Jinash Rouniyar This workflow corresponds to n8n.io template #11619 — 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": "oSwN9ry7V8KSNbcr",
"meta": {
"templateCredsSetupCompleted": true
},
"name": "Contextual AI Agent Orchestration",
"tags": [],
"nodes": [
{
"id": "3fc78343-dd47-4b2a-926c-f366c47fd1cc",
"name": "Create Agent",
"type": "n8n-nodes-contextualai.contextualAi",
"position": [
288,
112
],
"parameters": {
"agentName": "={{ $('Submit Agent Information').item.json['Agent Name'] }}",
"datastoreName": "={{ $('Submit Agent Information').item.json['Datastore Name'] }}",
"agentDescription": "={{ $('Submit Agent Information').item.json['Agent Description'] }}",
"binaryPropertyName": "={{ $json.binaryKeys }}"
},
"credentials": {
"contextualAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "9afd4c74-47cc-4ecf-ab7b-ef7df2e3d88b",
"name": "Submit Agent Information",
"type": "n8n-nodes-base.formTrigger",
"position": [
-160,
112
],
"parameters": {
"options": {},
"formTitle": "Create your AI Agent",
"formFields": {
"values": [
{
"fieldLabel": "Agent Name",
"placeholder": "e.g. Customer Support Bot, Research Assistant, or Code Reviewer",
"requiredField": true
},
{
"fieldLabel": "Agent Description",
"placeholder": "e.g. An AI assistant that helps answer customer questions and summarize responses.",
"requiredField": true
},
{
"fieldLabel": "Datastore Name",
"placeholder": "e.g. support_knowledge_base, research_docs, or company_manuals",
"requiredField": true
},
{
"fieldType": "file",
"fieldLabel": "Files",
"requiredField": true,
"acceptFileTypes": ".pdf, .docx, .doc, .ppt"
}
]
},
"formDescription": "Configure your personalized AI agent by providing the essential details below. Once submitted, it will automatically create an agent for you"
},
"typeVersion": 2.3
},
{
"id": "f0cae1d6-6af5-4d06-90d4-e754e7bd4d22",
"name": "Preprocessing Step",
"type": "n8n-nodes-base.code",
"position": [
64,
112
],
"parameters": {
"jsCode": "const item = $input.first(); \nconst binaries = item.binary;\nconst keys = Object.keys(binaries);\n\nreturn [\n {\n json: {\n binaryKeys: keys.join(', ')\n },\n binary: binaries\n }\n];"
},
"typeVersion": 2
},
{
"id": "db104ab2-8ccb-4ab0-aca1-88ed528e539f",
"name": "Agent Orchestrator",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
192,
448
],
"parameters": {
"options": {
"systemMessage": "You can use two Contextual AI tools: list_agents (to get agent names and IDs) and query (to ask an agent).\nFor each user query:\n\nIf an agent name is given, call list_agents, find the matching agent ID, and use it.\n\nIf no name is given, call list_agents, pick the most relevant agent by intent.\n\nCall query with that agent ID and the user\u2019s question.\n\nReturn the response grounded in the response from query tool.\n"
}
},
"typeVersion": 2.2
},
{
"id": "58ad8395-3866-4cf2-a101-c4e6dd24186a",
"name": "Chat",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
-160,
448
],
"parameters": {
"options": {}
},
"typeVersion": 1.3
},
{
"id": "37877c08-6529-4b48-a039-b4ccc8128231",
"name": "Google Gemini Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatGoogleGemini",
"position": [
32,
672
],
"parameters": {
"options": {}
},
"credentials": {
"googlePalmApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "96aa07fe-7b74-41ea-af24-41c0aaeb11e1",
"name": "Simple Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
192,
672
],
"parameters": {
"contextWindowLength": 10
},
"typeVersion": 1.3
},
{
"id": "6da227ee-1f51-447a-b13e-23ae53cdbd5b",
"name": "Query Agent Tool",
"type": "n8n-nodes-contextualai.contextualAiTool",
"position": [
336,
672
],
"parameters": {
"query": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Query', ``, 'string') }}",
"agentId": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('Agent_ID', ``, 'string') }}",
"resource": "Query"
},
"credentials": {
"contextualAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "212f4d81-bc46-4ee3-afa9-debfb6dca484",
"name": "List Agents Tool",
"type": "n8n-nodes-contextualai.contextualAiTool",
"position": [
480,
672
],
"parameters": {
"operation": "List Agents"
},
"credentials": {
"contextualAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
},
{
"id": "b4e66a87-4f35-4b62-80d8-2973d478d5c8",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
-736,
-144
],
"parameters": {
"width": 480,
"height": 1312,
"content": "# RAG Multi-Agent Orchestration with Contextual AI Query Tool and Gemini 2.5 Flash\n\n## PROBLEM \nManaging multiple RAG AI agents can be complex when each has its own purpose and vector database. \n- Manually tracking agents and deciding which one to query wastes time. \n- LLMs often struggle to determine which agent best fits a user\u2019s request. \n\n### This workflow enables automated multi-agent orchestration, dynamically selecting and querying the correct agent using Contextual AI Query Tool and Gemini 2.5 Flash. \n\n## How it works \n- A **form trigger** allows users to create new agents by specifying a name, description, datastore, and uploading files. \n- A new agent is created with the provided information and files are ingested in the datastore\n- We get the status of file ingestion every 30 seconds until the ingestion process is complete\n- When users send queries, the **Agent Orchestrator** identifies the most relevant agent to generate grounded, context-aware responses. \n\n**Note:** The document ingestion process is asynchronous and may take a few minutes before your agent has the document fully available in the datastore for querying.\n\n## How to set up \n- Create a free [Contextual AI account](https://app.contextual.ai/) and obtain your `CONTEXTUALAI_API_KEY`. \n- Add `CONTEXTUALAI_API_KEY` as an **environment variable** in n8n. \n- For the baseline model, we have used Gemini 2.5 Flash Model, you can find your Gemini API key[ here](https://ai.google.dev/gemini-api/docs/api-key) \n\n\n## How to customize the workflow \n- Replace the **Form Trigger** with a **Webhook Trigger** or manual input to integrate with custom systems. \n- Swap **Gemini 2.5 Flash** with another LLM provider\n- Update the wait time as per user requirement\n- Modify the system prompt to fine-tune how the orchestration logic selects and queries agents. \n- You can check out this [Contextual AI API reference](https://docs.contextual.ai/api-reference/agents/create-agent) for more details on agent creation and usage. \n- If you have feedback or need support, please email **feedback@contextual.ai**.\n"
},
"typeVersion": 1
},
{
"id": "96446a81-d558-4261-8f00-28678d5baa20",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-224,
-144
],
"parameters": {
"color": 7,
"width": 1808,
"height": 448,
"content": "### 1: Form Submission \nUsers submit a form to create a new RAG agent by providing the **agent name**, **description**, **datastore name** and uploading relevant **documents**. Then, we iterate over each files to check their ingestion status. The workflow completes when both files are ingested and agent is ready for query.\n\n**Note:** The document ingestion process is asynchronous and may take a few minutes before your agent has the document fully available in the datastore for querying.\n"
},
"typeVersion": 1
},
{
"id": "96195bc9-9ede-45eb-8c26-60a8bf46cee3",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
-224,
320
],
"parameters": {
"color": 7,
"width": 992,
"height": 496,
"content": "### Step 2: Query and Retrieval \nUsers can enter a query, and the workflow automatically identifies the most relevant RAG agent based on the query and provides users with grounded answers.\n"
},
"typeVersion": 1
},
{
"id": "7a94b05a-46ff-4b54-be80-c8038ca7052e",
"name": "Split Out",
"type": "n8n-nodes-base.splitOut",
"position": [
512,
112
],
"parameters": {
"options": {},
"fieldToSplitOut": "uploaded"
},
"typeVersion": 1
},
{
"id": "1d2b3eab-8937-4f20-9619-994cfa711a59",
"name": "Wait",
"type": "n8n-nodes-base.wait",
"position": [
960,
48
],
"parameters": {
"amount": 30
},
"typeVersion": 1.1
},
{
"id": "a32883c4-bee5-4922-854c-b4a3ae25226d",
"name": "If",
"type": "n8n-nodes-base.if",
"position": [
1408,
112
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "e53d6b6e-da22-40c3-8445-339cd867e1e6",
"operator": {
"name": "filter.operator.equals",
"type": "string",
"operation": "equals"
},
"leftValue": "={{ $json.metadata.parseJson().status }}",
"rightValue": "completed"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "4c41e78c-2c4b-4bd7-94a3-278d4335f086",
"name": "Iterate over each files",
"type": "n8n-nodes-base.splitInBatches",
"position": [
736,
112
],
"parameters": {
"options": {}
},
"typeVersion": 3
},
{
"id": "b48205ed-227d-468f-88e6-e7da8230427a",
"name": "Get Document Ingestion Status",
"type": "n8n-nodes-contextualai.contextualAi",
"position": [
1184,
-32
],
"parameters": {
"resource": "Datastore",
"operation": "Get Document Metadata",
"documentId": "={{ $json.documentId }}",
"datastoreId": "={{ $('Create Agent').item.json.datastoreIds[0] }}"
},
"credentials": {
"contextualAiApi": {
"name": "<your credential>"
}
},
"typeVersion": 1
}
],
"active": false,
"settings": {
"executionOrder": "v1"
},
"versionId": "25d222d7-7c79-49a1-a1e0-431e7e0839c4",
"connections": {
"If": {
"main": [
[
{
"node": "Iterate over each files",
"type": "main",
"index": 0
}
],
[
{
"node": "Wait",
"type": "main",
"index": 0
}
]
]
},
"Chat": {
"main": [
[
{
"node": "Agent Orchestrator",
"type": "main",
"index": 0
}
]
]
},
"Wait": {
"main": [
[
{
"node": "Get Document Ingestion Status",
"type": "main",
"index": 0
}
]
]
},
"Split Out": {
"main": [
[
{
"node": "Iterate over each files",
"type": "main",
"index": 0
}
]
]
},
"Create Agent": {
"main": [
[
{
"node": "Split Out",
"type": "main",
"index": 0
}
]
]
},
"Simple Memory": {
"ai_memory": [
[
{
"node": "Agent Orchestrator",
"type": "ai_memory",
"index": 0
}
]
]
},
"List Agents Tool": {
"ai_tool": [
[
{
"node": "Agent Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Query Agent Tool": {
"ai_tool": [
[
{
"node": "Agent Orchestrator",
"type": "ai_tool",
"index": 0
}
]
]
},
"Preprocessing Step": {
"main": [
[
{
"node": "Create Agent",
"type": "main",
"index": 0
}
]
]
},
"Iterate over each files": {
"main": [
[],
[
{
"node": "Wait",
"type": "main",
"index": 0
}
]
]
},
"Google Gemini Chat Model": {
"ai_languageModel": [
[
{
"node": "Agent Orchestrator",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Submit Agent Information": {
"main": [
[
{
"node": "Preprocessing Step",
"type": "main",
"index": 0
}
]
]
},
"Get Document Ingestion Status": {
"main": [
[
{
"node": "If",
"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.
contextualAiApigooglePalmApi
Pro
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
Managing multiple RAG AI agents can be complex when each has its own purpose and vector database. Manually tracking agents and deciding which one to query wastes time. LLMs often struggle to determine which agent best fits a user’s request. A form trigger allows users to create…
Source: https://n8n.io/workflows/11619/ — 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 workflow implements a Retrieval-Augmented Generation (RAG) system using Google Gemini's File Search API. It allows users to upload files to a dedicated search store and then ask questions about t
Form Trigger, Chat Trigger, Google Gemini Chat +4
The Recap AI - Whats App Chatbot Agent. Uses formTrigger, agent, lmChatGoogleGemini, toolThink. Event-driven trigger; 19 nodes.
Form Trigger, Agent, Google Gemini Chat +7
How it Works
Memory Buffer Window, Agent, Output Parser Structured +9
1_4_Core_Node. Uses stopAndError, compression, xml, crypto. Event-driven trigger; 63 nodes.
Stop And Error, Compression, XML +7
Lection 9 main. Uses formTrigger, chatTrigger, agent, lmChatOpenAi. Event-driven trigger; 55 nodes.
Form Trigger, Chat Trigger, Agent +7