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 →
{
"name": "KI-Agent Grundlagen (Lern-Workflow)",
"nodes": [
{
"parameters": {
"content": "## KI-Agent in n8n\n\nEin **KI-Agent** bekommt eine Nachricht (vom **Chat Trigger**), denkt mit einem **Sprachmodell** (LLM), merkt sich den Verlauf \u00fcber **Memory** und kann **Tools** benutzen (rechnen, im Web suchen, APIs aufrufen \u2026).\n\n\u26a0\ufe0f Die farbigen Linien **unten** (`ai_languageModel`, `ai_memory`, `ai_tool`) sind **keine** normalen Datenfl\u00fcsse, sondern **F\u00e4higkeiten**, die du an den Agent ansteckst.\n\n\u2139\ufe0f Das Sprachmodell braucht eine **Anthropic-Credential** (deinen API-Key).\n\n\u25b6\ufe0f **Testen:** \u00fcber den **Chat** starten \u2014 Sub-Nodes (Modell/Memory/Tool) laufen nur \u00fcber den Agent, nicht einzeln per \u201eTest step\".",
"height": 280,
"width": 560,
"color": 4
},
"id": "a1000000-0000-4000-8000-000000000001",
"name": "Doku: \u00dcberblick",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
-80,
-330
]
},
{
"parameters": {
"content": "## Chat Trigger\n\u00d6ffnet ein Chat-Fenster und startet den Agent bei **jeder Nachricht**. (Eine von vielen Trigger-Arten \u2014 siehe `n8n-grundlagen.json`.)",
"height": 120,
"width": 220,
"color": 5
},
"id": "a1000000-0000-4000-8000-000000000002",
"name": "Doku: Trigger",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
20,
-160
]
},
{
"parameters": {
"content": "## KI-Agent\nDas \u201eGehirn\": versteht die Anfrage, entscheidet ob ein **Tool** n\u00f6tig ist, und formuliert die Antwort.",
"height": 120,
"width": 240,
"color": 6
},
"id": "a1000000-0000-4000-8000-000000000003",
"name": "Doku: Agent",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
300,
-160
]
},
{
"parameters": {
"content": "**Sprachmodell (Claude)**\nDas eigentliche LLM \u2014 das \u201eDenken\". Braucht eine **Anthropic-Credential**.",
"height": 100,
"width": 200,
"color": 7
},
"id": "a1000000-0000-4000-8000-000000000004",
"name": "Doku: Modell",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
120,
150
]
},
{
"parameters": {
"content": "**Memory**\nMerkt sich den Gespr\u00e4chsverlauf, damit der Agent Kontext hat.",
"height": 100,
"width": 200,
"color": 7
},
"id": "a1000000-0000-4000-8000-000000000005",
"name": "Doku: Memory",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
340,
150
]
},
{
"parameters": {
"content": "**Tool (Rechner)**\nEine F\u00e4higkeit f\u00fcr den Agent. Es gibt viele Tools (HTTP, Suche, eigene Workflows \u2026). **Fast jeder Node kann ein Tool sein.**",
"height": 100,
"width": 220,
"color": 7
},
"id": "a1000000-0000-4000-8000-000000000006",
"name": "Doku: Tool",
"type": "n8n-nodes-base.stickyNote",
"typeVersion": 1,
"position": [
560,
150
]
},
{
"parameters": {
"options": {}
},
"id": "a1000000-0000-4000-8000-0000000000b1",
"name": "Chat-Nachricht (Trigger)",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"typeVersion": 1.4,
"position": [
80,
0
]
},
{
"parameters": {
"promptType": "auto",
"options": {
"systemMessage": "Du bist ein hilfreicher Assistent f\u00fcr ein Hackathon-Team. Antworte freundlich und kurz auf Deutsch. F\u00fcr Rechenaufgaben benutze den Rechner."
}
},
"id": "a1000000-0000-4000-8000-0000000000b2",
"name": "KI-Agent",
"type": "@n8n/n8n-nodes-langchain.agent",
"typeVersion": 3.1,
"position": [
360,
0
]
},
{
"parameters": {
"options": {}
},
"id": "a1000000-0000-4000-8000-0000000000b3",
"name": "Sprachmodell (Claude)",
"type": "@n8n/n8n-nodes-langchain.lmChatAnthropic",
"typeVersion": 1.5,
"position": [
180,
280
]
},
{
"parameters": {},
"id": "a1000000-0000-4000-8000-0000000000b4",
"name": "Ged\u00e4chtnis (Memory)",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"typeVersion": 1.4,
"position": [
400,
280
]
},
{
"parameters": {},
"id": "a1000000-0000-4000-8000-0000000000b5",
"name": "Rechner",
"type": "@n8n/n8n-nodes-langchain.toolCalculator",
"typeVersion": 1,
"position": [
620,
280
]
}
],
"connections": {
"Chat-Nachricht (Trigger)": {
"main": [
[
{
"node": "KI-Agent",
"type": "main",
"index": 0
}
]
]
},
"Sprachmodell (Claude)": {
"ai_languageModel": [
[
{
"node": "KI-Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Ged\u00e4chtnis (Memory)": {
"ai_memory": [
[
{
"node": "KI-Agent",
"type": "ai_memory",
"index": 0
}
]
]
},
"Rechner": {
"ai_tool": [
[
{
"node": "KI-Agent",
"type": "ai_tool",
"index": 0
}
]
]
}
},
"settings": {
"executionOrder": "v1"
}
}
For the full experience including quality scoring and batch install features for each workflow upgrade to Pro
About this workflow
KI-Agent Grundlagen (Lern-Workflow). Uses chatTrigger, agent, lmChatAnthropic, memoryBufferWindow. Chat trigger; 11 nodes.
Source: https://github.com/freddy-schuetz/hackathon-n8n-starter/blob/c389e97515f8fee487299b0daed1f885f049178b/examples/workflows/ai-agent-grundlagen.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.
✨ Intro This workflow shows how to go beyond a “plain” AI chatbot by:
This workflow dynamically chooses between two new powerful Anthropic models — Claude Opus 4 and Claude Sonnet 4 — to handle user queries, based on their complexity and nature, maintaining scalability
Think Tool. Uses stickyNote, agent, googleCalendarTool, memoryBufferWindow. Chat trigger; 28 nodes.
ModelRouter. Uses chatTrigger, agent, modelSelector, httpRequest. Chat trigger; 28 nodes.
This workflow introduces a conversational AI Cost Estimation Chatbot with built-in OCR document analysis and interactive form guidance. It helps users and teams handle pricing, measurement, and produc