This workflow follows the Chainllm → Google Drive 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": "AgenteIA",
"nodes": [
{
"parameters": {
"promptType": "define",
"text": "=Eres un asistente de conversaci\u00f3n experto y debes mantener el contexto de la conversaci\u00f3n previa.\n\nREGLAS DE FORMATO CR\u00cdTICAS:\n1. Tu respuesta debe estar SIEMPRE en formato JSON ESTRICTO.\n2. Todas las CLAVES (\"reply\", \"videos\", etc.) DEBEN usar comillas dobles.\n3. NO incluyas bloques de c\u00f3digo Markdown (ej., ```json).\n4. SOLO devuelve el objeto JSON, sin texto adicional.\n\nHISTORIAL COMPLETO DE CONVERSACI\u00d3N (\u00dasalo como contexto):\n{{ JSON.stringify($json.body.history) }}\n\nInstrucci\u00f3n: Utiliza el historial de conversaci\u00f3n anterior para responder a la \u00faltima pregunta planteada en ese historial. Aseg\u00farate de que el array 'videos' contenga contenido relevante si aplica, o un array vac\u00edo si no lo es.",
"batching": {}
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.7,
"position": [
-48,
112
],
"id": "72a1ad3d-5e24-4923-b3a2-2f02b8c50074",
"name": "Basic LLM Chain"
},
{
"parameters": {
"options": {}
},
"type": "@n8n/n8n-nodes-langchain.lmChatDeepSeek",
"typeVersion": 1,
"position": [
-48,
-336
],
"id": "57100a22-65fb-43c6-bfa8-4cb34dd8b186",
"name": "DeepSeek Chat Model",
"credentials": {
"deepSeekApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"respondWith": "json",
"responseBody": "={{ $json }}",
"options": {}
},
"type": "n8n-nodes-base.respondToWebhook",
"typeVersion": 1.4,
"position": [
1312,
112
],
"id": "4e72873f-743c-48ec-8c12-a31779675c39",
"name": "Respond to Webhook"
},
{
"parameters": {
"rules": {
"values": [
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"leftValue": "={{ $json.body.history[$json.body.history.length - 1].content }}",
"rightValue": "formato",
"operator": {
"type": "string",
"operation": "contains"
},
"id": "a753bbb0-9315-45f3-af38-5ec7de9fb390"
}
],
"combinator": "and"
}
},
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "108fdc4e-501a-4473-a3ce-1a1bf06ed4aa",
"leftValue": "={{ $json.body.history[$json.body.history.length - 1].content }}",
"rightValue": "genera",
"operator": {
"type": "string",
"operation": "contains"
}
}
],
"combinator": "and"
}
},
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "28ad394f-5ff0-4ce3-8946-19b160ab8856",
"leftValue": "={{ $json.body.history[$json.body.history.length - 1].content }}",
"rightValue": "investigaci\u00f3n",
"operator": {
"type": "string",
"operation": "contains"
}
}
],
"combinator": "and"
}
},
{
"conditions": {
"options": {
"caseSensitive": true,
"leftValue": "",
"typeValidation": "strict",
"version": 2
},
"conditions": [
{
"id": "69e63266-51d2-4add-90ef-493bf1f9e4ee",
"leftValue": "={{ $json.body.history[$json.body.history.length - 1].content }}",
"rightValue": "",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
}
}
],
"combinator": "and"
}
}
]
},
"options": {}
},
"type": "n8n-nodes-base.switch",
"typeVersion": 3.3,
"position": [
-400,
-560
],
"id": "672daf75-cb82-4684-bf32-c6f686f4c974",
"name": "Switch"
},
{
"parameters": {
"promptType": "define",
"text": "=Eres un asistente de investigaci\u00f3n experto. Tu \u00fanica tarea es analizar la solicitud del usuario y generar un documento de investigaci\u00f3n completo y detallado.\n\nREGLAS DE FORMATO CR\u00cdTICAS:\n1. NO incluyas ninguna explicaci\u00f3n, saludo o texto antes del JSON.\n2. SIEMPRE devuelve la respuesta en formato JSON estricto.\n3. NO uses bloques de c\u00f3digo Markdown (```json) ni saltos de l\u00ednea fuera del campo \"content\".\n4. El JSON debe tener EXACTAMENTE DOS campos: \"title\" (t\u00edtulo del documento) y \"content\" (el cuerpo del documento).\n5. REGLA CR\u00cdTICA DE ESTRUCTURA: El campo \"content\" debe estar dividido en las siguientes secciones obligatorias, utilizando t\u00edtulos en negrita y doble salto de l\u00ednea para separarlos:\n - **Introducci\u00f3n**\n - **Desarrollo del Tema**\n - **Conclusiones**\n - **Referencias/Bibliograf\u00eda**\n\nESTRUCTURA DE RESPUESTA REQUERIDA (JSON):\n{\n\u00a0 \u00a0 \"title\": \"[Genera un t\u00edtulo formal y conciso basado en la solicitud]\",\n\u00a0 \u00a0 \"content\": \"[Genera el documento completo. Utiliza saltos de l\u00ednea (\\n\\n) para separar p\u00e1rrafos y doble asterisco (**) para encabezados de secci\u00f3n y negritas. Aseg\u00farate de incluir y formatear las secciones Introducci\u00f3n, Desarrollo, Conclusiones y Bibliograf\u00eda.]\"\n}\n\nTema de la investigaci\u00f3n: {{ $json.final_prompt }}",
"batching": {}
},
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"typeVersion": 1.7,
"position": [
-48,
-544
],
"id": "968cb3ed-6927-4974-bba7-98334f168faf",
"name": "GenerarInvestigacion",
"alwaysOutputData": false
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "98ed43f3-8588-49dc-8df4-98722d72ac44",
"name": "plantilla_caratula",
"value": "UNIVERSIDAD SAN PABLO DE GUATEMALA Facultad de Ciencias Empresariales Escuela de Ingenier\u00eda en Ciencias y Sistemas de la Computaci\u00f3n. [IMAGEN DEL ESCUDO] <<TITULO_INVESTIGACION>> en el curso de Sistemas Expertos Impartido por el ingeniero Marco Alfredo Orozco de Paz ALUMNO Henry Morales --- [INICIO_CONTENIDO]",
"type": "string"
}
]
},
"includeOtherFields": true,
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
-640,
-160
],
"id": "5a5e7a88-7c58-4590-b5cb-253fe30f5f90",
"name": "generaFormato"
},
{
"parameters": {
"httpMethod": "POST",
"path": "/chat",
"responseMode": "responseNode",
"options": {}
},
"type": "n8n-nodes-base.webhook",
"typeVersion": 2.1,
"position": [
-880,
-160
],
"id": "f16ef7cc-b159-409e-afab-f83d578f47e8",
"name": "Webhook"
},
{
"parameters": {
"mode": "combine",
"combineBy": "combineByPosition",
"options": {}
},
"type": "n8n-nodes-base.merge",
"typeVersion": 3.2,
"position": [
240,
-176
],
"id": "0205045f-4665-4824-b738-befb9e4d86d5",
"name": "Merge"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "f2472787-50c4-44f2-8eb7-0b381fab47bb",
"name": "reply",
"value": "={{ JSON.parse($json.text).reply }}",
"type": "string"
},
{
"id": "d80c0ec1-2187-4401-95f2-ace756bd5872",
"name": "videos",
"value": "={{ JSON.parse($json.text).videos }}",
"type": "string"
}
]
},
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
320,
112
],
"id": "69d5e8ab-9f7c-4c5e-b822-95ecb493cbac",
"name": "FormatoChat"
},
{
"parameters": {
"jsCode": "for (const item of items) {\n // 1. Obtener la \u00faltima solicitud del historial\n const history = item.json.body.history;\n const fullPrompt = history[history.length - 1].content;\n \n // 2. Usar RegEx para encontrar el texto exacto entre comillas despu\u00e9s de \"genera investigaci\u00f3n sobre \"\n const match = fullPrompt.match(/genera investigaci\u00f3n sobre \"(.*)\"/i);\n \n // 3. Si encuentra la coincidencia, usa solo el tema (match[1]). Si no, usa el prompt completo.\n const final_prompt = match && match[1] ? match[1] : fullPrompt;\n\n // 4. Agregar el campo de salida limpio al \u00edtem\n item.json.final_prompt = final_prompt;\n}\n\nreturn items;"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
-208,
-544
],
"id": "e7a6b2b7-fe7a-42a0-8251-eeaaa89a2fba",
"name": "Code in JavaScript"
},
{
"parameters": {
"operation": "copy",
"fileId": {
"__rl": true,
"value": "1pMtYHJZyLGWgS9BoOdSLMEu0cO37tcKrxx1XkEgpsjw",
"mode": "id"
},
"name": "={{ $json.llm_data.title }}",
"options": {
"copyRequiresWriterPermission": false,
"description": ""
}
},
"type": "n8n-nodes-base.googleDrive",
"typeVersion": 3,
"position": [
624,
-176
],
"id": "a693ac5d-380e-494f-9709-14d8b9274710",
"name": "Copy file",
"credentials": {
"googleDriveOAuth2Api": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"authentication": "serviceAccount",
"operation": "update",
"documentURL": "={{ $json.id }}",
"simple": false,
"actionsUi": {
"actionFields": [
{
"action": "replaceAll",
"text": "{{titulo}}",
"replaceText": "={{ $json.llm_data.title }}"
},
{
"action": "replaceAll",
"text": "{{introduccion}}",
"replaceText": "={{ $json.llm_data.introduccion }}"
},
{
"action": "replaceAll",
"text": "{{desarrollo}}",
"replaceText": "={{ $json.llm_data.desarrollo }}"
},
{
"action": "replaceAll",
"text": "{{conclusiones}}",
"replaceText": "={{ $json.llm_data.conclusiones }}"
},
{
"action": "replaceAll",
"text": "{{referencias}}",
"replaceText": "={{ $json.llm_data.referencias }}"
}
]
}
},
"type": "n8n-nodes-base.googleDocs",
"typeVersion": 2,
"position": [
1104,
-352
],
"id": "4b6afaf0-39c1-4bc5-897b-ff35ca742d8b",
"name": "Update a document",
"credentials": {
"googleApi": {
"name": "<your credential>"
}
}
},
{
"parameters": {
"jsCode": "for (const item of $input.all()) {\n try {\n // 1. PASO DE PARSEO y CLAVE\n const jsonString = item.json.text;\n item.json.llm_data = JSON.parse(jsonString);\n item.json.mergeKey = 1; \n \n // 2. PASO DE SEGMENTACI\u00d3N (\u00a1SOLO CONTENIDO!)\n const content = item.json.llm_data.content;\n\n // A. Extraer Introducci\u00f3n (Contenido puro entre los encabezados)\n const introMatch = content.match(/\\*\\*Introducci\u00f3n\\*\\*\\s*\\n*(.*?)\\n*\\*\\*Desarrollo del Tema\\*\\*/s);\n let introduccion = introMatch && introMatch[1] ? introMatch[1].trim() : '';\n // YA NO A\u00d1ADE EL T\u00cdTULO \"Introducci\u00f3n\\n\\n\"\n\n // B. Extraer Desarrollo del Tema\n const desarrolloMatch = content.match(/\\*\\*Desarrollo del Tema\\*\\*\\s*\\n*(.*?)\\n*\\*\\*Conclusiones\\*\\*/s);\n let desarrollo = desarrolloMatch && desarrolloMatch[1] ? desarrolloMatch[1].trim() : '';\n // YA NO A\u00d1ADE EL T\u00cdTULO\n\n // C. Extraer Conclusiones\n const concMatch = content.match(/\\*\\*Conclusiones\\*\\*\\s*\\n*(.*?)\\n*\\*\\*Referencias\\/Bibliograf\u00eda\\*\\*/s);\n let conclusiones = concMatch && concMatch[1] ? concMatch[1].trim() : '';\n // YA NO A\u00d1ADE EL T\u00cdTULO\n\n // D. Extraer Referencias/Bibliograf\u00eda\n const refMatch = content.match(/\\*\\*Referencias\\/Bibliograf\u00eda\\*\\*\\s*\\n*(.*)$/s);\n let referencias = refMatch && refMatch[1] ? refMatch[1].trim() : '';\n // YA NO A\u00d1ADE EL T\u00cdTULO\n\n // 3. Crear el nuevo objeto llm_data con campos segmentados\n item.json.llm_data = {\n ...item.json.llm_data,\n introduccion: introduccion,\n desarrollo: desarrollo,\n conclusiones: conclusiones,\n referencias: referencias,\n };\n \n // Eliminamos el campo 'content' original\n delete item.json.llm_data.content;\n\n } catch (error) {\n console.error('Error al analizar o segmentar el JSON:', error);\n }\n}\nreturn $input.all();"
},
"type": "n8n-nodes-base.code",
"typeVersion": 2,
"position": [
400,
-176
],
"id": "3fc2d705-c78a-4661-9555-2f2744c6550c",
"name": "Code in JavaScript1"
},
{
"parameters": {
"mode": "combine",
"fieldsToMatchString": "mergeKey",
"options": {}
},
"type": "n8n-nodes-base.merge",
"typeVersion": 3.2,
"position": [
928,
-352
],
"id": "0fffd1ff-0a65-4a45-a6aa-f26b13ca7aa8",
"name": "Merge1"
},
{
"parameters": {
"assignments": {
"assignments": [
{
"id": "491d2d3e-6e94-4669-b72c-9855bd3ad6c8",
"name": "mergeKey",
"value": 1,
"type": "number"
}
]
},
"includeOtherFields": true,
"options": {}
},
"type": "n8n-nodes-base.set",
"typeVersion": 3.4,
"position": [
784,
-176
],
"id": "c65412dc-e24d-4247-9606-3e1a1a95264b",
"name": "SetMergeKey"
}
],
"connections": {
"DeepSeek Chat Model": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain",
"type": "ai_languageModel",
"index": 0
},
{
"node": "GenerarInvestigacion",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Basic LLM Chain": {
"main": [
[
{
"node": "FormatoChat",
"type": "main",
"index": 0
}
]
]
},
"Switch": {
"main": [
[
{
"node": "Code in JavaScript",
"type": "main",
"index": 0
}
],
[
{
"node": "Code in JavaScript",
"type": "main",
"index": 0
}
],
[
{
"node": "Code in JavaScript",
"type": "main",
"index": 0
}
],
[
{
"node": "Basic LLM Chain",
"type": "main",
"index": 0
}
]
]
},
"GenerarInvestigacion": {
"main": [
[
{
"node": "Merge",
"type": "main",
"index": 0
}
]
]
},
"generaFormato": {
"main": [
[
{
"node": "Switch",
"type": "main",
"index": 0
},
{
"node": "Merge",
"type": "main",
"index": 1
}
]
]
},
"Webhook": {
"main": [
[
{
"node": "generaFormato",
"type": "main",
"index": 0
}
]
]
},
"Merge": {
"main": [
[
{
"node": "Code in JavaScript1",
"type": "main",
"index": 0
}
]
]
},
"FormatoChat": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
},
"Code in JavaScript": {
"main": [
[
{
"node": "GenerarInvestigacion",
"type": "main",
"index": 0
}
]
]
},
"Copy file": {
"main": [
[
{
"node": "SetMergeKey",
"type": "main",
"index": 0
}
]
]
},
"Update a document": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
},
"Code in JavaScript1": {
"main": [
[
{
"node": "Copy file",
"type": "main",
"index": 0
},
{
"node": "Merge1",
"type": "main",
"index": 0
}
]
]
},
"Merge1": {
"main": [
[
{
"node": "Update a document",
"type": "main",
"index": 0
}
]
]
},
"SetMergeKey": {
"main": [
[
{
"node": "Merge1",
"type": "main",
"index": 1
}
]
]
}
},
"active": true,
"settings": {
"executionOrder": "v1"
},
"versionId": "3521d091-97b2-4dab-8c65-02afebdf85ea",
"meta": {
"templateCredsSetupCompleted": true
},
"id": "sTYK22OQWGCd5Fxx",
"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.
deepSeekApigoogleApigoogleDriveOAuth2Api
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
AgenteIA. Uses chainLlm, lmChatDeepSeek, googleDrive, googleDocs. Webhook trigger; 15 nodes.
Source: https://github.com/bautista3383/Examen-Final-Sistemas-Expertos---Agentes-de-IA/blob/332a2f6de30f8f7a247947f3d84b5752414e6e93/AgenteIA.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.
Awesome N8N Templates. Uses lmChatDeepSeek, splitOut, removeDuplicates, outputParserAutofixing. Webhook trigger; 26 nodes.
Lm Chat Deep Seek, Output Parser Autofixing, Output Parser Structured +6
L&D_AgentsAI_ATIVO. Uses httpRequest, agent, googleCalendarTool, toolSerpApi. Webhook trigger; 93 nodes.
HTTP Request, Agent, Google Calendar Tool +9
ANIS_HUB 1. Uses gmail, googleDrive, googleSheets, httpRequest. Webhook trigger; 89 nodes.
Gmail, Google Drive, Google Sheets +3
The best content automation template in the market is now even better—with “deep research” on time-sensitive topics\! Unlike most n8n content automation templates that are mainly for “demo purposes,”
OpenAI, HTTP Request, XML +11
What if AI didn't just write content—but actually thought about how to write it? This n8n workflow revolutionizes content creation by deploying multiple specialized AI agents that handle every aspect
Tool Http Request, Anthropic Chat, Airtable +7