This workflow follows the Agent → Chat Trigger 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": "multi_agent-Cloudmcpserver_sse_rag",
"nodes": [
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"typeVersion": 1.1,
"position": [
-1080,
220
],
"id": "859ad202-2ab0-4407-890c-eb6560886f5c",
"name": "When chat message received"
},
{
"parameters": {},
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"typeVersion": 1.3,
"position": [
-340,
440
],
"id": "a30265c1-9b65-418e-89d0-c10584864f33",
"name": "Simple Memory"
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "2966067a-7bb6-4316-8ad9-2b7681c7e0b9",
"leftValue": "={{$json[\"output\"]}}",
"rightValue": "",
"operator": {
"type": "string",
"operation": "equals",
"name": "filter.operator.equals"
}
}
],
"combinator": "and"
},
"options": {}
},
"type": "n8n-nodes-base.if",
"typeVersion": 2.2,
"position": [
0,
0
],
"id": "d708ae26-2d8d-4a00-b66c-3a1679fe1d53",
"name": "If"
},
{
"parameters": {
"promptType": "define",
"text": "=\u0e04\u0e38\u0e13\u0e40\u0e1b\u0e47\u0e19 data engineer \u0e17\u0e35\u0e48\u0e08\u0e30\u0e23\u0e31\u0e1a\u0e1c\u0e25\u0e25\u0e31\u0e1e\u0e18\u0e4c\u0e08\u0e32\u0e01 AI agent \u0e01\u0e48\u0e2d\u0e19\u0e2b\u0e19\u0e49\u0e32\u0e21\u0e32\u0e1b\u0e23\u0e30\u0e21\u0e27\u0e25\u0e1c\u0e25\nInput \u0e17\u0e35\u0e48\u0e44\u0e14\u0e49\u0e23\u0e31\u0e1a: {{ $('RAG Agent').first().json.output }}\n\n\u0e07\u0e32\u0e19\u0e02\u0e2d\u0e07\u0e04\u0e38\u0e13: \u0e41\u0e1b\u0e25\u0e07\u0e02\u0e49\u0e2d\u0e21\u0e39\u0e25\u0e19\u0e35\u0e49\u0e40\u0e1b\u0e47\u0e19 JSON object \u0e17\u0e35\u0e48\u0e40\u0e2b\u0e21\u0e32\u0e30\u0e2a\u0e21\u0e2a\u0e33\u0e2b\u0e23\u0e31\u0e1a MongoDB\n\n\u0e01\u0e23\u0e38\u0e13\u0e32 return \u0e40\u0e09\u0e1e\u0e32\u0e30 JSON object \u0e42\u0e14\u0e22\u0e15\u0e23\u0e07 \u0e44\u0e21\u0e48\u0e15\u0e49\u0e2d\u0e07\u0e21\u0e35 markdown code blocks \u0e2b\u0e23\u0e37\u0e2d\u0e04\u0e33\u0e2d\u0e18\u0e34\u0e1a\u0e32\u0e22\n\nJSON structure \u0e17\u0e35\u0e48\u0e15\u0e49\u0e2d\u0e07\u0e01\u0e32\u0e23:\n1. \u0e21\u0e35\u0e42\u0e04\u0e23\u0e07\u0e2a\u0e23\u0e49\u0e32\u0e07\u0e17\u0e35\u0e48\u0e0a\u0e31\u0e14\u0e40\u0e08\u0e19\n2. \u0e40\u0e2b\u0e21\u0e32\u0e30\u0e2a\u0e33\u0e01\u0e32\u0e23\u0e40\u0e01\u0e47\u0e1a\u0e43\u0e19 MongoDB\n3. \u0e23\u0e2d\u0e07\u0e23\u0e31\u0e1a\u0e01\u0e32\u0e23 query \u0e44\u0e14\u0e49\u0e2d\u0e22\u0e48\u0e32\u0e07\u0e21\u0e35\u0e1b\u0e23\u0e30\u0e2a\u0e34\u0e17\u0e18\u0e34\u0e20\u0e32\u0e1e\n4. \u0e43\u0e0a\u0e49\u0e20\u0e32\u0e29\u0e32\u0e44\u0e17\u0e22\u0e40\u0e1b\u0e47\u0e19\u0e2b\u0e25\u0e31\u0e01",
"options": {
"systemMessage": "=\u0e04\u0e38\u0e13\u0e40\u0e1b\u0e47\u0e19 data engineer \u0e17\u0e33\u0e2b\u0e19\u0e49\u0e32\u0e17\u0e35\u0e48\u0e2a\u0e23\u0e49\u0e32\u0e07 valid JSON object \u0e40\u0e1e\u0e37\u0e48\u0e2d\u0e43\u0e0a\u0e49\u0e19\u0e33\u0e40\u0e02\u0e49\u0e32 MongoDB\n\n\u0e01\u0e0e\u0e2a\u0e33\u0e04\u0e31\u0e0d:\n- return \u0e40\u0e09\u0e1e\u0e32\u0e30 valid JSON object \u0e40\u0e17\u0e48\u0e32\u0e19\u0e31\u0e49\u0e19\n- \u0e2b\u0e49\u0e32\u0e21\u0e43\u0e2a\u0e48\u0e04\u0e33\u0e2d\u0e18\u0e34\u0e1a\u0e32\u0e22 markdown \u0e2b\u0e23\u0e37\u0e2d\u0e02\u0e49\u0e2d\u0e04\u0e27\u0e32\u0e21\u0e43\u0e14\u0e46\n- \u0e15\u0e23\u0e27\u0e08\u0e2a\u0e2d\u0e1a JSON syntax \u0e43\u0e2b\u0e49\u0e16\u0e39\u0e01\u0e15\u0e49\u0e2d\u0e07 (comma, quotes, brackets)\n- \u0e2b\u0e49\u0e32\u0e21 wrap \u0e14\u0e49\u0e27\u0e22 ```json``` \u0e2b\u0e23\u0e37\u0e2d array\n\n\u0e15\u0e31\u0e27\u0e2d\u0e22\u0e48\u0e32\u0e07 output \u0e17\u0e35\u0e48\u0e16\u0e39\u0e01\u0e15\u0e49\u0e2d\u0e07:\n{\"_id\": \"analysis_001\", \"data\": {...}}"
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.9,
"position": [
280,
220
],
"id": "d1a90147-ad85-4d86-8070-1f7fc34e82cd",
"name": "Data Engineer Agent"
},
{
"parameters": {
"jsCode": "const staticData = $getWorkflowStaticData('global');\n\n// \u0e15\u0e23\u0e27\u0e08\u0e2a\u0e2d\u0e1a\u0e27\u0e48\u0e32\u0e40\u0e1b\u0e47\u0e19\u0e23\u0e2d\u0e1a\u0e41\u0e23\u0e01\u0e2b\u0e23\u0e37\u0e2d\u0e44\u0e21\u0e48\nif (!staticData.current_session_data) {\n // \u0e23\u0e2d\u0e1a\u0e41\u0e23\u0e01: \u0e40\u0e01\u0e47\u0e1a\u0e02\u0e49\u0e2d\u0e21\u0e39\u0e25\u0e17\u0e31\u0e49\u0e07\u0e2b\u0e21\u0e14\u0e08\u0e32\u0e01 input\n staticData.current_session_data = {\n original_question: $json[\"chatInput\"],\n sessionId: $json[\"userId\"] || \"default-session\",\n startTime: new Date().toISOString()\n };\n}\n\n// \u0e2a\u0e48\u0e07\u0e2d\u0e2d\u0e01\u0e02\u0e49\u0e2d\u0e21\u0e39\u0e25\u0e17\u0e35\u0e48\u0e04\u0e07\u0e2d\u0e22\u0e39\u0e48\u0e17\u0e38\u0e01\u0e23\u0e2d\u0e1a\nreturn [{\n sessionId: staticData.current_session_data.sessionId,\n original_question: staticData.current_session_data.original_question\n}];\n"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
-780,
220
],
"id": "ff270044-17c4-4f51-88b8-526407f0ef0b",
"name": "Question memorizer"
},
{
"parameters": {
"model": {
"__rl": true,
"value": "openai/gpt-oss-120b",
"mode": "list",
"cachedResultName": "openai/gpt-oss-120b"
},
"options": {
"maxTokens": 8000
}
},
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"typeVersion": 1.2,
"position": [
-520,
440
],
"id": "72452aa6-2d56-4cc6-b2d0-0528252102d6",
"name": "OpenAI Chat Model by OpenRouter",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"model": {
"__rl": true,
"value": "qwen/qwen-max",
"mode": "list",
"cachedResultName": "qwen/qwen-max"
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"typeVersion": 1.2,
"position": [
380,
440
],
"id": "dbd9730f-2126-47f0-ac50-c08585cd6363",
"name": "OpenAI Chat Model by OpenRouter-1",
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"descriptionType": "manual",
"toolDescription": "List all documentation sources currently stored",
"connectionType": "sse",
"operation": "executeTool",
"toolName": "list_sources"
},
"type": "n8n-nodes-mcp.mcpClientTool",
"typeVersion": 1,
"position": [
-160,
440
],
"id": "2e67a747-2624-428a-9790-8f1852a9bc4f",
"name": "list_sources",
"credentials": {
"mcpClientSseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"descriptionType": "manual",
"toolDescription": "Add all supported files from a directory to the RAG database",
"connectionType": "sse",
"operation": "executeTool",
"toolName": "add_directory",
"toolParameters": "={{ (function() {\n const paramsString = $fromAI('Tool_Parameters');\n \n // Default path\n const defaultParams = {\n path: \"/home/user/documents\"\n };\n \n if (!paramsString) {\n return defaultParams;\n }\n \n try {\n // \u0e25\u0e2d\u0e07\u0e41\u0e1b\u0e25\u0e07 JSON \u0e01\u0e48\u0e2d\u0e19\n const params = JSON.parse(paramsString);\n return {\n path: params.path || defaultParams.path\n };\n } catch (e) {\n // \u0e16\u0e49\u0e32\u0e41\u0e1b\u0e25\u0e07 JSON \u0e44\u0e21\u0e48\u0e44\u0e14\u0e49 \u0e16\u0e37\u0e2d\u0e27\u0e48\u0e32\u0e40\u0e1b\u0e47\u0e19 path string \u0e42\u0e14\u0e22\u0e15\u0e23\u0e07\n if (typeof paramsString === 'string' && paramsString.trim().length > 0) {\n return {\n path: paramsString.trim()\n };\n }\n return defaultParams;\n }\n})() }}"
},
"type": "n8n-nodes-mcp.mcpClientTool",
"typeVersion": 1,
"position": [
-20,
440
],
"id": "85c4ee62-99bb-4f19-adb6-73f69f20abd7",
"name": "add_directory",
"credentials": {
"mcpClientSseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"descriptionType": "manual",
"toolDescription": "Search through stored documentation",
"connectionType": "sse",
"operation": "executeTool",
"toolName": "search_documentation",
"toolParameters": "={{ (function() {\n const paramsString = $fromAI('Tool_Parameters');\n \n // Default parameters\n const defaultParams = {\n query: \"search term\",\n limit: 10\n };\n \n if (!paramsString) {\n return defaultParams;\n }\n \n try {\n // \u0e25\u0e2d\u0e07\u0e41\u0e1b\u0e25\u0e07 JSON \u0e01\u0e48\u0e2d\u0e19\n const params = JSON.parse(paramsString);\n return {\n query: params.query || params.search || params.keyword || defaultParams.query,\n limit: params.limit || params.max || params.count || defaultParams.limit\n };\n } catch (e) {\n // \u0e16\u0e49\u0e32\u0e41\u0e1b\u0e25\u0e07 JSON \u0e44\u0e21\u0e48\u0e44\u0e14\u0e49 \u0e16\u0e37\u0e2d\u0e27\u0e48\u0e32\u0e40\u0e1b\u0e47\u0e19 search string \u0e42\u0e14\u0e22\u0e15\u0e23\u0e07\n if (typeof paramsString === 'string' && paramsString.trim().length > 0) {\n return {\n query: paramsString.trim(),\n limit: defaultParams.limit\n };\n }\n return defaultParams;\n }\n})() }}"
},
"type": "n8n-nodes-mcp.mcpClientTool",
"typeVersion": 1,
"position": [
120,
440
],
"id": "bf4fd4fb-4037-43aa-be5a-1c118161dce9",
"name": "search_documentation",
"credentials": {
"mcpClientSseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"promptType": "define",
"text": "={{$json[\"original_question\"]}}",
"options": {
"systemMessage": "=CORE IDENTITY\nYou are a professional Documentation Management Agent with RAG (Retrieval-Augmented Generation) capabilities. Your primary goals are EFFICIENT DOCUMENTATION MANAGEMENT, ACCURATE RETRIEVAL, and COMPREHENSIVE KNOWLEDGE BASE BUILDING.\n\nTOOL USAGE REQUIREMENT\nYou MUST use documentation tools for every documentation-related request. You cannot provide documentation analysis without using the actual tools.\n\nAVAILABLE TOOLS\n1. add_documentation - Add documentation from a URL to the RAG database\n2. search_documentation - Search through stored documentation\n3. list_sources - List all documentation sources currently stored\n4. add_directory - Add all supported files from a directory to the RAG database\n\nMANDATORY OUTPUT FORMAT\nEvery operation MUST include:\nREQUEST: [restate user request]\nTOOL SELECTED: [which tool and why]\nPARAMETERS USED:\n[exact parameters in JSON format]\nEXECUTION STATUS: [SUCCESS/FAILED/ERROR]\nRAW RESULTS: [actual response from tool]\nACTION SUMMARY: [what was accomplished or what went wrong]\nNEXT STEPS: [recommendations for follow-up actions if needed]\n\nWORKFLOW PATTERNS\n\nPattern 1: Adding New Documentation\n- Use add_documentation for URLs\n- Use add_directory for local folders\n- Always verify with list_sources after adding\n\nPattern 2: Searching Documentation\n- First use list_sources to understand available sources\n- Then use search_documentation with relevant queries\n- Refine search if initial results insufficient\n\nPattern 3: Knowledge Base Maintenance\n- Regularly list_sources to audit content\n- Search before adding to avoid duplicates\n- Build comprehensive coverage systematically\n\nERROR HANDLING\n- If add fails: Check URL/path validity\n- If search returns empty: Try broader terms\n- If tool errors: Report exact error message\n- Always provide alternative approaches",
"maxIterations": 30
}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 1.9,
"position": [
-440,
220
],
"id": "5ef6f293-d28e-4524-a741-236aa5bef488",
"name": "RAG Agent",
"alwaysOutputData": false,
"retryOnFail": true
}
],
"connections": {
"When chat message received": {
"main": [
[
{
"node": "Question memorizer",
"type": "main",
"index": 0
}
]
]
},
"Simple Memory": {
"ai_memory": [
[
{
"node": "RAG Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"If": {
"main": [
[
{
"node": "Question memorizer",
"type": "main",
"index": 0
}
],
[
{
"node": "Data Engineer Agent",
"type": "main",
"index": 0
}
]
]
},
"Question memorizer": {
"main": [
[
{
"node": "RAG Agent",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model by OpenRouter": {
"ai_languageModel": [
[
{
"node": "RAG Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"OpenAI Chat Model by OpenRouter-1": {
"ai_languageModel": [
[
{
"node": "Data Engineer Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"list_sources": {
"ai_tool": [
[
{
"node": "RAG Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"add_directory": {
"ai_tool": [
[
{
"node": "RAG Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"search_documentation": {
"ai_tool": [
[
{
"node": "RAG Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"RAG Agent": {
"main": [
[
{
"node": "If",
"type": "main",
"index": 0
}
]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "c9673bc8-53bf-43cd-905b-66351c41ee0c",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "hyzMkUCp2qK0rfoV",
"tags": [
{
"createdAt": "2025-09-04T04:31:20.574Z",
"updatedAt": "2025-09-04T04:31:20.574Z",
"id": "C7wWNpLYb8cT5j44",
"name": "in-progress"
}
]
}
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.
mcpClientSseApiopenAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
multi_agent-Cloudmcpserver_sse_rag. Uses chatTrigger, memoryBufferWindow, agent, lmChatOpenAi. Chat trigger; 11 nodes.
Source: https://github.com/kuan31898/2025-n8n-mcp/blob/c22d48eae33490ce46efabf5064bcee0f04920b7/backup@6sept2025/n8n-mcpserver-sse-rag/multi_agent_Cloudmcpserver_sse_rag.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 n8n workflow template uses community nodes and is only compatible with the self-hosted version of n8n.
Chat Trigger, Agent, N8N Nodes Mcp +4
This template obtains all the possible tools from Bright Data MCP, process this through chatbot, then run any tool based on the user's query
Agent, Chat Trigger, OpenAI Chat +6
💥🛠️Build a Web Search Chatbot with GPT-4o and MCP Brave Search. Uses agent, chatTrigger, n8n-nodes-mcp, memoryBufferWindow. Chat trigger; 15 nodes.
Agent, Chat Trigger, N8N Nodes Mcp +2
Disclaimer: This workflow only works with local installations of n8n because it uses a community MCP node
Agent, Chat Trigger, N8N Nodes Mcp +2
Disclaimer: This template is for self-hosted n8n instances only.
Chat Trigger, Agent, OpenAI Chat +3