Davide Boizza This workflow corresponds to n8n.io template #9193 — we link there as the canonical source.
This workflow follows the Agent → Gmail 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": "mYOyC65TtRuJhvzs",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Send Gmail replies by analyzing Thread ID",
"tags": [],
"nodes": [
{
"id": "c1469689-3e67-4a07-9d5e-374552d164e7",
"name": "Gmail Trigger",
"type": "n8n-nodes-base.gmailTrigger",
"position": [
-464,
-272
],
"parameters": {
"filters": {},
"pollTimes": {
"item": [
{
"mode": "everyMinute"
}
]
}
},
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "072aba43-704d-4663-a6e7-a5f7e9ad942d",
"name": "Get a thread",
"type": "n8n-nodes-base.gmail",
"position": [
128,
-64
],
"parameters": {
"options": {
"returnOnlyMessages": true
},
"resource": "thread",
"threadId": "={{ $json.threadId }}",
"operation": "get"
},
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
},
"typeVersion": 2.1
},
{
"id": "972b0ab6-a6a4-4409-bc7c-4b06f3f06255",
"name": "Create a draft",
"type": "n8n-nodes-base.gmail",
"position": [
1232,
-64
],
"parameters": {
"message": "={{ $json.output }}",
"options": {
"ccList": "={{ $ifEmpty($('Gmail Trigger').item.json.Cc, '') }}",
"sendTo": "={{ $('Gmail Trigger').item.json.From }}",
"threadId": "={{ $('Gmail Trigger').item.json.threadId }}"
},
"subject": "={{ $('Gmail Trigger').item.json.Subject }}",
"resource": "draft",
"emailType": "html"
},
"credentials": {
"gmailOAuth2": {
"name": "<your credential>"
}
},
"typeVersion": 2.1
},
{
"id": "9b255334-0339-4d8c-932d-3e7b759476b4",
"name": "Merge",
"type": "n8n-nodes-base.merge",
"position": [
960,
-256
],
"parameters": {
"mode": "combine",
"options": {},
"combineBy": "combineAll"
},
"typeVersion": 3.2
},
{
"id": "67777b0c-98d3-46dd-bc8c-036946048aa1",
"name": "OpenAI Chat Model1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
-304,
144
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-5-nano",
"cachedResultName": "gpt-5-nano"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "7e958566-cf0f-4b20-bb7a-1e5115aa0635",
"name": "Email Classifier",
"type": "@n8n/n8n-nodes-langchain.textClassifier",
"position": [
-224,
-64
],
"parameters": {
"options": {
"fallback": "discard",
"enableAutoFixing": true,
"systemPromptTemplate": "Please classify the text provided by the user into one of the following categories: {categories}, and use the provided formatting instructions below. Don't explain, and only output the json."
},
"inputText": "={{ $json.snippet }}",
"categories": {
"categories": [
{
"category": "ok",
"description": "Non-advertising, non-newsletter, non-spam emails that deserve a response"
}
]
}
},
"typeVersion": 1.1
},
{
"id": "84a4f669-17af-4dbb-8ccf-660751316e38",
"name": "Replying email Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
560,
-64
],
"parameters": {
"text": "={{ JSON.stringify($json.data) }}",
"options": {
"systemMessage": "You are an expert in replying to emails that arrive in my inbox. \nYou will be provided with the conversation history. \nAnalyze it carefully and draft a reply. \n\n- Do not invent information. \n- If some variables are missing or unknown, insert clear placeholders like {{variable}}. \n- Your output must be ONLY the email reply in valid HTML format, without any preamble, notes, or explanations. "
},
"promptType": "define"
},
"typeVersion": 2.2
},
{
"id": "7496e0e8-05d8-4f37-b6a7-abe91d7fea73",
"name": "Code",
"type": "n8n-nodes-base.code",
"position": [
336,
-64
],
"parameters": {
"jsCode": "const results = [];\n\nfor (const item of $input.all()) {\n let emails = item.json;\n\n // Normalizza in array:\n if (!emails) {\n emails = [];\n } else if (!Array.isArray(emails)) {\n // Se \u00e8 avvolto in una propriet\u00e0 comune\n if (Array.isArray(emails.emails)) {\n emails = emails.emails;\n } else if (Array.isArray(emails.messages)) {\n emails = emails.messages;\n } else if (typeof emails === 'object' && (emails.id || emails.snippet || emails.From || emails.To)) {\n // singolo oggetto email -> mettilo in array\n emails = [emails];\n } else {\n // fallback: non iterabile\n emails = [];\n }\n }\n\n for (const email of emails) {\n results.push({\n snippet: email.snippet ?? '',\n from: email.From ?? email.from ?? '',\n to: email.To ?? email.to ?? '',\n });\n }\n}\n\n// Output unico con propriet\u00e0 \"data\"\nreturn [\n {\n json: {\n data: results,\n },\n },\n];\n\n\n"
},
"typeVersion": 2
},
{
"id": "6d5cc052-a0b6-403d-a6ea-efaed97ab351",
"name": "Anthropic Sonnet 4.5",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"position": [
496,
144
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "claude-sonnet-4-5-20250929",
"cachedResultName": "Claude Sonnet 4.5"
},
"options": {}
},
"credentials": {
"anthropicApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "a9070e89-5e15-4110-b9a7-3f88fdde7990",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-480,
-848
],
"parameters": {
"color": 3,
"width": 704,
"height": 208,
"content": "## AI-Powered Gmail Assistant: send replies by analyzing Thread ID with Sonnet 4.5\n \nThis workflow automates **analyzing Gmail threads and drafting AI-powered replies** with the new model Anthropic **Sonnet 4.5**.\n\nThis workflow automates the process of analyzing incoming emails and generating context-aware draft replies by examining the entire email thread."
},
"typeVersion": 1
},
{
"id": "9c238691-41c5-4f32-98f6-6ac1239063b1",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-480,
-608
],
"parameters": {
"width": 704,
"height": 224,
"content": "## STEP 1\n \n* **Gmail OAuth2 Credentials:** The workflow uses the same Gmail account for the trigger, fetching threads, and creating drafts. Configure this in the \"Gmail Trigger,\" \"Get a thread,\" and \"Create a draft\" nodes.\n * **OpenAI API Credentials:** Required for both the \"Email Classifier\". Provide your API key in the respective OpenAI Chat Model nodes.\n \n * **Anthropic API Credentials:** Required for the main \"Replying email Agent.\" Provide your API key in the respective Antrhopic Chat Model nodes."
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "9ddd07a7-fb73-4016-b1d1-90e59be8be84",
"connections": {
"Code": {
"main": [
[
{
"node": "Replying email Agent",
"type": "main",
"index": 0
}
]
]
},
"Merge": {
"main": [
[
{
"node": "Create a draft",
"type": "main",
"index": 0
}
]
]
},
"Get a thread": {
"main": [
[
{
"node": "Code",
"type": "main",
"index": 0
}
]
]
},
"Gmail Trigger": {
"main": [
[
{
"node": "Merge",
"type": "main",
"index": 0
},
{
"node": "Email Classifier",
"type": "main",
"index": 0
}
]
]
},
"Email Classifier": {
"main": [
[
{
"node": "Get a thread",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model1": {
"ai_languageModel": [
[
{
"node": "Email Classifier",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Anthropic Sonnet 4.5": {
"ai_languageModel": [
[
{
"node": "Replying email Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Replying email Agent": {
"main": [
[
{
"node": "Merge",
"type": "main",
"index": 1
}
]
]
}
}
}
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.
anthropicApigmailOAuth2openAiApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
This workflow automates analyzing Gmail threads and drafting AI-powered replies with the new model Anthropic Sonnet 4.5.
Source: https://n8n.io/workflows/9193/ — 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 template and YouTube video goes over 5 different implementations of evaluations within n8n. Categorization Correctness Tools used String similarity Helpfulness
Evaluation, Evaluation Trigger, Google Gemini Chat +8
RESUME SCREENER. Uses agent, outputParserStructured, lmChatOpenAi, gmailTrigger. Event-driven trigger; 37 nodes.
Agent, Output Parser Structured, OpenAI Chat +7
AI Agents Vs AI Workflow. Uses lmChatOpenAi, gmailTrigger, gmail, gmailTool. Event-driven trigger; 30 nodes.
OpenAI Chat, Gmail Trigger, Gmail +7
E-mail Assistant Agent. Uses gmailTrigger, textClassifier, gmail, lmChatOpenAi. Event-driven trigger; 29 nodes.
Gmail Trigger, Text Classifier, Gmail +6
This workflow is ideal for: Customer support teams looking to reduce manual response time SaaS companies that frequently receive product inquiries E-commerce stores with common customer questions abou
OpenAI Chat, Output Parser Structured, Google Sheets Tool +5