This workflow follows the Agent → OpenAI Chat recipe pattern — see all workflows that pair these two integrations.
The workflow JSON
Copy or download the full n8n JSON below. Paste it into a new n8n workflow, add your credentials, activate. Full import guide →
{
"name": "LINE-Desktop-MCP-Demo-chatbot",
"nodes": [
{
"parameters": {
"promptType": "define",
"text": "Check the latest messages and discussions in the LINE group [Project Discussion], determine whether a reply is needed based on the rules below, send an appropriate response automatically via LINE-desktop-MCP, and then terminate the agent execution after completion.\n\n---\n\n\u6aa2\u67e5 line group [\u5c08\u6848\u8a0e\u8ad6] \u4e2d\u7684\u6700\u65b0\u5c0d\u8a71\u8207\u8a0e\u8ad6\u5167\u5bb9, \u4f9d\u898f\u5247\u5224\u65b7\u4e26\u7d66\u4e88\u9069\u7576\u7684\u5c0d\u8a71\u56de\u61c9\u4e26\u81ea\u52d5\u9001\u51fa. \u5b8c\u6210\u5f8c\u5373\u7d50\u675f agent \u57f7\u884c",
"options": {
"systemMessage": "You are C3 (West-3), a chatbot assistant working for a sales manager.\nYou can use the LINE-desktop-MCP tool to read conversations in the LINE group, detect whether any response is needed, and send suitable replies using the same tool.\n\nPlease follow these rules to decide whether to send a message\n\n(You don\u2019t need to output these reasoning steps through LINE MCP)\n1. Check the most recent speaker.\n\u3000If it\u2019s yourself (C3), do not reply, to avoid flooding the chat.\n2. Avoid replying repeatedly to the same question if it has already been answered.\n\n\u3000If the customer repeats a question, only respond if new information is needed.\n3. When encountering a question you\u2019re not sure about, politely acknowledge it and tag @Max to ask for his input.\n\nGroup members\n1. Geoffrey Wang \u2014 Client\n2. Max \u2014 Sales Manager\n3. C3 \u2014 Max\u2019s AI assistant bot\n\nTone and Style\n1. Keep replies polite and professional, concise, and free from unnecessary filler words.\n2. Respond in the same language used by the person who asked the question.\n\nBasic Q&A Knowledge Base (can be extended via RAG or other tools)\n1. The ongoing project Project.T is expected to enter UAT on November 5.\n2. The company has been established for 3 years.\n3. Available meeting times next week: Thursday afternoon and Friday morning.\n4. Training courses beyond the project\u2019s included hours are subject to separate quotation and billing.\n\n---\n\n\u4f60\u662f\u4e00\u4f4d\u696d\u52d9\u4e3b\u7ba1\u7684\u5c0f\u5e6b\u624b\u6a5f\u5668\u4eba, \u4ee3\u865fC3 (\u897f\u4e09), \u4f60\u6703\u4f7f\u7528 LINE-desktop-MCP \u5de5\u5177, \u8b80\u53d6 LINE\u7fa4\u7d44\u4e2d\u7684\u5c0d\u8a71, \u4ee5\u4fbf\u4e86\u89e3\u6709\u6c92\u6709\u9700\u8981\u56de\u61c9\u7684\u5730\u65b9. \u4e26\u7528 LINE-desktop-MCP \u5de5\u5177\u9001\u51fa\u9069\u7576\u56de\u61c9. \n\n## \u8acb\u9075\u5b88\u4ee5\u4e0b\u898f\u5247, \u6c7a\u5b9a\u8981\u4e0d\u8981\u9001\u51fa\u5c0d\u8a71, \u4f46\u4e0d\u9700\u8981\u5c07\u9019\u4e9b\u898f\u5247\u7684\u904e\u7a0b\u900f\u904e LINE MCP \u9001\u51fa\n\n1. \u9700\u5148\u6aa2\u67e5\u8fd1\u671f\u6700\u5f8c\u4e00\u500b\u767c\u8a00\u8005, \u5982\u679c\u662f\u81ea\u5df1C3, \u5247\u4e0d\u8981\u767c\u8a00, \u907f\u514d\u9020\u6210\u6d17\u983b\n2. \u5df2\u7d93\u56de\u8986\u904e\u7684\u554f\u984c, \u907f\u514d\u4e3b\u52d5\u91cd\u8907\u56de\u8986. \u5982\u679c\u5ba2\u6236\u91cd\u8907\n3. \u9047\u5230\u4e0d\u78ba\u5b9a\u7b54\u6848\u7684\u554f\u984c, \u9664\u4e86\u5148\u7d66\u4e88\u79ae\u8c8c\u6027\u5730\u56de\u8986\u5916, \u4e5f\u53ef\u4ee5 tag @max \u63d0\u9192\u4ed6\u51fa\u9762\u56de\u7b54\n\n## \u7fa4\u7d44\u4e2d\u7684\u4eba\n1. Geoffrey Wang: \u5ba2\u6236\n2. Max: \u696d\u52d9\u4e3b\u7ba1\n3. C3: Max\u7684AI\u6a5f\u5668\u4eba\u5c0f\u5e6b\u624b\n\n## \u89d2\u8272\u7528\u8a9e\u98a8\u683c\n1. \u5e36\u6709\u5c08\u696d\u8207\u79ae\u8c8c\u7684\u56de\u8986, \u4e26\u4fdd\u6301\u7c21\u77ed\u907f\u514d\u4e0d\u9700\u8981\u7684\u8d05\u5b57\u592a\u591a\n2. \u4f9d\u7167\u767c\u554f\u8005\u7684\u6240\u4f7f\u7528\u8a9e\u8a00\u505a\u56de\u8986\n\n## \u7c21\u6613 QA \u554f\u984c\u96c6 (\u53ef\u518d\u900f\u904e RAG \u6216\u5176\u4ed6\u5de5\u5177\u64f4\u5145)\n1. \u9032\u884c\u4e2d\u7684\u5c08\u6848 Project.T , \u9810\u8a08 11/5 \u53ef\u4ee5\u9032\u5165 UAT\n2. \u516c\u53f8\u6210\u7acb 3 \u5e74\n3. \u4e0b\u5468\u53ef\u4ee5\u958b\u6703\u7684\u6642\u9593, \u9031\u56db\u4e0b\u5348\u8207\u5468\u4e94\u65e9\u4e0a\n4. \u6559\u80b2\u8a13\u7df4\u8ab2\u7a0b, \u9664\u5c08\u6848\u4e2d\u6240\u9644\u5e36\u6642\u6578\u5916, \u5176\u9918\u9700\u53e6\u5916\u5831\u50f9\u8a08\u8cbb"
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 2.2,
"position": [
208,
0
],
"id": "5b62ffe4-20b6-4aaa-ac2c-965a2c414e24",
"name": "AI Agent",
"alwaysOutputData": true,
"executeOnce": true,
"notesInFlow": true
},
{
"parameters": {
"model": {
"__rl": true,
"value": "gpt-5-mini",
"mode": "list",
"cachedResultName": "gpt-5-mini"
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"typeVersion": 1.2,
"position": [
80,
240
],
"id": "ebbbd39b-5032-43d2-9268-03b5d5116a65",
"name": "OpenAI Chat Model",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"sessionIdType": "customKey",
"sessionKey": "7788"
},
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"typeVersion": 1.3,
"position": [
256,
240
],
"id": "4f37b772-86e9-4143-acd3-0ba75fcdaebf",
"name": "Simple Memory"
},
{
"parameters": {
"endpointUrl": "http://host.docker.internal:3000/mcp",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.mcpClientTool",
"typeVersion": 1.2,
"position": [
480,
240
],
"id": "f61e06f6-7981-4bd0-a64c-e504dbc3bba4",
"name": "MCP Client",
"notes": "LINE-Deskto-MCP:\n\nhttps://github.com/dtwang/line-desktop-mcp"
},
{
"parameters": {
"rule": {
"interval": [
{
"field": "minutes",
"minutesInterval": 10
},
{}
]
}
},
"type": "n8n-nodes-base.scheduleTrigger",
"typeVersion": 1.2,
"position": [
16,
0
],
"id": "337f3aa7-dc83-440b-93ea-38d29691bbf7",
"name": "Schedule Trigger"
},
{
"parameters": {
"content": "\ud83d\udccc **LINE \u696d\u52d9\u5c0f\u52a9\u624b Chatbot** \n\u81ea\u52d5\u6aa2\u67e5 LINE \u7fa4\u7d44\u8a0a\u606f\u4e26\u7531 AI \u751f\u6210\u56de\u8986\u3002 \n\u4f7f\u7528 [LINE-Desktop-MCP](https://github.com/dtwang/line-desktop-mcp) \u63a7\u5236 LINE \u684c\u9762\u7248\uff0c \u6bcf 10 \u5206\u9418\u89f8\u767c\u6aa2\u67e5\uff0c\u53ef\u81ea\u8a02\u6a21\u578b\u8207\u8a9e\u6c23\u3002\n\n\u9069\u5408\u5ba2\u670d\u3001\u793e\u7fa4\u8207\u696d\u52d9\u81ea\u52d5\u5316\u61c9\u7528\u3002 \n\n\ud83e\udd16 **LINE Business Assistant Chatbot** \nMonitors LINE chats and uses AI to reply automatically. Powered by [LINE-Desktop-MCP](https://github.com/dtwang/line-desktop-mcp). Runs every 10 min, customizable model and tone. \n\nPerfect for customer support and group automation.",
"height": 352,
"width": 304
},
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-352,
0
],
"id": "cc0ca8fa-76ce-4845-a484-1dbf93498c62",
"name": "Sticky Note"
}
],
"connections": {
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Simple Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"MCP Client": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Schedule Trigger": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "33e9a213-f8ef-4f17-9ff1-3ee00ecfed64",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "XH1zyo9FmqvJO18z",
"tags": []
}
Credentials you'll need
Each integration node will prompt for credentials when you import. We strip credential IDs before publishing — you'll add your own.
openAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
LINE-Desktop-MCP-Demo-chatbot. Uses agent, lmChatOpenAi, memoryBufferWindow, mcpClientTool. Scheduled trigger; 6 nodes.
Source: https://github.com/dtwang/line-desktop-mcp/blob/aa8d7323fc273aac5095efa6706c60e83b85d03b/doc_media/LINE-Desktop-MCP-Demo-chatbot-sample.json — original creator credit. Request a take-down →
Related workflows
Workflows that share integrations, category, or trigger type with this one. All free to copy and import.
This workflow automatically generates professional, branded LinkedIn posts using your custom Figma designs. Perfect for marketers, agencies, content creators, and businesses who want to maintain consi
Memory Buffer Window, Perplexity Tool, Tool Think +7
This template helps you manage your calendars by enriching each calendar event with data from Perplexity, OpenWeatherMap, & Open AI, and sending notifications on your Telegram.
Memory Buffer Window, OpenAI Chat, Agent +7
Build an MCP server with Airtable. Uses chatTrigger, agent, memoryBufferWindow, mcpClientTool. Chat trigger; 13 nodes.
Chat Trigger, Agent, Memory Buffer Window +4
Build an MCP server with Airtable. Uses chatTrigger, agent, memoryBufferWindow, mcpClientTool. Chat trigger; 13 nodes.
Chat Trigger, Agent, Memory Buffer Window +4
This template is designed for anyone who wants to integrate MCP with their AI Agents using Airtable. Whether you're a developer, a data analyst, or an automation enthusiast, if you're looking to lever
Chat Trigger, Agent, Memory Buffer Window +4