nGoogle: Yes, I can see your screen. I see a spreadsheet with sales data. What would you like to do?\nA: I want to create a pivot table to analyze the sales performance of each country. Can you guide me through it step by step?\nGoogle: Sure! First, please select all of the data in your spreadsheet, including the headers. Have you done that?\nA: All right, I’ve selected all the data. What should I do next?\nGoogle: Great! Now that you’ve selected all the data, click on the Insert tab at the top of the screen, and then select the Pivot Table option.\nA: OK, done. What’s next?\nGoogle: I see you’ve opened the Create Pivot Table window. Now, select whether you want to insert the pivot table into a new sheet or an existing sheet, \nA: Let it be in new sheet\n\nGoogle: Then click Create.\nA: Yes, I did that.\nGoogle: Now, since you want to analyze sales performance by country, in the Pivot Table Editor, under Rows, click Add and choose the Country field. Then, under Values, click Add and select the Sales field.\nA: OK, done.\nGoogle: Great! You have now created a pivot table to find the sales per country. Is there anything else I can help you with?\nWow, that was crazy! Google just watched our screen, understood what we needed and helped us solve the problem step by step.\nHere’s a quick note: Each session lasts only 10 minutes and will automatically stop. If your problem isn’t solved, don’t worry, you can just start a new session and give a quick recap of what you’ve done. Since Google doesn’t remember past chats, this helps it pick up where you left off and continue helping you.\nOkay, Let’s start a new one.\nNow, let's see how Google can assist with another tool. This time, I’ll ask for help with Photoshop.\nSo let’s go back to Google Studio.\nClick here to start the session.\nI’m going to share my screen again.\nHey google, I need some help with Photoshop. Can you see my screen?\nGoogle: \nA: I want to remove this person from this picture. Can you tell me how to do that step by step?\nGoogle: Sure! To remove the person in the background, first, select the Object Selection Tool from the toolbar on the left side of the screen. It looks like an arrow pointing to a rectangle with a dotted outline. Do you see the tool?\nA: Yes. I’ve selected the “object selection tool”. Now what should I do next?\nGoogle: Now, click on the person you want to remove.\nA: OK.\nGoogle: Once you click on the person, a selection of dotted lines will appear around them. Did that happen?\nA: Yeah, I can see the dotted lines around the person.\nGoogle: Now press the Delete key on your keyboard. Photoshop should automatically fill in the space with Content-Aware Fill.\nA: Wow! The person is gone. That was super easy.\nGoogle: You're welcome! Is there anything else I can assist you with today?\nOkay, let’s try something different. I’ll ask Google to guide me through creating another user on this PC. \nLet me go back to Google Studio and share my screen.\nA: Hey Google, I need to add a new user to my windows 10 PC. Can you help me with that?\nGoogle: Yes! To add a new user on your Windows PC, first, open the Settings app. Click on the Start button at the bottom left of your screen, it’s the Windows logo,\nthen click the gear icon.\nA: OK, done. next?\nGoogle: Now, click on the Accounts icon. (It looks like two people.)\nA: Got it. I’m in the Accounts section. What’s next?\nGoogle: Now, on the left side of the screen, under Accounts, click on Family & Other Users.\nA: OK.\nGoogle: Now that you're on the Family & Other Users page, you should see a button that says Add someone else to this PC. Can you see that?\nA: Yes, I see it.\nGoogle: Great! Click on that button to continue creating a new user account.\nGoogle: Now, it’s asking for a username and a password. Type in the username you’d like and a password, then click Next.\nA: okay, Looks like the new user account has been created.\nGoogle: Great! Is there anything else I can help you with?\nA: Nope, that’s it. Thank you!\nAlright, that was pretty cool! \nSo, after using Google AI studio, I must say it’s super helpful in learning how to use a new software. It gives clear guidance and makes things easier. Of course it’s not flawless. Sometimes, it might provide incorrect steps, but if you notice an issue, just point it out, and it will correct itself. \nBut overall, it does a great job walking you through each step.\nSo that’s it guys.\n(This is how you can use Google AI to learn any new software.)\nHope you liked this video.\nIf you found this video helpful, do give it a thumbs up and subscribe to my channel. \nThanks for watching\nI’ll see you in the next video\nTake care \nBye - Bye…\n\nSCRIPT 3: “How to” type script\n\nOutline:\nTo clone a web app\nWe’re going to do 4 parts\nPart 1 - Generate the Web app structure.\nStep 1: Create an account on Lovable\nStep 2: Add a screenshot of the reference web app\nStep 3: Get the prompt to create the app.\nDemo of testing the app\nPart 2: Make the App Work\nStep 1: Connect the Web App to a Database\nSign up for Supabase using GitHub.\nCreate an organization in Supabase.\nStart a new project in Supabase.\nLink the project with Supabase.\nStep 2: Get your OpenAI API Key.\nStep 3: Add the API Key to the project.\nPart 3: Customize Your Web App\nStep 1: Generate a History Page \nStep 2: Generate an Authentication/Sign-up Page\nStep 3: Generate a Landing Page\nStep 4: Edit any element in the web app.\nPart 4: Publish your web app.\nDemo: Testing the app.\n\n\n\nIntro:\n\nWhat if I told you that you could clone any web app — design, features, and all — without writing a single line of code? Sounds wild, right? Well, watch this…\n\nYep, that just happened — and it’s all thanks to a crazy-powerful AI tool I just found. \nIn this video, I’m going to show you how to clone a fully functional web application in just minutes - the user login, database, custom design, and even publishing it online with a free domain — all with zero coding!\nI’ll walk you through how to customize everything to make it your own and deploy it live with just one click.\nSo if you’ve ever dreamed of launching your own app without the tech headache, you need to see this.\nI’m A from my channel, and let’s get started!\nTo clone a web app\nWe’re going to do 4 parts\nThe first part is to “Generate the structure of the Web App”\nSo in this part, we’ll generate the complete structure of the web app, including its design and layout.\nTo do that\nWe’re going to do 3 steps\nThe first step is to “Create an account on Lovable.”\nThis is the AI tool we’re going to use to generate a web app.\nSo to create an account,\nClick the link in the description,\nNow click ‘sign up.’\nAnd sign in with your Google account.\nOkay, so now we’ve successfully created our account on Lovable.\nOnce you’ve created your account.\nLet’s go to step 2, which is to “Add a screenshot of the reference app.”\nTo clone any web app, you need to provide the AI tool with a reference image. This helps the AI analyze the design and layout, so it can create an accurate replica of the app.\nFirst, you need to take a screenshot of the app you want to clone.\nIn this video, I’m going to clone 11 Labs, a popular text-to-speech AI tool.\nSo let’s go to Elevenlabs and take a screenshot of it.\nIf you’re using Windows, just press Windows + Print Screen. For Mac, you can press Command + Shift + 4\nOnce you’ve taken the screenshot,\nGo back to lovable\nClick here and upload the screenshot.\nOnce you’re done,\nNext we need to add the prompt to guide the AI to clone/generate the app for you.\nSo, let’s go to step 3, which is to “Get the prompt to create the app.”\nTo generate the app, we need to give clear instructions to Lovable AI about what it should do. The more details we provide, the better the AI will understand and build the app exactly how we want it.\nTo make this easier, I’ve created a ChatGPT prompt that will act as a chatbot and help us write the perfect description for our app. \nI’ve given the prompt in the description,\nJust copy the prompt\nThen go to chatgpt and paste it here\nAnd click send.\nNow ChatGPT will act as an AI assistant and ask you a few simple questions about the app you’re cloning, like the type of app and its primary function.\nJust answer them one by one.\nFirst, it will ask the type of application that we’re cloning,\nSince I’m creating a text-to-speech web application. I’ll enter that.\nThen, it will ask about the primary function of the app\nI’m going to enter, “It’ll convert text to speech with different voice models.”\nNext, ChatGPT will provide a list of the 10 best API providers and ask us to select one. This will vary depending on the type of app you’re trying to create. Since I’m cloning a text to speech tool, the AI cannot build the backend for the app. That’s why we need to select a text to speech API. \nIf you want to use another API that is not on the list, you can mention it/them here.\nI’m going to go with ‘open AI’, since it provides high-quality voice generation compared to the others.\nNow based on your answers, ChatGPT will generate the detailed prompt to generate your webapp.\nOnce you’ve got the prompt, we can generate the web app using Lovable\nSo to generate the web app, \njust copy the prompt, go back to Lovable AI, and paste it here.\nThen, click here to start generating the web app.\nAs you can see, the tool has started writing the code for our app.\nOnce it's done, \nYou can see our web application is ready, and it looks exactly like the screenshot we provided.\nSo now we’ve successfully generated the structure of our app.\nNext, let’s test the app and see if it works\nJust enter a text into the input field and click Generate.\nAs you can see, The app is not working and has not generated the voice. \nLet’s ask Lovable AI why the app is not working and how to solve this. You can ask your question in a chat over here.\nAs you can see, it’s saying that API integration is not done. \nIf you see here, it also suggests the solution.\nTo make the app work, we need to store the open AI API key securely in a database and integrate the open AI API Key. \nSo, why do we need to do this? If we add the API key directly to our app, it will be visible to the public in the source code. This means anyone accessing the site could misuse it. So to keep the API key secure, we need to set up a backend server for our web app.\nNext, let’s go to part 2 of this tutorial, which is to “Create the backend for our app.”\nA backend is essential to making any app fully functional. It handles storage, API integrations, and all the core functionalities. Without a backend, the app won’t work properly.\nSo, to create a backend for our app\nFirst we need to “Connect our web app to Supabase.”\nSupabase is a backend integration feature available in Lovable,\nwhich lets us manage the database, user login, and file storage easily.\nSo, to connect it\nGo to supabase and click ‘connect supabase’\n(Again click ‘connect supabase’)\nAnd sign in with your GitHub account\nNow click ‘authorize supabase’\nAnd it will ask you to create an organization.\nThis will be needed to connect supabase to your web app.\nSo let’s click ‘create organization.’\nClick ‘’create” again.\nThen click ‘authorize lovable’\nNow we’ve linked supabase to lovable.\nTo finish setting up the supabase connection,\nWe need to create a project in supabase.\nThis is where all your app’s data will be securely stored and managed.\nTo create it,\nGo to “connect to a project” \nAnd click “Create new project.”\nHere, give a name for the project\nAdd a password \nand click ‘Create a project.’\nNow, we’ve created the project \nNext, we need to connect the project/this to our app.\nTo connect it, go to supabase \nClick here and you’ll see the project which you’ve created.\nclick ‘connect’\nAnd click ‘connect again.’\nSo now we’ve successfully connected our web app to supabase.\nLet’s ask Lovable for the next steps to make our app work. \nAs you can see, it’s suggesting that we need the OpenAI API key to enable the app’s text-to-speech functionality.\nBut remember, these steps might be different depending on the type of app you’re creating.\nNext, let’s see how to get the open ai api key\nTo get the api key \nClick here\nIt’ll take you to open ai\nNow click ‘login’ \nAnd sign in with your Google account.\nHere, click ‘create new secret key’\nThe,n enter a name and click create key \nAnd the api key will be created.\nOnce you’ve got the API key\nTo add this key to your app,\nCopy it, and go back to “lovable.\nClick ‘Add API Key.’ \nPaste the key into the input field and click ‘Submit.’ \nOnce you’ve added the api key, \nThe tool will secure your/the API key in the database.\nAnd it’s done. \nSo now we’ve successfully created the backend for our app and it’s ready to use.\nTesting the app\nOkay, let’s test our app and see how it works.\nFirst, we’ll enter the text in this field and click ‘generate’\nThe voice has been generated. Let’s play it\nWell, that sounds great, actually! We can download the audio by clicking here.\nWe can also try out different voice models. Let’s select a new one and adjust the stability setting to make the voice sound more natural.\nAnd click generate.\nThere it is; the audio has been generated with the selected voice.\nNow that we’ve tested our app, it’s working perfectly.\nNext, let’s go to part 3 of this tutorial, which is to “Customize your web app”\nNow, if you see here, our app doesn’t have the option to store the generated audio files.\nHistory page \nNext, let’s see how to create a history page on our app where all the generated audio files are displayed, like this. \nSo to create the history page,\nLet’s go to the chat \nI’m going to instruct lovable to create a history page where I can see all the generated audio and ask it to store them in supabase (bucket).\nAs you can see, the tool has started writing SQL commands to create a Storage Bucket for the Audio files.\nOnce it’s done\nWe’ll get the “apply changes” button here\nClick on it \nAs you can see, the history page has been added to our app.\nTo check if the history page works, let’s generate a new voice. \nThen, if we refresh this page and go to history\nYou can see that, we have the generated audio file.\nNow that we’ve successfully created a history page for our app.\nLet’s confirm/test if the audio files are stored properly in the database.\nTo check that\nLet’s go to supabase dashboard\nGo here and click ‘storage’\nHere, there will be a folder called Audio Files, where all the generated audio files will be displayed.\nOkay\nRight now, anyone can access and use our web app. But what if we want only registered users to access it?\nSignup page\nNext, let’s see how to create the authentication page for our app, which is basically a sign-in page for the users.\nTo create it\nLike we did before, let’s instruct the tool to create the authentication page for our web.\nThen click here\nYou can see that, the tool has started working on it, and it also created a table to store user data.\nOnce that’s done\nClick ‘apply changes’\nAnd you can see that, the signup page has been added to our web app.\nLet’s sign up \nFirst, go to the signup page\nEnter the email and password\nThen click ‘sign up’\nAnd that’s it we’ve signed up.\nLet’s check if our user details are stored on the database\nTo do that\nLet’s click ‘users’\nHere you can see that, the user details are stored here\nSo now, any registered user can log in to the app and access it securely.\nNow, we have created the authentication page for our web app.\nAlright!\nLanding page\nNext, let’s create a landing page for our web app.\nThis is the first page users will see when they visit your app, so it’s important to make it visually appealing.\nSo to create the landing page\nLet’s instruct lovable to add a landing page with the features required and click here\nNow you can see that, our landing page has been created.\nIf we click this button, \nIt’ll take us to the sign up page.\nSo now we’ve successfully created the landing page for our web app.\nLet’s say you want to change any element on your app\nHow will you do that?\nCustomize the app\nNext, let’s see how you can customize the elements on your app.\nTo do that\nGo to ‘edit’ here\nNow click the element which you want to change \nI’m going to change this content \nYou’ll see an editing section like this \nJust make the changes here and it’ll be added to your app.\nNow, let’s say you want to change the color of this button. Just click on it, and you’ll get options to adjust the color and even the size.\nIn the same way, you can customize any element on your website.\nOkay, once you’re done with the customization\nPublish the app\nLet’s go to the final part, which is to “publish your web app.”\nOnce you publish your web app, it’ll be live on the internet, and anyone can access it from anywhere.\nTo publish the app\nJust click ‘publish’\nAnd you’ll be given a domain for your web app.\nClick ’publish’ again\nAnd your app will be published and live on the internet. \nIf you click here, it’ll take you to the app\nNow, let’s check if everything works fine\nClick ‘try for free’\nAnd then sign in\nEnter the text, click ‘generate’\nAnd the voice will be generated.\nSo we’ve successfully published our app and made it live on the internet.\nSo that’s it, guys\nThis is how you can clone any web app using AI.\nHope you liked this video\nMake sure you subscribe to my channel for more videos like this one\nThanks for watching \nI’ll see you in the next video\nTake care\nBye - Bye…\n"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "fd7a505d-d33c-457e-a382-6c2bf6bf2adb",
"name": "Google Gemini 聊天模型",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
384,
336
],
"parameters": {
"options": {}
},
"credentials": {
"googlePalmApi": {
"id": "icZMUpIbeHJl5HFg",
"name": "Google Gemini(PaLM) Api account 3"
}
},
"typeVersion": 1
},
{
"id": "0886153a-eb9d-4088-a6d6-7e954a0c8be5",
"name": "Google Gemini 聊天模型1",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
480,
848
],
"parameters": {
"options": {}
},
"credentials": {
"googlePalmApi": {
"id": "icZMUpIbeHJl5HFg",
"name": "Google Gemini(PaLM) Api account 3"
}
},
"typeVersion": 1
},
{
"id": "d0faae97-b7b6-4b65-8e40-7c6198487fef",
"name": "WebSearch Summary",
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"position": [
-128,
624
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "models/gemini-2.5-flash",
"cachedResultName": "models/gemini-2.5-flash"
},
"options": {},
"messages": {
"values": [
{
"role": "model",
"content": "=You are a research assistant and YouTube script researcher.\n\n## INPUTS\n• Topic / Video Title → {{ $('Tavily1').item.json.query }}\n• Scraped source texts → {{ JSON.stringify($node[\"Tavily1\"].json.results) }} ← n8n expression; leave unchanged.\n\n## TASK\n1. Read every source in **scraped source texts**. \n2. Extract the most relevant points that match the exact topic above—ignore off‑topic info. \n3. Create a **detailed research summary** (≈ 1000 words) in *plain English*.\n\n## LANGUAGE & TONE\n• Aim for a 6th‑to‑8th‑grade reading level. \n• Short sentences (≤ 18 words). \n• Prefer common words (“use” > “utilize”). \n• No jargon unless you explain it in one simple sentence. \n• Do **not** mention any host or channel name.\n\n## STRUCTURE\nReturn plain text with one blank line between sections:\n\n**Intro (≤ 30 words)** – a catchy hook that frames why the viewer should care.\n\n**Main Summary** – depends on the topic type \n◦ **If the topic is a *listicle*** → \n – Remove duplicate items across sources. \n – For each unique item, write: \n • Item #X – Name (bold the name) \n – Key spec / feature #1 \n – Key spec / feature #2 \n – One real‑world benefit or use case (1 sentence) \n\n◦ **If the topic is a *how‑to* guide** → \n – Compare all methods across sites, pick the best approach. \n – Present clear **Step #1, Step #2 …** instructions, adding 1‑sentence tips where helpful. \n\n◦ **If the topic is *experimental* (comparison, challenge, versus, etc.)** → \n – Summarise the experiment setup in 2‑3 sentences. \n – List the contenders / variables. \n – Highlight the most surprising finding and why it matters. \n\n**Mini‑Conclusion (2 sentences)** – wrap up value and tease that the full script will dive deeper.\n\n## OUTPUT RULES\n• Plain text only—no Markdown fences, no JSON. \n• At least **1000 words**. \n• Each bullet line starts with “– ” (en dash + space). \n• One blank line between Intro, Main Summary, and Mini‑Conclusion.\n"
},
{
"content": "={{ JSON.stringify($node[\"Tavily1\"].json[\"results\"]) }}\n"
}
]
}
},
"credentials": {
"googlePalmApi": {
"id": "icZMUpIbeHJl5HFg",
"name": "Google Gemini(PaLM) Api account 3"
}
},
"typeVersion": 1
},
{
"id": "234f5279-1eda-4b11-a9d3-070a420310f2",
"name": "Websearch",
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"position": [
-144,
160
],
"parameters": {
"modelId": {
"__rl": true,
"mode": "list",
"value": "models/gemini-2.5-flash",
"cachedResultName": "models/gemini-2.5-flash"
},
"options": {},
"messages": {
"values": [
{
"role": "model",
"content": "=You are a research assistant. Your task is to extract information from these websites scraped content for the given title {{ $('On form submission').item.json['Provide topic'] }}, analyze the information, and prepare a summarized content.\nRules:\n- Ensure the summary is based solely on the given title.\n- Generate a single, cohesive summary from the extracted information.\n- If the topic is a listicle, remove duplicate items/products and compile a summary with the remaining unique products.\n- If the information includes different methods across websites, select the best method and create a concise summary based on it."
},
{
"content": "={{ JSON.stringify($node[\"Tavily\"].json[\"results\"]) }}\n"
}
]
}
},
"credentials": {
"googlePalmApi": {
"id": "icZMUpIbeHJl5HFg",
"name": "Google Gemini(PaLM) Api account 3"
}
},
"typeVersion": 1
},
{
"id": "f40cef41-9e1d-4c43-8f9f-9904c067f674",
"name": "便签",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1584,
80
],
"parameters": {
"width": 656,
"height": 912,
"content": "# 🤖 YouTube Content Generator (Shorts & Long Videos)\nThis workflow is a content multiplier that automatically creates scripts for two different video formats from a single topic. When you submit an idea, it generates both a concise script for a YouTube Short and a detailed script for a long-form video, saving both directly to Google Docs.\n\n## 📝 Core Functions\n### The workflow splits into two specialized paths after you submit your topic:\n### Branch 1 (For YouTube Shorts):\n- This path is designed for speed and impact.\n- It performs a quick web search on the topic.\n- An AI Agent (Gemini) immediately processes the information to write a short, punchy, and engaging script perfect for the Shorts format.\n### Branch 2 (For Long-Form Videos):\n- This path is designed for depth and structure.\n- It conducts a more detailed web search and summarizes the key findings.\n- It then uses AI to generate a proper video outline (e.g., Intro, Main Points, Call to Action, Outro).\n- A final AI Agent uses this detailed outline to write a comprehensive, well-structured script suitable for a full-length YouTube video.\n## 🛠️ Setup Instructions\n### To use this workflow, you'll need to configure the following credentials:\n- On form submission Trigger: Configure this with your form or provide data manually. It should contain the video topic.\n- Tavily Nodes: Add your Tavily API key for web search capabilities.\n- Google Gemini Chat Model: Connect your Google AI credentials for the AI processing steps.\n- Google Docs Nodes: Authenticate your Google account (using OAuth2) to allow n8n to create and update documents in your Google Drive.\n## 🚀 How to Use\n- Activate the workflow.\n- Submit your video topic through the connected form.\n- Wait for the workflow to execute both branches.\n- Check your Google Drive and Docs. You will find two new documents: one script for your YouTube Short and another for your long-form video.\n\n\n"
},
"typeVersion": 1
},
{
"id": "752c776e-9b2f-485e-92c7-e0a746699eac",
"name": "便签1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-480,
48
],
"parameters": {
"color": 4,
"width": 1760,
"height": 432,
"content": "# For Youtube Short Videos"
},
"typeVersion": 1
},
{
"id": "83824059-0df4-49c6-bf2e-1258f4be79b5",
"name": "便签3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-464,
512
],
"parameters": {
"color": 4,
"width": 1760,
"height": 448,
"content": "# For Youtube Long Videos"
},
"typeVersion": 1
},
{
"id": "e6006ebf-5819-4df3-a908-1957d9a67353",
"name": "便签2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-432,
112
],
"parameters": {
"width": 256,
"height": 224,
"content": "## Add Tavily API KEYS"
},
"typeVersion": 1
},
{
"id": "8b852ef8-8b16-426e-9fee-837403cbe09b",
"name": "便签4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-400,
576
],
"parameters": {
"width": 256,
"height": 224,
"content": "## Add Tavily API KEYS"
},
"typeVersion": 1
}
],
"pinData": {},
"connections": {
"If": {
"main": [
[
{
"node": "Tavily",
"type": "main",
"index": 0
}
],
[
{
"node": "Tavily1",
"type": "main",
"index": 0
}
]
]
},
"Tavily": {
"main": [
[
{
"node": "Websearch",
"type": "main",
"index": 0
}
]
]
},
"Tavily1": {
"main": [
[
{
"node": "WebSearch Summary",
"type": "main",
"index": 0
}
]
]
},
"AI Agent": {
"main": [
[
{
"node": "Create Doc",
"type": "main",
"index": 0
}
]
]
},
"AI Agent1": {
"main": [
[
{
"node": "Create Doc1",
"type": "main",
"index": 0
}
]
]
},
"Websearch": {
"main": [
[
{
"node": "Reference Script",
"type": "main",
"index": 0
}
]
]
},
"Create Doc": {
"main": [
[
{
"node": "Update Doc",
"type": "main",
"index": 0
}
]
]
},
"Create Doc1": {
"main": [
[
{
"node": "Update Doc1",
"type": "main",
"index": 0
}
]
]
},
"Create Outline": {
"main": [
[
{
"node": "Reference Script1",
"type": "main",
"index": 0
}
]
]
},
"Reference Script": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Reference Script1": {
"main": [
[
{
"node": "AI Agent1",
"type": "main",
"index": 0
}
]
]
},
"WebSearch Summary": {
"main": [
[
{
"node": "Create Outline",
"type": "main",
"index": 0
}
]
]
},
"On form submission": {
"main": [
[
{
"node": "If",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Google Gemini Chat Model1": {
"ai_languageModel": [
[
{
"node": "AI Agent1",
"type": "ai_languageModel",
"index": 0
},
{
"node": "Create Outline",
"type": "ai_languageModel",
"index": 0
}
]
]
}
}
}常见问题
如何使用这个工作流？
复制上方的 JSON 配置代码，在您的 n8n 实例中创建新工作流并选择「从 JSON 导入」，粘贴配置后根据需要修改凭证设置即可。
这个工作流适合什么场景？
高级 - 内容创作, 多模态 AI
需要付费吗？
本工作流完全免费，您可以直接导入使用。但请注意，工作流中使用的第三方服务（如 OpenAI API）可能需要您自行付费。
相关工作流推荐
WordPress博客自动化专业版（深度研究）v2.1市场
使用GPT-4o、Perplexity AI和多语言支持自动化SEO优化的博客创建
If
Set
Xml
+27
125 节点Daniel Ng
内容创作
使用Gemini AI和Elementor为多个客户生成并安排SEO博客文章
使用Gemini AI和Elementor为多个客户生成并安排SEO博客文章
If
N8n
Set
+12
39 节点Zain Khan
内容创作
使用Gemini AI、DeepSeek和GoAPI为电商生成产品图片和视频
使用Gemini AI、DeepSeek和GoAPI为电商生成产品图片和视频
If
Set
Html
+13
50 节点Dhruv Dalsaniya
内容创作
使用 Telegram 和 Google Gemini 的个人助理机器人
基于 Telegram 和 Google Gemini 的多模态个人AI助手，集成生产力工具
Set
Code
Merge
+16
83 节点Anshul Chauhan
内容创作
使用Groq、Gemini和Slack审批系统自动化RSS到Medium发布
通过Groq、Gemini和Slack审批系统实现RSS到Medium发布的自动化流程
If
Set
Code
+16
41 节点ObisDev
内容创作
LinkedIn和X病毒内容自动引擎
使用AI生成和发布自动创建LinkedIn和X的病毒内容
If
Set
Wait
+26
156 节点Diptamoy Barman
内容创作
