Automate With Marc This workflow corresponds to n8n.io template #7679 — we link there as the canonical source.
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 →
{
"id": "TgpCq3JAieEaFdGJ",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Email News MCP Template",
"tags": [],
"nodes": [
{
"id": "0606f766-255e-469c-8e6c-5751537ed3ab",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
192,
-160
],
"parameters": {
"options": {
"systemMessage": "You are a helpful email assistant.\n\n##Tool\nUse attached Email MCP Tool for emails when asked\n\nUse attached Email MCP Tool for "
}
},
"typeVersion": 2.2
},
{
"id": "225b0350-6eae-45fc-a158-da9961b8aafe",
"name": "When chat message received",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
0,
-160
],
"parameters": {
"options": {}
},
"typeVersion": 1.3
},
{
"id": "80fcfcad-1310-4cf2-a4df-bf6746339cfd",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
48,
48
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4.1-mini"
},
"options": {}
},
"typeVersion": 1.2
},
{
"id": "7e3db391-7ede-4e92-9593-7a1288938d80",
"name": "Simple Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
224,
48
],
"parameters": {},
"typeVersion": 1.3
},
{
"id": "1b9a577c-3401-4081-be39-d5051922df38",
"name": "Send a message in Gmail",
"type": "n8n-nodes-base.gmailTool",
"position": [
-144,
480
],
"parameters": {
"sendTo": "<<<REPLACE_WITH_EMAIL>>>",
"message": "<<<REPLACE_WITH_MESSAGE>>>",
"options": {},
"subject": "<<<REPLACE_WITH_SUBJECT>>>"
},
"typeVersion": 2.1
},
{
"id": "fa6ae7d8-3d4d-4bd0-a4f9-d1d295f5f14b",
"name": "Send a message in Gmail1",
"type": "n8n-nodes-base.gmailTool",
"position": [
64,
480
],
"parameters": {
"sendTo": "<<<REPLACE_WITH_EMAIL>>>",
"message": "<<<REPLACE_WITH_MESSAGE>>>",
"options": {},
"subject": "<<<REPLACE_WITH_SUBJECT>>>"
},
"typeVersion": 2.1
},
{
"id": "252988a9-b546-4e1a-9d6f-338618b5781b",
"name": "Send a message in Gmail2",
"type": "n8n-nodes-base.gmailTool",
"position": [
256,
480
],
"parameters": {
"sendTo": "<<<REPLACE_WITH_EMAIL>>>",
"message": "<<<REPLACE_WITH_MESSAGE>>>",
"options": {},
"subject": "<<<REPLACE_WITH_SUBJECT>>>"
},
"typeVersion": 2.1
},
{
"id": "722718e7-8a84-44b4-98e3-a6eb53902a7c",
"name": "Search in Tavily",
"type": "@tavily/n8n-nodes-tavily.tavilyTool",
"position": [
512,
480
],
"parameters": {
"query": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Query', ``, 'string') }}",
"options": {}
},
"typeVersion": 1
},
{
"id": "0bba3a97-e1ee-46f5-abec-7713d6ff2948",
"name": "Message a model in Perplexity",
"type": "n8n-nodes-base.perplexityTool",
"position": [
688,
480
],
"parameters": {
"options": {},
"messages": {
"message": [
{
"content": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('message0_Text', ``, 'string') }}"
}
]
},
"simplify": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Simplify_Output', ``, 'boolean') }}",
"requestOptions": {}
},
"typeVersion": 1
},
{
"id": "60b275c9-9e2f-4e3c-bc11-2477fe0bc951",
"name": "News MCP Server",
"type": "@n8n/n8n-nodes-langchain.mcpTrigger",
"position": [
544,
256
],
"parameters": {
"path": "<<<REPLACE_WITH_PATH>>>"
},
"typeVersion": 2
},
{
"id": "946b0a9d-590f-4633-ac98-ce983bbb205f",
"name": "Email MCP Server",
"type": "@n8n/n8n-nodes-langchain.mcpTrigger",
"position": [
-96,
256
],
"parameters": {
"path": "<<<REPLACE_WITH_PATH>>>"
},
"typeVersion": 2
},
{
"id": "34bff09d-95d1-446f-88cb-1c664d1ad754",
"name": "Email MCP Client",
"type": "@n8n/n8n-nodes-langchain.mcpClientTool",
"position": [
544,
48
],
"parameters": {
"endpointUrl": "<<<REPLACE_WITH_ENDPOINT_URL>>>",
"serverTransport": "httpStreamable"
},
"typeVersion": 1.1
},
{
"id": "57587695-df6b-461d-8596-6561ce295f79",
"name": "News MCP Client",
"type": "@n8n/n8n-nodes-langchain.mcpClientTool",
"position": [
384,
48
],
"parameters": {
"endpointUrl": "<<<REPLACE_WITH_ENDPOINT_URL>>>",
"serverTransport": "httpStreamable"
},
"typeVersion": 1.1
},
{
"id": "2e931983-39af-4b1d-9a16-e30cd536ff0b",
"name": "Search in Tavily1",
"type": "@tavily/n8n-nodes-tavily.tavilyTool",
"position": [
848,
480
],
"parameters": {
"query": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Query', ``, 'string') }}",
"options": {}
},
"typeVersion": 1
},
{
"id": "c8fc2868-c029-454f-b47c-6cf2a4f2fb7c",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1024,
-432
],
"parameters": {
"width": 736,
"height": 1808,
"content": "AI Agent MCP for Email & News Research \n\nBuild a chat-first MCP-powered research and outreach agent. This workflow lets you ask questions in an n8n chat, then the agent researches news (via Tavily + Perplexity through an MCP server) and drafts emails (via Gmail through a separate MCP server). It uses OpenAI for reasoning and short-term memory for coherent, multi\u2011turn conversations.\n\nWatch build along videos for workflows like these on: www.youtube.com/@automatewithmarc\n\nWhat this template does\n\nChat-native trigger: Start a conversation and ask for research or an email draft.\n\nMCP client tools: The agent talks to two MCP servers \u2014 one for Email work, one for News research.\n\nNews research stack: Uses Tavily (search) and Perplexity (LLM retrieval/answers) behind a News MCP server.\n\nEmail stack: Uses Gmail Tool to generate and send messages via an Email MCP server.\n\nReasoning + memory: OpenAI Chat Model + Simple Memory for context-aware, multi-step outputs.\n\nHow it works (node map)\n\nWhen chat message received \u2192 collects your prompt and routes it to the agent.\n\nAI Agent (system prompt = \u201chelpful email assistant\u201d) \u2192 orchestrates tools via MCP Clients.\n\nOpenAI Chat Model \u2192 reasoning/planning for research or email drafting.\n\nSimple Memory \u2192 keeps recent chat context for follow-ups.\n\nNews MCP Server exposes:\n\nTavily Tool (Search) and Perplexity Tool (Ask) for up-to-date findings.\n\nEmail MCP Server exposes:\n\nGmail Tool (To, Subject, Message via AI fields) to send or draft emails.\n\nThe MCP Clients (News/Email) plug into the Agent, so your single chat prompt can research and then draft/send emails in one flow.\n\nRequirements\n\nn8n (Cloud or self\u2011hosted)\n\nOpenAI API key for the Chat Model (set on the node)\n\nTavily, Perplexity, and Gmail credentials (connected on their respective tool nodes)\n\nPublicly reachable MCP Server endpoints (provided in the MCP Client nodes)\n\nSetup (quick start)\n\nImport the template and open it in the editor.\n\nConnect credentials on: OpenAI, Tavily, Perplexity, and Gmail tool nodes.\n\nConfirm MCP endpoints in both MCP Client nodes (News/Email) and leave transport as httpStreamable unless you have special requirements.\n\nRun the workflow. In chat, try:\n\n\u201cFind today\u2019s top stories on Kubernetes security and draft an intro email to Acme.\u201d\n\n\u201cSummarize the latest AI infra trends and email a 3\u2011bullet update to my team.\u201d\n\nInputs & outputs\n\nInput: Natural-language prompt via chat trigger.\n\nTools used: News MCP (Tavily + Perplexity), Email MCP (Gmail).\n\nOutput: A researched summary and/or a drafted/sent email, returned in the chat and executed via Gmail when requested.\n\nWhy teams will love it\n\nOne prompt \u2192 research + outreach: No tab\u2011hopping between tools.\n\nUp-to-date answers: Pulls current info through Tavily/Perplexity.\n\nEmail finalization: Converts findings into send-ready drafts via Gmail.\n\nContext-aware: Memory keeps threads coherent across follow-ups.\n\nPro tips\n\nUse clear verbs in your prompt: \u201cResearch X, then email Y with Z takeaways.\u201d\n\nFor safer runs, point Gmail to a test inbox first (or disable send and only draft).\n\nAdd guardrails in the Agent\u2019s system message to match your voice/tone."
},
"typeVersion": 1
},
{
"id": "226bc7c3-d026-4dea-adec-1d8fc5a5481b",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-144,
-304
],
"parameters": {
"color": 5,
"width": 928,
"height": 512,
"content": "Agent & MCP Client"
},
"typeVersion": 1
},
{
"id": "4d9280da-af9b-4eab-be1a-9c25a6258022",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-256,
224
],
"parameters": {
"color": 6,
"width": 672,
"height": 512,
"content": "Email MCP Server"
},
"typeVersion": 1
},
{
"id": "f55f5515-090b-4c3d-9e60-49e0588292a4",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
432,
224
],
"parameters": {
"color": 7,
"width": 672,
"height": 512,
"content": "News Research MCP Server"
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"connections": {
"Simple Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"News MCP Client": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Email MCP Client": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Search in Tavily": {
"ai_tool": [
[
{
"node": "News MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Search in Tavily1": {
"ai_tool": [
[
{
"node": "News MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
},
"Send a message in Gmail": {
"ai_tool": [
[
{
"node": "Email MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
},
"Send a message in Gmail1": {
"ai_tool": [
[
{
"node": "Email MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
},
"Send a message in Gmail2": {
"ai_tool": [
[
{
"node": "Email MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
},
"When chat message received": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Message a model in Perplexity": {
"ai_tool": [
[
{
"node": "News MCP Server",
"type": "ai_tool",
"index": 0
}
]
]
}
}
}
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
AI Agent MCP for Email & News Research
Source: https://n8n.io/workflows/7679/ — 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 is designed for Shopify store owners and e-commerce managers who want to automate their store operations through an intelligent AI assistant. The workflow creates a conversational in
Shopify Tool, Discord, Telegram +10
Description: Turn your WooCommerce store into a conversational AI assistant — create products, place orders, run reports and manage coupons using natural language via n8n + an MCP Server.
Agent, OpenAI Chat, Mcp Client Tool +4
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
The Project starter bot takes the hassle out of launching projects by automatically creating a well-structured folder system in Dropbox and sending timely notifications through Slack and Gmail. By com
Chat Trigger, Agent, Memory Buffer Window +7
Personal Assistant MCP server. Uses chatTrigger, lmChatGoogleGemini, memoryBufferWindow, mcpTrigger. Chat trigger; 20 nodes.
Chat Trigger, Google Gemini Chat, Memory Buffer Window +6