This workflow follows the Agent → Documentdefaultdataloader 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": "n8n telegram RAG",
"nodes": [
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"typeVersion": 1,
"position": [
1792,
224
],
"id": "6c207a00-ccc3-4c16-ba3b-38478da7dc13",
"name": "Google Gemini Chat Model",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {},
"type": "@n8n/n8n-nodes-langchain.embeddingsGoogleGemini",
"typeVersion": 1,
"position": [
2144,
448
],
"id": "4cb8447a-dcf9-4f04-adc3-6159da61309f",
"name": "Embeddings Google Gemini",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.memoryManager",
"typeVersion": 1.1,
"position": [
1360,
0
],
"id": "c30cb0da-8c6a-408f-b1ec-05ac2fb5e7c3",
"name": "Chat Memory Manager"
},
{
"parameters": {
"mode": "retrieve-as-tool",
"toolDescription": "Use this to search and answer user queries",
"tableName": {
"__rl": true,
"value": "documents",
"mode": "list",
"cachedResultName": "documents"
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"typeVersion": 1.3,
"position": [
2144,
256
],
"id": "74d35bb2-ba95-46c0-a234-56a75f7b3f19",
"name": "RAG_Internal_Knowledge_Base",
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {},
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"typeVersion": 1.3,
"position": [
1360,
240
],
"id": "d2321648-dd7a-4cd5-917e-1564e555f25d",
"name": "Simple Memory1"
},
{
"parameters": {
"chatId": "={{ $('Set Session ID').item.json.sessionId }}",
"text": "={{ $json.draft_answer }}",
"additionalFields": {
"appendAttribution": false
}
},
"type": "n8n-nodes-base.telegram",
"typeVersion": 1.2,
"position": [
3584,
16
],
"id": "deaa6e3f-34a5-4562-b5e0-e6836882938a",
"name": "Send a text message",
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"promptType": "define",
"text": "={{ $json.query }}\nYou are a STRICT RAG-based assistant. \nYou do NOT use your own knowledge. \nYou do NOT invent answers. \nYou only answer using information retrieved from the tool:\n\"RAG_Internal_Knowledge_Base\".\n\n====================================================\n### ABSOLUTE RULES\n- You MUST use the RAG tool for EVERY user question.\n- You MUST NOT answer from your own memory or general training.\n- If the tool returns no relevant context, say:\n \"I could not find this information in the knowledge base.\"\n\n- Never guess.\n- Never hallucinate.\n- Never apologize for limitations beyond:\n \u201cThis information does not appear in the knowledge base.\u201d\n\n====================================================\n### HOW TO USE THE RAG TOOL\nAlways call the RAG tool using:\n\nRAG_Internal_Knowledge_Base({\n \"q\": \"<the user's query or a refined version>\"\n})\n\nExamples:\nUser: \"Summarize the pdf.\"\n\u2192 RAG_Internal_Knowledge_Base({ \"q\": \"summarize the PDF\" })\n\nUser: \"What does the onboarding document say?\"\n\u2192 RAG_Internal_Knowledge_Base({ \"q\": \"onboarding document content\" })\n\n====================================================\n### AFTER TOOL CALL\nOnce you receive retrieved document chunks:\n- Read them carefully.\n- Synthesize them into a clear, structured answer.\n- Only use retrieved content.\n- If chunks do not contain the answer, say:\n \u201cThe knowledge base does not contain this information.\u201d\n\n====================================================\n### RESPONSE STYLE\n- Be concise.\n- Use bullets or short paragraphs.\n- Stay factual.\n- Stay grounded only in tool results.\n- Never use outside world knowledge.\n- Never answer without running the tool first.\n\n====================================================\n### IMPORTANT\nIf a user asks something not connected to the documents:\n- STILL call the RAG tool.\n- If the tool returns nothing:\n \u201cI could not find relevant information in the knowledge base.\u201d\n",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 3,
"position": [
1984,
0
],
"id": "fc7b6d94-2efd-4838-a595-123aca798e06",
"name": "AI Agent"
},
{
"parameters": {
"updates": [
"message"
],
"additionalFields": {}
},
"type": "n8n-nodes-base.telegramTrigger",
"typeVersion": 1.2,
"position": [
96,
0
],
"id": "14b93131-ff1c-4d2f-abe6-aa8be4df66d1",
"name": "Telegram Trigger",
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "485243bf-1436-456d-88ca-04f96aa4483f",
"leftValue": "={{ $json.content.parts[0].text }}",
"rightValue": "bad",
"operator": {
"type": "string",
"operation": "equals"
}
}
],
"combinator": "or"
},
"options": {}
},
"type": "n8n-nodes-base.if",
"typeVersion": 2.2,
"position": [
2880,
0
],
"id": "bc864773-fd98-4c83-b775-37f02bcbfa2f",
"name": "If"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "78ae84e9-6379-4967-9776-20181428d78f",
"name": "query",
"value": "={{ $('Telegram Trigger').item.json.message.text }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
1696,
0
],
"id": "292b8297-24f0-4a30-bd71-98e574ee177e",
"name": "Edit Fields1"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "ae8cbfc6-e253-4cab-847e-21264f223384",
"name": "draft_answer",
"value": "={{ $json.output }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
2304,
0
],
"id": "d01c743a-b545-4f43-98f9-2eac263f1b6a",
"name": "Edit Fields3"
},
{
"parameters": {
"modelId": {
"__rl": true,
"value": "models/gemini-2.5-flash",
"mode": "list",
"cachedResultName": "models/gemini-2.5-flash"
},
"messages": {
"values": [
{
"content": "=Here is a draft answer that was rated as bad:\n{{ $json.draft_answer }}\n\nImprove it to be clearer, more helpful, and more accurate.\nReturn ONLY the improved answer. No explanations.\n"
}
]
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"typeVersion": 1,
"position": [
3056,
-224
],
"id": "b2f5ccd3-e775-4d1d-bcbb-e21f328f596f",
"name": "Improver Model",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"modelId": {
"__rl": true,
"value": "models/gemini-2.5-flash",
"mode": "list",
"cachedResultName": "models/gemini-2.5-flash"
},
"messages": {
"values": [
{
"content": "=Draft answer:\n{{$json.draft_answer}}\n\nYou are an answer evaluator.\nReturn only one word:\ngood\nbad\n"
}
]
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"typeVersion": 1,
"position": [
2496,
0
],
"id": "373bfee3-e69f-4ead-baf3-c998f8bdac1f",
"name": "Evaluator",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "71248aea-1e92-46cc-97ec-9f41cb5c0aed",
"name": "draft_answer",
"value": "={{ $('Edit Fields3').item.json.draft_answer }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
3200,
16
],
"id": "8e923185-22d7-4c90-b57c-f8fd59432009",
"name": "Edit Fields"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "6f4b7463-9f6a-4f84-b4b7-9a4074ce43ab",
"name": "final_answer",
"value": "={{ $json.content.parts[0].text }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
3376,
-224
],
"id": "d7de5394-81d6-4bb2-afe8-e84f21b67681",
"name": "Set Final Answer (Improved)"
},
{
"parameters": {
"rule": {
"interval": [
{
"field": "seconds",
"secondsInterval": 2
}
]
}
},
"type": "n8n-nodes-base.scheduleTrigger",
"typeVersion": 1.3,
"position": [
-48,
976
],
"id": "c0386789-0c08-4dc0-b558-0591ad3b96e2",
"name": "Schedule Trigger"
},
{
"parameters": {
"resource": "fileFolder",
"filter": {
"folderId": {
"__rl": true,
"value": "15P-KRMVzTSos7Bh7HVTKmdVGAZ-nXgDA",
"mode": "id"
}
},
"options": {}
},
"type": "n8n-nodes-base.googleDrive",
"typeVersion": 3,
"position": [
160,
976
],
"id": "680bd2ad-c153-4d5f-9503-fc4b4be7d51c",
"name": "Search files and folders",
"credentials": {
"googleDriveOAuth2Api": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"operation": "download",
"fileId": {
"__rl": true,
"mode": "id",
"value": "={{ $json.id }}"
},
"options": {}
},
"type": "n8n-nodes-base.googleDrive",
"typeVersion": 3,
"position": [
368,
976
],
"id": "406f2c85-2915-4da6-a0ab-2e90cafe1c5f",
"name": "Download file",
"credentials": {
"googleDriveOAuth2Api": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"mode": "insert",
"tableName": {
"__rl": true,
"value": "documents",
"mode": "list",
"cachedResultName": "documents"
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.vectorStoreSupabase",
"typeVersion": 1.3,
"position": [
576,
976
],
"id": "e515ea12-0371-4462-85a0-ef10a9738a0c",
"name": "Supabase Vector Store",
"credentials": {
"supabaseApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.textSplitterRecursiveCharacterTextSplitter",
"typeVersion": 1,
"position": [
368,
1392
],
"id": "ef0c5922-95ac-41a1-8c77-53e02238a8cf",
"name": "Recursive Character Text Splitter"
},
{
"parameters": {
"dataType": "binary",
"textSplittingMode": "custom",
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.documentDefaultDataLoader",
"typeVersion": 1.1,
"position": [
720,
1184
],
"id": "b22af7b3-596e-44d2-9c5e-af02e84479b0",
"name": "Default Data Loader"
},
{
"parameters": {},
"type": "@n8n/n8n-nodes-langchain.embeddingsGoogleGemini",
"typeVersion": 1,
"position": [
576,
1168
],
"id": "02c2612c-6432-4f05-b8b2-bdb529889a3c",
"name": "Embeddings Google Gemini1",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "bde8203d-6fac-4172-beb0-416df51565b3",
"name": "sessionId",
"value": "={{$json.message.chat.id}}",
"type": "number"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
1152,
0
],
"id": "0c68c57f-994b-4097-a29d-119a73b0c12e",
"name": "Set Session ID"
},
{
"parameters": {
"content": "# Data Ingestion pipeline",
"height": 768,
"width": 1136,
"color": 4
},
"type": "n8n-nodes-base.stickyNote",
"position": [
-144,
880
],
"typeVersion": 1,
"id": "72173cc7-f056-47a7-b073-f5a57227ca23",
"name": "Sticky Note1"
},
{
"parameters": {
"url": "={{ $json.url }}",
"options": {
"response": {
"response": {
"fullResponse": true,
"responseFormat": "text"
}
}
}
},
"type": "n8n-nodes-base.httpRequest",
"typeVersion": 4.3,
"position": [
944,
-320
],
"id": "e9339a6d-41b0-494a-a905-9252b5129aaa",
"name": "HTTP Request"
},
{
"parameters": {
"chatId": "={{ $('Telegram Trigger').item.json.message.chat.id }}",
"text": "={{ $json.content.parts[0].text }}",
"additionalFields": {
"appendAttribution": false,
"parse_mode": "HTML"
}
},
"type": "n8n-nodes-base.telegram",
"typeVersion": 1.2,
"position": [
1840,
-320
],
"id": "06a7e1d7-3565-4ddb-9bdb-cd7f84d31714",
"name": "Send Summarization Message",
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"modelId": {
"__rl": true,
"value": "models/gemini-2.5-flash",
"mode": "list",
"cachedResultName": "models/gemini-2.5-flash"
},
"messages": {
"values": [
{
"content": "=You will receive raw HTML from a webpage.\n\nTASK:\nExtract ONLY human-readable article text.\nDo NOT include:\n- HTML tags\n- JavaScript\n- CSS\n- Boilerplate\n- Menus, headers, footers\n- Code examples\n- Explanations\n- Meta descriptions\n- Ads\n- Python code or suggestions\n\nVery important:\nReturn ONLY the extracted plain text from the article.\nNo introductions, no explanations, no notes.\n\nHTML:\n{{ $json.data }}\n"
}
]
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"typeVersion": 1,
"position": [
1136,
-320
],
"id": "396e076f-eeb2-45e5-9d16-d04475bc6d98",
"name": "Extract Human Readable Text",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"mode": "expression",
"numberOutputs": 3,
"output": "={{\n $json.message.text.startsWith('/help') ? 2 :\n $json.message.text.match(/https?:\\/\\/\\S+/) ? 0 :\n $json.message.text.startsWith('/summarize') ? 0 :\n $json.message.text.startsWith('/analyze') ? 0 :\n $json.message.text.startsWith('/explain') ? 0 :\n $json.message.text.startsWith('/insights') ? 0 :\n 1\n}}"
},
"type": "n8n-nodes-base.switch",
"typeVersion": 3.3,
"position": [
368,
-16
],
"id": "e3766468-7c41-4045-8c62-09cf16adb408",
"name": "Switch"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "1e04be32-d683-4297-9251-dc58212f5cba",
"name": "url",
"value": "={{ $json.message.text.match(/https?:\\/\\/\\S+/)?.[0] || \"\" }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
704,
-256
],
"id": "525098da-f46c-4b60-95e0-321737ad0c8f",
"name": "Edit Fields2"
},
{
"parameters": {
"chatId": "={{ $json.message.chat.id }}",
"text": "\ud83e\udd16 Available Commands\n\n/help - for all commands\n\n/summarize <url> -> Summarize the webpage at url\n\n/analyze <url> -> Analyze the webpage at url\n\n/explain <url> -> Explain the webpage at url\n\n/insights <url> -> Insights of the webpage at url\n\n---\n\n\ud83d\udd17 Send any URL\nThe bot will automatically:\n\n\u2022 Fetch the webpage \n\u2022 Extract the article text \n\u2022 Summarize it \n\u2022 Send the summary back \n\n---\n\n\ud83d\udcc4 Ask anything about uploaded documents\nThe bot uses RAG + Gemini to answer strictly from your knowledge base.\n\n---\n\n\ud83d\udcc2 Auto-Ingestion\nThe bot ingests PDFs from Google Drive every 2 seconds and stores them in Supabase Vector DB.\n\n---\n\n\ud83d\udd01 Self-Correcting Answers\nEvery response is: \n1. Evaluated \n2. Improved \n3. Re-sent with higher accuracy \n\n---\n\n\ud83d\ude80 More features coming soon...",
"additionalFields": {
"appendAttribution": false
}
},
"type": "n8n-nodes-base.telegram",
"typeVersion": 1.2,
"position": [
672,
112
],
"id": "e58d095c-0dac-4f59-9a8a-6d75d228c0ee",
"name": "Send a text message1",
"credentials": {
"telegramApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"modelId": {
"__rl": true,
"value": "models/gemini-2.5-flash",
"mode": "list",
"cachedResultName": "models/gemini-2.5-flash"
},
"messages": {
"values": [
{
"content": "=Firstly check if {{ $('Switch').item.json.message.text }} contains one of the following /commands:\n/analyze, /summarize, /explain or /insights.\n\nalways mention the command only at the top of the message\n\nYou are an AI assistant that ONLY responds to the following slash commands. Do not engage in general conversation or answer questions outside of these commands.\n\n## Available Commands:\n\n**/summarize**\n- Fetch and read the content from the provided URL\n- Provide a concise summary of the main points (3-5 key takeaways)\n- Keep the summary brief and focused on essential information\n- Use bullet points for clarity\n\n**/analyze**\n- Fetch and read the content from the provided URL\n- Perform a detailed analysis of the content\n- Examine structure, arguments, data, methodology, or key themes\n- Identify strengths, weaknesses, biases, or patterns\n- Provide critical insights and evaluation\n\n**/explain**\n- Fetch and read the content from the provided URL\n- Break down complex concepts into simple, easy-to-understand language\n- Use analogies and examples where appropriate\n- Address technical terms and jargon in an educational manner\n\n**/insights**\n- Fetch and read the content from the provided URL\n- Extract actionable insights and implications\n- Identify trends, opportunities, or key takeaways\n- Focus on \"so what?\" - why this matters and what can be learned\n- Provide forward-looking perspectives\n\n## Critical Rules:\n- ONLY respond when one of the four commands above is used\n- If no valid command is detected, respond with: \"Please use one of the available commands: /summarize, /analyze, /explain, or /insights\"\n- Do not answer general questions or engage in conversation\n- Execute the command immediately without asking for confirmation\n{{ $json.content.parts[0].text }}\n"
}
]
},
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.googleGemini",
"typeVersion": 1,
"position": [
1488,
-320
],
"id": "ac99352f-d016-4139-9c23-bab4c3127479",
"name": "Command Performer",
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"content": "# Main Pipeline",
"height": 1056,
"width": 4096,
"color": 6
},
"type": "n8n-nodes-base.stickyNote",
"position": [
-144,
-416
],
"typeVersion": 1,
"id": "7042f327-8c9d-433c-be60-6ac8d92b0220",
"name": "Sticky Note"
}
],
"connections": {
"Google Gemini Chat Model": {
"ai_languageModel": [
[
{
"node": "AI Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Embeddings Google Gemini": {
"ai_embedding": [
[
{
"node": "RAG_Internal_Knowledge_Base",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Chat Memory Manager": {
"main": [
[
{
"node": "Edit Fields1",
"type": "main",
"index": 0
}
]
]
},
"RAG_Internal_Knowledge_Base": {
"ai_tool": [
[
{
"node": "AI Agent",
"type": "ai_tool",
"index": 0
}
]
]
},
"Simple Memory1": {
"ai_memory": [
[
{
"node": "Chat Memory Manager",
"type": "ai_memory",
"index": 0
}
]
]
},
"AI Agent": {
"main": [
[
{
"node": "Edit Fields3",
"type": "main",
"index": 0
}
]
]
},
"Telegram Trigger": {
"main": [
[
{
"node": "Switch",
"type": "main",
"index": 0
}
]
]
},
"If": {
"main": [
[
{
"node": "Improver Model",
"type": "main",
"index": 0
}
],
[
{
"node": "Edit Fields",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields1": {
"main": [
[
{
"node": "AI Agent",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields3": {
"main": [
[
{
"node": "Evaluator",
"type": "main",
"index": 0
}
]
]
},
"Evaluator": {
"main": [
[
{
"node": "If",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields": {
"main": [
[
{
"node": "Send a text message",
"type": "main",
"index": 0
}
]
]
},
"Improver Model": {
"main": [
[
{
"node": "Set Final Answer (Improved)",
"type": "main",
"index": 0
}
]
]
},
"Set Final Answer (Improved)": {
"main": [
[
{
"node": "Send a text message",
"type": "main",
"index": 0
}
]
]
},
"Schedule Trigger": {
"main": [
[
{
"node": "Search files and folders",
"type": "main",
"index": 0
}
]
]
},
"Search files and folders": {
"main": [
[
{
"node": "Download file",
"type": "main",
"index": 0
}
]
]
},
"Download file": {
"main": [
[
{
"node": "Supabase Vector Store",
"type": "main",
"index": 0
}
]
]
},
"Default Data Loader": {
"ai_document": [
[
{
"node": "Supabase Vector Store",
"type": "ai_document",
"index": 0
}
]
]
},
"Embeddings Google Gemini1": {
"ai_embedding": [
[
{
"node": "Supabase Vector Store",
"type": "ai_embedding",
"index": 0
}
]
]
},
"Recursive Character Text Splitter": {
"ai_textSplitter": [
[
{
"node": "Default Data Loader",
"type": "ai_textSplitter",
"index": 0
}
]
]
},
"Set Session ID": {
"main": [
[
{
"node": "Chat Memory Manager",
"type": "main",
"index": 0
}
]
]
},
"HTTP Request": {
"main": [
[
{
"node": "Extract Human Readable Text",
"type": "main",
"index": 0
}
]
]
},
"Extract Human Readable Text": {
"main": [
[
{
"node": "Command Performer",
"type": "main",
"index": 0
}
]
]
},
"Switch": {
"main": [
[
{
"node": "Edit Fields2",
"type": "main",
"index": 0
}
],
[
{
"node": "Set Session ID",
"type": "main",
"index": 0
}
],
[
{
"node": "Send a text message1",
"type": "main",
"index": 0
}
]
]
},
"Edit Fields2": {
"main": [
[
{
"node": "HTTP Request",
"type": "main",
"index": 0
}
]
]
},
"Command Performer": {
"main": [
[
{
"node": "Send Summarization Message",
"type": "main",
"index": 0
}
]
]
}
},
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "0c812053-ce54-4395-b70b-b38fe73295ba",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "IJV5OKbBk8x0lBwi",
"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.
googleDriveOAuth2ApigooglePalmApisupabaseApitelegramApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
How this works
This workflow enables you to build an intelligent Telegram bot that delivers context-aware responses by retrieving relevant information from your custom knowledge base, saving time on manual queries and enhancing user interactions with accurate, personalised answers. It's ideal for developers, customer support teams, or content creators who manage Telegram channels and need AI-driven assistance without coding from scratch. The core step involves the AI Agent using Google Gemini's chat model and embeddings to query a Supabase vector store, pulling precise data before crafting replies via Telegram.
Use this when you want an event-driven bot for real-time Q&A on specific topics, like product FAQs or internal docs, especially if you're already using Supabase for storage. Avoid it for high-volume traffic without scaling resources, or if you need multi-channel support beyond Telegram. Common variations include swapping Google Gemini for another LLM or adding webhooks for external data ingestion.
About this workflow
n8n telegram RAG. Uses lmChatGoogleGemini, embeddingsGoogleGemini, memoryManager, vectorStoreSupabase. Event-driven trigger; 32 nodes.
Source: https://github.com/Edublackk/self-correcting-rag-chatbot/blob/main/workflows/n8n_workflow.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.
Api Schema Extractor. Uses manualTrigger, httpRequest, splitOut, textSplitterRecursiveCharacterTextSplitter. Event-driven trigger; 88 nodes.
HTTP Request, Text Splitter Recursive Character Text Splitter, Document Default Data Loader +9
Wait Splitout. Uses manualTrigger, httpRequest, splitOut, textSplitterRecursiveCharacterTextSplitter. Event-driven trigger; 88 nodes.
HTTP Request, Text Splitter Recursive Character Text Splitter, Document Default Data Loader +9
This workflow automates the process of discovering and extracting APIs from various services, followed by generating custom schemas. It works in three distinct stages: research, extraction, and schema
HTTP Request, Text Splitter Recursive Character Text Splitter, Document Default Data Loader +9
A lightweight, self-hosted AI assistant built entirely in n8n. Multi-channel messaging (Telegram, WhatsApp, Gmail), persistent memory, task management, and autonomous work — all in a single visual wor
Telegram Trigger, OpenRouter Chat, Data Table +20
Alfred (funcional). Uses gmailTool, googleCalendarTool, gmail, embeddingsOpenAi. Event-driven trigger; 83 nodes.
Gmail Tool, Google Calendar Tool, Gmail +24