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 →
{
"id": "DnHvQ3KL8v8r5L5Z",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Telegram Chat with Buffering",
"tags": [],
"nodes": [
{
"id": "a3cc74e9-c696-48de-a04e-d48555641897",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1640,
-800
],
"parameters": {
"color": 7,
"width": 220,
"height": 280,
"content": "## 1. Receive Message\n\n"
},
"typeVersion": 1
},
{
"id": "ff18667d-0a31-4768-acf8-ed0d53b2f382",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
160,
-840
],
"parameters": {
"color": 7,
"width": 600,
"height": 520,
"content": "## 3. AI Assistant\n"
},
"typeVersion": 1
},
{
"id": "ce90f954-19b6-4224-ae88-b20c4da639e6",
"name": "Reply",
"type": "n8n-nodes-base.telegram",
"position": [
920,
-700
],
"parameters": {
"text": "={{ $json.output }}",
"chatId": "={{ $('Receive Message').item.json.message.chat.id }}",
"additionalFields": {
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "6f46d89b-034c-47ea-a217-8d007bec1531",
"name": "Receive Message",
"type": "n8n-nodes-base.telegramTrigger",
"position": [
-1580,
-680
],
"parameters": {
"updates": [
"message"
],
"additionalFields": {}
},
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.1
},
{
"id": "0f391daa-0e74-4058-8923-52f3c050c9ad",
"name": "Wait 10 Seconds",
"type": "n8n-nodes-base.wait",
"position": [
-1000,
-580
],
"parameters": {
"amount": 10
},
"typeVersion": 1.1
},
{
"id": "8e6495d8-db6e-4692-ade5-45239049de34",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1320,
-760
],
"parameters": {
"color": 7,
"width": 1400,
"height": 440,
"content": "## 2. Buffer Incoming Messages"
},
"typeVersion": 1
},
{
"id": "d4876fd2-2e0b-4f82-9dc3-553f926310bd",
"name": "Add to Queued Messages",
"type": "n8n-nodes-base.supabase",
"position": [
-1240,
-680
],
"parameters": {
"tableId": "message_queue",
"fieldsUi": {
"fieldValues": [
{
"fieldId": "user_id",
"fieldValue": "={{ $json.message.chat.id }}"
},
{
"fieldId": "message",
"fieldValue": "={{ $json.message.text }}"
},
{
"fieldId": "message_id",
"fieldValue": "={{ $json.message.message_id }}"
}
]
}
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "a2eeb77f-2d74-44ac-9812-c3659d2e2803",
"name": "No Operation, do nothing",
"type": "n8n-nodes-base.noOp",
"position": [
-340,
-460
],
"parameters": {},
"typeVersion": 1
},
{
"id": "638fc82e-aba1-4deb-b506-33dcf4746896",
"name": "Aggregate",
"type": "n8n-nodes-base.aggregate",
"position": [
220,
-700
],
"parameters": {
"options": {},
"fieldsToAggregate": {
"fieldToAggregate": [
{
"fieldToAggregate": "message"
}
]
}
},
"typeVersion": 1
},
{
"id": "772f60e5-e52f-4779-aa03-e4d532ee4b5c",
"name": "Delete Queued Messages",
"type": "n8n-nodes-base.supabase",
"position": [
-100,
-700
],
"parameters": {
"filters": {
"conditions": [
{
"keyName": "user_id",
"keyValue": "={{ $json.user_id }}",
"condition": "eq"
}
]
},
"tableId": "message_queue",
"operation": "delete"
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "16b46a70-85a0-4c8c-94ba-172ebe9aafa4",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
860,
-780
],
"parameters": {
"color": 7,
"width": 280,
"height": 260,
"content": "## 4. Send Reply\n\n\n"
},
"typeVersion": 1
},
{
"id": "9162f110-465f-4cd6-9f03-17751d7e43a4",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
380,
-460
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1.2
},
{
"id": "b47ef0c9-725b-4837-b9e9-96a4ff2b3636",
"name": "Sort by Message ID",
"type": "n8n-nodes-base.sort",
"position": [
-580,
-680
],
"parameters": {
"options": {},
"sortFieldsUi": {
"sortField": [
{
"fieldName": "message_id"
}
]
}
},
"typeVersion": 1
},
{
"id": "1aa80c99-eec8-4174-bcf3-c6873354ed0f",
"name": "Get Queued Messages",
"type": "n8n-nodes-base.supabase",
"position": [
-780,
-680
],
"parameters": {
"filters": {
"conditions": [
{
"keyName": "user_id",
"keyValue": "={{ $('Receive Message').item.json.message.from.id }}",
"condition": "eq"
}
]
},
"tableId": "message_queue",
"operation": "getAll",
"returnAll": true
},
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "85050328-b5aa-47fe-802c-7d9f31f225cb",
"name": "Check Most Recent Message",
"type": "n8n-nodes-base.if",
"position": [
-360,
-680
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "loose"
},
"combinator": "and",
"conditions": [
{
"id": "8852bab7-230e-442a-a4a2-994e979c8f9f",
"operator": {
"type": "number",
"operation": "equals"
},
"leftValue": "={{ $input.last().json.message_id }}\n",
"rightValue": "={{ $('Receive Message').item.json.message.message_id }}"
}
]
},
"looseTypeValidation": true
},
"typeVersion": 2.2
},
{
"id": "bed86d81-bb57-42ce-aaa7-4bdc21e1651c",
"name": "AI Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
420,
-700
],
"parameters": {
"text": "={{ $json.message.join(String.fromCharCode(10)) }}",
"options": {},
"promptType": "define"
},
"typeVersion": 1.7
},
{
"id": "4f468a14-fbea-44ec-a2b8-e4b3785c0362",
"name": "Postgres Chat Memory",
"type": "@n8n/n8n-nodes-langchain.memoryPostgresChat",
"position": [
560,
-460
],
"parameters": {
"sessionKey": "={{ $('Receive Message').item.json.message.chat.id }}",
"sessionIdType": "customKey"
},
"credentials": {
"postgres": {
"name": "<your credential>"
}
},
"typeVersion": 1.3
},
{
"id": "610516e8-d4ad-448e-ac97-17aad1a31862",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
-2420,
-820
],
"parameters": {
"width": 700,
"height": 420,
"content": "## Allow Users to Send a Sequence of Messages to an AI Agent in Telegram with Supabase\n### Use Case\nWhen creating chatbots that interface through applications such as **Telegram** and **WhatsApp**, users can often sends multiple shorter messages in quick succession, in place of a single, longer message. This workflow accounts for this behaviour.\n### What it Does\nThis workflow allows users to send several messages in quick succession, treating them as one coherent conversation instead of separate messages requiring individual responses. \n### How it Works\n1. When messages arrive, they are stored in a **Supabase PostgreSQL** table\n2. The system waits briefly to see if additional messages arrive\n3. If no new messages arrive within the waiting period, all queued messages are:\n - Combined and processed as a single conversation\n - Responded to with one unified reply\n - Deleted from the queue"
},
"typeVersion": 1
},
{
"id": "c8bd8777-fb0f-4941-8674-f5bb7c264506",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1640,
-1060
],
"parameters": {
"width": 520,
"height": 220,
"content": "### Setup\n1. Create a table in Supabase called **message_queue**. It needs to have the following columns: **user_id** (`uint8`), **message** (`text`), and **message_id** (`uint8`)\n2. Add your **Telegram**, **Supabase**, **OpenAI**, and **PostgreSQL** credentials\n3. Activate the workflow and test by sending multiple messages the Telegram bot in one go\n4. Wait ten seconds after which you will receive a single reply to all of your messages"
},
"typeVersion": 1
},
{
"id": "24604fc7-7957-4e20-8303-b31f2ce1e257",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1060,
-700
],
"parameters": {
"color": 5,
"width": 220,
"height": 280,
"content": "### Modification\nChange the value of *Wait Amount* to vary the buffering window"
},
"typeVersion": 1
},
{
"id": "24f388f3-5655-4bd4-9c30-978efb2dc400",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
180,
-480
],
"parameters": {
"color": 5,
"width": 340,
"height": 140,
"content": "### Modification\nReplace this sub-node \nto use a different language\n model"
},
"typeVersion": 1
},
{
"id": "3db12526-6b97-4e3a-b53d-987f5d20c46e",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"position": [
380,
-800
],
"parameters": {
"color": 5,
"width": 340,
"height": 240,
"content": "### Modification\nAdd a **System Message** to tailor the chatbot to your use case"
},
"typeVersion": 1
}
],
"active": true,
"settings": {
"callerPolicy": "workflowsFromSameOwner",
"executionOrder": "v1"
},
"versionId": "e415eb18-1bb9-426b-b759-0ba269db1f8f",
"connections": {
"AI Agent": {
"main": [
[
{
"node": "Reply",
"type": "main",
"index": 0
}
]
]
},
"Aggregate": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Receive Message": {
"main": [
[
{
"node": "Add to Queued Messages",
"type": "main",
"index": 0
}
]
]
},
"Wait 10 Seconds": {
"main": [
[
{
"node": "Get Queued Messages",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Sort by Message ID": {
"main": [
[
{
"node": "Check Most Recent Message",
"type": "main",
"index": 0
}
]
]
},
"Get Queued Messages": {
"main": [
[
{
"node": "Sort by Message ID",
"type": "main",
"index": 0
}
]
]
},
"Postgres Chat Memory": {
"ai_memory": [
[
{
"node": "AI Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Add to Queued Messages": {
"main": [
[
{
"node": "Wait 10 Seconds",
"type": "main",
"index": 0
}
]
]
},
"Delete Queued Messages": {
"main": [
[
{
"node": "Aggregate",
"type": "main",
"index": 0
}
]
]
},
"Check Most Recent Message": {
"main": [
[
{
"node": "Delete Queued Messages",
"type": "main",
"index": 0
}
],
[
{
"node": "No Operation, do nothing",
"type": "main",
"index": 0
}
]
]
}
}
}
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.
openAiApipostgressupabaseApitelegramApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
This workflow enables seamless, buffered conversations in Telegram, ensuring AI responses are thoughtful and context-aware without overwhelming the chat. It's ideal for developers or businesses building interactive bots that handle user queries intelligently, using integrations like Telegram and Supabase to store and retrieve message history. The key step involves queuing incoming messages in Supabase for processing, followed by an AI-driven reply that maintains conversation flow, delivering reliable engagement even during high-volume interactions.
Use this workflow when creating persistent chat experiences, such as customer support bots or virtual assistants, where response delays prevent rushed outputs. Avoid it for real-time applications like live gaming, as the built-in 10-second wait and buffering introduce intentional pauses. Common variations include swapping Supabase for another database or integrating additional AI models for specialised responses.
About this workflow
Telegram Chat with Buffering. Uses stickyNote, telegram, telegramTrigger, supabase. Event-driven trigger; 22 nodes.
Source: https://github.com/Zie619/n8n-workflows — 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.
Get notification each time cryptoreaches X. Uses telegramTrigger, telegram, agent, supabase. Event-driven trigger; 34 nodes.
Telegram Trigger, Telegram, Agent +6
When creating chatbots that interface through applications such as Telegram and WhatsApp, users can often sends multiple shorter messages in quick succession, in place of a single, longer message. Thi
Telegram, Telegram Trigger, Supabase +3
Bitlab-Chatbot. Uses telegramTrigger, telegram, snowflake, httpRequest. Event-driven trigger; 87 nodes.
Telegram Trigger, Telegram, Snowflake +13
Personal Assistant. Uses memoryBufferWindow, agent, agentTool, httpRequestTool. Event-driven trigger; 77 nodes.
Memory Buffer Window, Agent, Agent Tool +15
System Architecture Two integrated N8N workflows providing automated US stock portfolio management through Telegram:
Output Parser Autofixing, OpenAI Chat, Perplexity +10