Complete Youtube
Generate comprehensive YouTube video summaries and insights with this AI-powered workflow, transforming how you interact with video content. This automation connects YouTube to OpenAI, allowing you to search for videos, extract their data, and then leverage advanced AI to process and summarize that information. The workflow begins with a manual trigger or a chat message received, initiating an AI Agent that orchestrates the entire process. It uses the YouTube node to get video details, then processes this data through a series of nodes including httpRequest for finding video data, splitInBatches for looping over items, and memoryBufferWindow for storing and retrieving information, ensuring efficient handling of video content. Finally, the OpenAI LLM node generates intelligent responses based on the video data. This workflow is ideal for content creators, researchers, and marketers who need to quickly understand video content without watching entire videos, saving significant time and effort in content analysis and research by automating the tedious task of video data extraction and summarization.
Workflow JSON
{"id": "XSyVFC1tsGSxNwX9", "meta": {"instanceId": "60ad864624415060d2d0a0e71acd8b3b40e4ee2e9ef4b439d9937d3d33537a96"}, "name": "Complete Youtube", "tags": [], "nodes": [{"id": "fd74706b-609b-4723-b4a4-067e1b064194", "name": "AI Agent", "type": "@n8n/n8n-nodes-langchain.agent", "position": [300, 60], "parameters": {"options": {"systemMessage": "=You help youtube creators find trending videos based on a specific niche.\n\nVerify if the user informed a niche before doing anything. If not, then ask him for one by giving him suggestions for him to select from.\n\nAfter you know what type of content the user might produce, use the \"youtube_search\" tool up to 3 times with different search terms based on the user's content type and niche.\n\nThe tool will answer with a list of videos from the last 2 days that had the most amount of relevancy. It returns a list of json's covering each video's id, view count, like count, comment count, description, channel title, tags and channel id. Each video is separated by \"### NEXT VIDEO FOUND: ###\".\n\nYou should then proceed to understand the data received then provide the user with insightful data of what could be trending from the past 2 days. Provide the user links to the trending videos which should be in this structure:\n\nhttps://www.youtube.com/watch?v={video_id}\n\nto reach the channel's link you should use:\n\nhttps://www.youtube.com/channel/{channel_id}\n\nFind patterns in the tags, titles and especially in the related content for the videos found.\n\nYour mission isn't to find the trending videos. It's to provide the user with valuable information of what is trending in that niche in terms of content news. Remember to provide the user with the numbers of views, likes and comments while commenting about any video. So you should not talk about any particular video, focus rather in explaining the overall senario of all that was found.\n\nExample of response:\n\n\"It seems like what is trending in digital marketing right now is talking about mental triggers, since 3 of the most trending videos in the last 2 days were about...\""}}, "typeVersion": 1.6}, {"id": "ced4b937-b590-4727-b1f2-a5e88b96091a", "name": "chat_message_received", "type": "@n8n/n8n-nodes-langchain.chatTrigger", "position": [80, 60], "webhookId": "ff9622a4-a6ec-4396-b9de-c95bd834c23c", "parameters": {"options": {}}, "typeVersion": 1.1}, {"id": "35a91359-5007-407d-a750-d6642e595690", "name": "youtube_search", "type": "@n8n/n8n-nodes-langchain.toolWorkflow", "position": [540, 180], "parameters": {"name": "youtube_search", "workflowId": {"__rl": true, "mode": "list", "value": "N9DveO781xbNf8qs", "cachedResultName": "Youtube Search Workflow"}, "description": "Call this tool to search for trending videos based on a query.", "jsonSchemaExample": "{\n\t\"search_term\": \"some_value\"\n}", "specifyInputSchema": true}, "typeVersion": 1.2}, {"id": "42f41096-531d-4587-833a-6f659ef78dd0", "name": "openai_llm", "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi", "position": [260, 180], "parameters": {"options": {}}, "typeVersion": 1}, {"id": "e4bda5b9-abd4-4cd6-8c95-126a01aa6e21", "name": "window_buffer_memory", "type": "@n8n/n8n-nodes-langchain.memoryBufferWindow", "position": [400, 180], "parameters": {}, "typeVersion": 1.2}, {"id": "f6d86c5a-393a-4bcf-bdaf-3b06c79fa51d", "name": "Sticky Note1", "type": "n8n-nodes-base.stickyNote", "position": [0, 0], "parameters": {"color": 7, "width": 693.2572054941234, "height": 354.53098948245656, "content": "Main Workflow"}, "typeVersion": 1}, {"id": "4ddbc3f0-e3d7-4ce4-a732-d731c05024d2", "name": "find_video_data1", "type": "n8n-nodes-base.httpRequest", "position": [700, 720], "parameters": {"url": "https://www.googleapis.com/youtube/v3/videos?", "options": {}, "sendQuery": true, "queryParameters": {"parameters": [{"name": "key", "value": "={{ $env[\"GOOGLE_API_KEY\"] }}"}, {"name": "id", "value": "={{ $json.id.videoId }}"}, {"name": "part", "value": "contentDetails, snippet, statistics"}]}}, "typeVersion": 4.2}, {"id": "fdb28635-801d-4ce0-8919-11446c6a7a82", "name": "get_videos1", "type": "n8n-nodes-base.youTube", "position": [280, 560], "parameters": {"limit": 3, "filters": {"q": "={{ $json.query.search_term }}", "regionCode": "US", "publishedAfter": "={{ new Date(Date.now() - 2 * 24 * 60 * 60 * 1000).toISOString() }}"}, "options": {"order": "relevance", "safeSearch": "moderate"}, "resource": "video"}, "credentials": {"youTubeOAuth2Api": {"id": "", "name": "[Your youTubeOAuth2Api]"}}, "typeVersion": 1}, {"id": "60e9e61d-0e5e-4212-8b55-71299aeec4d5", "name": "response1", "type": "n8n-nodes-base.set", "position": [1100, 500], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "b9b9117b-ea14-482e-a13b-e68b8e6b441d", "name": "response", "type": "string", "value": "={{ $input.all() }}"}]}}, "typeVersion": 3.4}, {"id": "254a6740-8b25-4898-9795-4c3f0009471f", "name": "group_data1", "type": "n8n-nodes-base.set", "position": [1160, 700], "parameters": {"options": {}, "assignments": {"assignments": [{"id": "47c172ad-90c8-4cf6-a9f5-50607e04cc90", "name": "id", "type": "string", "value": "={{ $json.items[0].id }}"}, {"id": "9e639efa-0714-4b06-9847-f7b4b2fbef59", "name": "viewCount", "type": "string", "value": "={{ $json.items[0].statistics.viewCount }}"}, {"id": "93328f00-91b8-425b-ad0f-a330b2f95242", "name": "likeCount", "type": "string", "value": "={{ $json.items[0].statistics.likeCount }}"}, {"id": "015b0fb2-2a98-464c-a21b-51100616f26a", "name": "commentCount", "type": "string", "value": "={{ $json.items[0].statistics.commentCount }}"}, {"id": "cf1e1ec3-a138-42b8-8747-d249afa58dd3", "name": "description", "type": "string", "value": "={{ $json.items[0].snippet.description }}"}, {"id": "c5c9a3a2-b820-4932-a38a-e21102992215", "name": "title", "type": "string", "value": "={{ $json.items[0].snippet.title }}"}, {"id": "38216ead-1f8d-4f93-b6ad-5ef709a1ad2a", "name": "channelTitle", "type": "string", "value": "={{ $json.items[0].snippet.channelTitle }}"}, {"id": "ff34194d-3d46-43a8-9127-84708987f536", "name": "tags", "type": "string", "value": "={{ $json.items[0].snippet.tags.join(', ') }}"}, {"id": "e50b0f7b-3e37-4557-8863-d68d4fa505c8", "name": "channelId", "type": "string", "value": "={{ $json.items[0].snippet.channelId }}"}]}}, "typeVersion": 3.4}, {"id": "124c19a9-cbbd-4010-be37-50523c05f64b", "name": "save_data_to_memory1", "type": "n8n-nodes-base.code", "position": [1360, 700], "parameters": {"mode": "runOnceForEachItem", "jsCode": "const workflowStaticData = $getWorkflowStaticData('global');\n\nif (typeof workflowStaticData.lastExecution !== 'object') {\n workflowStaticData.lastExecution = {\n response: \"\"\n };\n}\n\nfunction removeEmojis(text) {\n return text.replace(/[\\u{1F600}-\\u{1F64F}|\\u{1F300}-\\u{1F5FF}|\\u{1F680}-\\u{1F6FF}|\\u{2600}-\\u{26FF}|\\u{2700}-\\u{27BF}]/gu, '');\n}\n\nfunction cleanDescription(description) {\n return description\n .replace(/https?:\\/\\/\\S+/g, '')\n .replace(/www\\.\\S+/g, '')\n .replace(/ +/g, ' ')\n .trim();\n}\n\nconst currentItem = { ...$input.item };\n\nif (currentItem.description) {\n currentItem.description = cleanDescription(currentItem.description);\n}\n\nlet sanitizedItem = JSON.stringify(currentItem)\n .replace(/\\\\r/g, ' ')\n .replace(/https?:\\/\\/\\S+/g, '')\n .replace(/www\\.\\S+/g, '')\n .replace(/\\\\n/g, ' ')\n .replace(/\\n/g, ' ')\n .replace(/\\\\/g, '')\n .replace(/ +/g, ' ')\n .trim();\n\nif (workflowStaticData.lastExecution.response) {\n workflowStaticData.lastExecution.response += ' ### NEXT VIDEO FOUND: ### ';\n}\n\nworkflowStaticData.lastExecution.response += removeEmojis(sanitizedItem);\n\nreturn workflowStaticData.lastExecution;\n"}, "typeVersion": 2}, {"id": "67f92ec4-71c0-49df-a0ea-11d2e3cf0f94", "name": "retrieve_data_from_memory1", "type": "n8n-nodes-base.code", "position": [780, 500], "parameters": {"jsCode": "const workflowStaticData = $getWorkflowStaticData('global');\n\nconst lastExecution = workflowStaticData.lastExecution;\n\nreturn lastExecution;"}, "typeVersion": 2}, {"id": "685820ba-b089-4cdc-984d-52f134754b5c", "name": "loop_over_items1", "type": "n8n-nodes-base.splitInBatches", "position": [500, 560], "parameters": {"options": {}}, "typeVersion": 3}, {"id": "3d4d5a4b-d06b-41db-bb78-a64a266d5308", "name": "if_longer_than_3_", "type": "n8n-nodes-base.if", "position": [880, 720], "parameters": {"options": {}, "conditions": {"options": {"version": 2, "leftValue": "", "caseSensitive": true, "typeValidation": "strict"}, "combinator": "and", "conditions": [{"id": "08ba3db9-6bcf-47f8-a74d-9e26f28cb08f", "operator": {"type": "boolean", "operation": "true", "singleValue": true}, "leftValue": "={{ \n (() => {\n const duration = $json.items[0].contentDetails.duration;\n\n // Helper function to convert ISO 8601 duration to seconds\n const iso8601ToSeconds = iso8601 => {\n const match = iso8601.match(/PT(?:(\\d+)H)?(?:(\\d+)M)?(?:(\\d+)S)?/);\n const hours = parseInt(match[1] || 0, 10);\n const minutes = parseInt(match[2] || 0, 10);\n const seconds = parseInt(match[3] || 0, 10);\n return hours * 3600 + minutes * 60 + seconds;\n };\n\n // Convert duration to seconds\n const durationInSeconds = iso8601ToSeconds(duration);\n\n // Check if greater than 210 seconds (3 minutes 30 seconds)\n return durationInSeconds > 210;\n })() \n}}", "rightValue": ""}]}}, "typeVersion": 2.2}, {"id": "7c6b8b82-fd6c-4f44-bccf-88c5a76f0319", "name": "Sticky Note2", "type": "n8n-nodes-base.stickyNote", "position": [0, 420], "parameters": {"color": 5, "width": 1607, "height": 520, "content": "This part should be abstracted to another workflow and called inside the \"youtube_search\" tool of the main AI Agent."}, "typeVersion": 1}], "active": false, "pinData": {}, "settings": {"executionOrder": "v1"}, "versionId": "cea84238-2b82-4a32-85dd-0c71ad685d47", "connections": {"openai_llm": {"ai_languageModel": [[{"node": "AI Agent", "type": "ai_languageModel", "index": 0}]]}, "get_videos1": {"main": [[{"node": "loop_over_items1", "type": "main", "index": 0}]]}, "group_data1": {"main": [[{"node": "save_data_to_memory1", "type": "main", "index": 0}]]}, "youtube_search": {"ai_tool": [[{"node": "AI Agent", "type": "ai_tool", "index": 0}]]}, "find_video_data1": {"main": [[{"node": "if_longer_than_3_", "type": "main", "index": 0}]]}, "loop_over_items1": {"main": [[{"node": "retrieve_data_from_memory1", "type": "main", "index": 0}], [{"node": "find_video_data1", "type": "main", "index": 0}]]}, "if_longer_than_3_": {"main": [[{"node": "group_data1", "type": "main", "index": 0}], [{"node": "loop_over_items1", "type": "main", "index": 0}]]}, "save_data_to_memory1": {"main": [[{"node": "loop_over_items1", "type": "main", "index": 0}]]}, "window_buffer_memory": {"ai_memory": [[{"node": "AI Agent", "type": "ai_memory", "index": 0}]]}, "chat_message_received": {"main": [[{"node": "AI Agent", "type": "main", "index": 0}]]}, "retrieve_data_from_memory1": {"main": [[{"node": "response1", "type": "main", "index": 0}]]}}}How to Import This Workflow
- 1Copy the workflow JSON above using the Copy Workflow JSON button.
- 2Open your n8n instance and go to Workflows.
- 3Click Import from JSON and paste the copied workflow.
Don't have an n8n instance? Start your free trial at n8nautomation.cloud
Related Templates
Text to Speech (OpenAI)
Converts text into natural-sounding speech using OpenAI's Text-to-Speech API. It sends your input text to OpenAI and receives an audio file in return. This is useful for creating audio versions of articles, generating voiceovers for videos, or providing accessibility features for web content. Quickly transform written content into engaging audio.
LangChain - Example - Code Node Example
Explore a basic LangChain agent that answers questions using a custom tool. This workflow connects n8n's AI nodes and custom code nodes to OpenAI for language model interactions. It's useful for developers building custom AI assistants or researchers experimenting with agentic workflows. This saves development time by providing a ready-to-use example of a LangChain agent.
AI-Powered Candidate Shortlisting Automation for ERPNext
Automate AI-powered candidate shortlisting for ERPNext job applications. This workflow connects ERPNext, Google Gemini, WhatsApp, and Outlook to process resumes, evaluate candidates, and communicate outcomes. Recruiters and HR departments can use this to efficiently screen applicants, automatically reject unqualified candidates, and send acceptance notifications. It significantly reduces manual review time and streamlines the hiring process.