존재하는 지원 포털을 활용한 IT 지원 어시스턴트 채팅 로봇 구축
고급
이것은Support분야의자동화 워크플로우로, 16개의 노드를 포함합니다.주로 If, Set, SplitOut, Aggregate, HttpRequest 등의 노드를 사용하며. 존재하는 지원 포털을 활용한 IT 지원 어시스턴트 채팅 로봇 구축
사전 요구사항
- •대상 API의 인증 정보가 필요할 수 있음
- •OpenAI API Key
사용된 노드 (16)
카테고리
워크플로우 미리보기
노드 연결 관계를 시각적으로 표시하며, 확대/축소 및 이동을 지원합니다
워크플로우 내보내기
다음 JSON 구성을 복사하여 n8n에 가져오면 이 워크플로우를 사용할 수 있습니다
{
"meta": {
"instanceId": "408f9fb9940c3cb18ffdef0e0150fe342d6e655c3a9fac21f0f644e8bedabcd9",
"templateCredsSetupCompleted": true
},
"nodes": [
{
"id": "8f203423-b063-4918-a6ec-dad3ac7d1a20",
"name": "채팅 메시지 수신 시",
"type": "@n8n/n8n-nodes-langchain.chatTrigger",
"position": [
860,
-100
],
"webhookId": "c82193c7-163c-4556-942f-81c80037e0ea",
"parameters": {
"options": {}
},
"typeVersion": 1.1
},
{
"id": "d9f2e90f-128b-458b-b3cf-79db2ec08633",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1000,
100
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1.2
},
{
"id": "4f752502-8589-4e31-bbe1-4b8395e7325a",
"name": "Simple Memory",
"type": "@n8n/n8n-nodes-langchain.memoryBufferWindow",
"position": [
1160,
100
],
"parameters": {},
"typeVersion": 1.3
},
{
"id": "61ca5a4b-3661-4330-ac4c-e09e75dd764c",
"name": "Acuity Support Search API",
"type": "n8n-nodes-base.httpRequest",
"position": [
1840,
80
],
"parameters": {
"url": "https://2al21hjwoz-dsn.algolia.net/1/indexes/*/queries?x-algolia-agent=Algolia%20for%20JavaScript%20(3.35.1)%3B%20Browser%20(lite)%3B%20instantsearch.js%201.12.1%3B%20Zendesk%20Integration%20(2.32.0)%3B%20JS%20Helper%20(2.28.1)&x-algolia-application-id=2AL21HJWOZ&x-algolia-api-key=c3c07dd7fb575008575163c085a62b92",
"method": "POST",
"options": {},
"jsonBody": "={{\n{\n \"requests\":[\n {\n \"indexName\":\"Zendesk 4-25\",\n \"params\": \"query=\" + $json.query + \"&hitsPerPage=5&page=0&facets=%5B%22locale.locale%22%2C%22label_names%22%2C%22category.title%22%5D&tagFilters=&facetFilters=%5B%22locale.locale%3Aen-us%22%5D\"\n }\n ]\n}\n}}",
"sendBody": true,
"sendHeaders": true,
"specifyBody": "json",
"headerParameters": {
"parameters": [
{
"name": "Accept-Language",
"value": "en"
},
{
"name": "Cache-Control",
"value": "no-cache"
},
{
"name": "Connection",
"value": "keep-alive"
},
{
"name": "Origin",
"value": "https://help.acuityscheduling.com"
},
{
"name": "Referer",
"value": "https://help.acuityscheduling.com/"
},
{
"name": "User-Agent",
"value": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/134.0.0.0 Safari/537.36"
},
{
"name": "accept",
"value": "application/json"
}
]
}
},
"typeVersion": 4.2
},
{
"id": "8ecd6287-982c-4754-9300-4c6d54202273",
"name": "관련 필드 추출",
"type": "n8n-nodes-base.set",
"position": [
2560,
80
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "a6973f14-e17d-46b0-9c5b-c6d9967dbf99",
"name": "title",
"type": "string",
"value": "={{ $json.title }}"
},
{
"id": "88092adb-7f63-4daa-8c7a-cbd85750e180",
"name": "body",
"type": "string",
"value": "={{ $json.body_safe }}"
},
{
"id": "12718897-a73d-4c3a-bcfb-b17c890458ec",
"name": "url",
"type": "string",
"value": "=https://help.acuityscheduling.com/hc/en-us/articles/{{ $json.id }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "bf5855b2-8e73-4c29-b277-adee63e8bf59",
"name": "결과를 항목으로 변환",
"type": "n8n-nodes-base.splitOut",
"position": [
2360,
80
],
"parameters": {
"options": {},
"fieldToSplitOut": "results[0].hits"
},
"typeVersion": 1
},
{
"id": "c9329816-bbe0-4de7-b6fb-fa87783f6a5c",
"name": "결과가 있습니까?",
"type": "n8n-nodes-base.if",
"position": [
2040,
80
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "f5d7e890-f00a-4252-8588-c6662e71790c",
"operator": {
"type": "array",
"operation": "lengthGt",
"rightType": "number"
},
"leftValue": "={{ $json.results[0]?.hits ?? [] }}",
"rightValue": 0
}
]
}
},
"typeVersion": 2.2
},
{
"id": "860a178a-d500-4291-acfc-9c9f4638d6c7",
"name": "빈 응답",
"type": "n8n-nodes-base.set",
"position": [
2360,
260
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "0ce36950-83d9-4964-8763-f329a4cda5a8",
"name": "response",
"type": "array",
"value": "[]"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "c9f2a08b-88c2-4287-994c-f7af58e98301",
"name": "응답 통합",
"type": "n8n-nodes-base.aggregate",
"position": [
2760,
80
],
"parameters": {
"options": {},
"aggregate": "aggregateAllItemData",
"destinationFieldName": "response"
},
"typeVersion": 1
},
{
"id": "5f1f8874-7022-4ea1-b0a7-de42c4f800a1",
"name": "Knowledgebase Tool",
"type": "@n8n/n8n-nodes-langchain.toolWorkflow",
"position": [
1320,
100
],
"parameters": {
"name": "acuity_support_search",
"workflowId": {
"__rl": true,
"mode": "id",
"value": "={{ $workflow.id }}"
},
"description": "Call this tool to query AcuityScheduling's Support Center Search API.",
"workflowInputs": {
"value": {
"query": "={{ /*n8n-auto-generated-fromAI-override*/ $fromAI('query', ``, 'string') }}"
},
"schema": [
{
"id": "query",
"type": "string",
"display": true,
"removed": false,
"required": false,
"displayName": "query",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "defineBelow",
"matchingColumns": [],
"attemptToConvertTypes": false,
"convertFieldsToString": false
}
},
"typeVersion": 2.1
},
{
"id": "3913ddaa-852e-4463-a072-fe8be22bc184",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
720,
-300
],
"parameters": {
"color": 7,
"width": 780,
"height": 580,
"content": "## 1. Simple Chatbot with Knowledgebase Tool\n[Learn more about AI agents](https://docs.n8n.io/integrations/builtin/cluster-nodes/root-nodes/n8n-nodes-langchain.agent)\n\nThe AI agent node is the simplest and recommended way to create user-friendly chatbots in n8n. Here, we'll define a support agent which can answer AcuityScheduling.com questions. To ensure the answers are accurate and up-to-date, we'll connect it to the support knowledgebase via a custom workflow tool."
},
"typeVersion": 1
},
{
"id": "e24d75f9-6d3c-4bca-b67f-33737ee969ee",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
1540,
-140
],
"parameters": {
"color": 7,
"width": 700,
"height": 440,
"content": "## 2. Use your Existing Help Portal Search\n[Read more about the HTTP request tool](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.httprequest)\n\nThe concept of RAG need to be synonymous with vector stores! In truth, many companies with a decent enough support website are able to leverage this existing knowledgebase for support agents. This saves time, money and effort and additional avoids maintenance of a vector store where syncs and updates are common."
},
"typeVersion": 1
},
{
"id": "f5feebf1-fd6d-4558-a868-7ea4f852386c",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
2260,
-140
],
"parameters": {
"color": 7,
"width": 720,
"height": 600,
"content": "## 3. Clean up the Results to Optimise Tokens\n[Read more about the aggregate node](https://docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.aggregate)\n\nOf course, the results are intended for the website format but by using the custom workflow tool, we can edit it down to suit our chat scenario and save LLM costs (in terms of tokens) whilst we're at it. "
},
"typeVersion": 1
},
{
"id": "8132de59-9b47-460a-9cb9-f2ec83123a3f",
"name": "AcuityScheduling Support Chatbot",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1060,
-100
],
"parameters": {
"options": {
"systemMessage": "You are a support assistant for the SaaS company, AcuityScheduling.com. Your task is to openly help the user with any questions regarding the AcuityScheduling service however, you are restricted to only this service. If the user asks questions unrelated to AcuityScheduling, you may ask them for clarification, explain you are not able to help them out of scope or redirect them to support@acuityScheduling.com. Be factual in your answer, tap into the resources or tools available and do not rely on your training data (which might be out-of-date). When returning a response to the user, you are encouraged to share the URL of the knowledgebase page where the user can explore the documentation for themselves."
}
},
"typeVersion": 1.8
},
{
"id": "564bde38-25ea-4969-aa3f-bff66ec2782f",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
260,
-840
],
"parameters": {
"width": 440,
"height": 1120,
"content": "## Try it Out!\n### This n8n template demonstrates how you can leverage existing support site search to power your Support Chatbots and agents.\n\nBuilding a support chatbot need not be complicated! If building and indexing vector stores or duplicating data isn't necessarily your thing, an alternative implementation of the [RAG](https://www.databricks.com/glossary/retrieval-augmented-generation-rag) approach is to leverage existing knowledge-bases such as support portals.\n\n### How it works\n* A simple AI agent is connected with chat trigger to receive user queries.\n* The AI agent is instructed to fetch information from the knowledge-base via the attached custom workflow tool (aka \"knowledgebase tool\").\n* There is no step to replicate the entire support articles database into a vector store. You may choose not too because of time, cost and maintainence involved.\n* Instead, the tool leverages the existing support portal's search API to retrieve knowledge-base articles.\n* Finally, the search results are formatted before sending an aggregated response back to the agent.\n\n### How to use?\n* Customise the subworkflow to work with your own support portal API and format accordingly.\n* Try the following queries\n * How do I connect my icloud to acuityScheduling?\n * How do I download past invoices for my Acuity account?\n\n### Requirements\n* OpenAI for LLM.\n* If your organisation's APIs require authorisation, you may need to add custom credentials as necessary.\n\n### Customising this workflow\n* Add additional tools to reach other parts of your internal knowledgebase.\n* Not using OpenAI? Feel free to swap but ensure the LLM has tools/function calling support.\n\n\n### Need Help?\nJoin the [Discord](https://discord.com/invite/XPKeKXeB7d) or ask in the [Forum](https://community.n8n.io/)!\n\nHappy Hacking!"
},
"typeVersion": 1
},
{
"id": "a918718f-915d-4d5c-a7c2-a015b8a84bbb",
"name": "KnowledgeBase Tool Subworkflow",
"type": "n8n-nodes-base.executeWorkflowTrigger",
"position": [
1620,
80
],
"parameters": {
"workflowInputs": {
"values": [
{
"name": "query"
}
]
}
},
"typeVersion": 1.1
}
],
"pinData": {},
"connections": {
"c9329816-bbe0-4de7-b6fb-fa87783f6a5c": {
"main": [
[
{
"node": "bf5855b2-8e73-4c29-b277-adee63e8bf59",
"type": "main",
"index": 0
}
],
[
{
"node": "860a178a-d500-4291-acfc-9c9f4638d6c7",
"type": "main",
"index": 0
}
]
]
},
"4f752502-8589-4e31-bbe1-4b8395e7325a": {
"ai_memory": [
[
{
"node": "8132de59-9b47-460a-9cb9-f2ec83123a3f",
"type": "ai_memory",
"index": 0
}
]
]
},
"bf5855b2-8e73-4c29-b277-adee63e8bf59": {
"main": [
[
{
"node": "8ecd6287-982c-4754-9300-4c6d54202273",
"type": "main",
"index": 0
}
]
]
},
"d9f2e90f-128b-458b-b3cf-79db2ec08633": {
"ai_languageModel": [
[
{
"node": "8132de59-9b47-460a-9cb9-f2ec83123a3f",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"5f1f8874-7022-4ea1-b0a7-de42c4f800a1": {
"ai_tool": [
[
{
"node": "8132de59-9b47-460a-9cb9-f2ec83123a3f",
"type": "ai_tool",
"index": 0
}
]
]
},
"8ecd6287-982c-4754-9300-4c6d54202273": {
"main": [
[
{
"node": "c9f2a08b-88c2-4287-994c-f7af58e98301",
"type": "main",
"index": 0
}
]
]
},
"61ca5a4b-3661-4330-ac4c-e09e75dd764c": {
"main": [
[
{
"node": "c9329816-bbe0-4de7-b6fb-fa87783f6a5c",
"type": "main",
"index": 0
}
]
]
},
"8f203423-b063-4918-a6ec-dad3ac7d1a20": {
"main": [
[
{
"node": "8132de59-9b47-460a-9cb9-f2ec83123a3f",
"type": "main",
"index": 0
}
]
]
},
"a918718f-915d-4d5c-a7c2-a015b8a84bbb": {
"main": [
[
{
"node": "61ca5a4b-3661-4330-ac4c-e09e75dd764c",
"type": "main",
"index": 0
}
]
]
}
}
}자주 묻는 질문
이 워크플로우를 어떻게 사용하나요?
위의 JSON 구성 코드를 복사하여 n8n 인스턴스에서 새 워크플로우를 생성하고 "JSON에서 가져오기"를 선택한 후, 구성을 붙여넣고 필요에 따라 인증 설정을 수정하세요.
이 워크플로우는 어떤 시나리오에 적합한가요?
고급 - 지원
유료인가요?
이 워크플로우는 완전히 무료이며 직접 가져와 사용할 수 있습니다. 다만, 워크플로우에서 사용하는 타사 서비스(예: OpenAI API)는 사용자 직접 비용을 지불해야 할 수 있습니다.
관련 워크플로우 추천
빌드 커스텀 n8n 워크플로우 MCP 서버
구축할 수 있는 개별 n8n 워크플로우 MCP 서버
If
N8n
Set
+
If
N8n
Set
46 노드Jimleuk
기타
Qdrant, Mistral.ai, OpenAI를 사용하여 세금 법률 어시스턴트를 구축합니다.
Qdrant, Mistral.ai, OpenAI를 사용하여 세금 법률 어시스턴트를 구축합니다.
Set
Wait
Filter
+
Set
Wait
Filter
38 노드Jimleuk
금융
OpenAI 및 Redis 세션 관리를 사용한 채팅 지원 고객 인증
OpenAI와 Redis 세션 관리를 사용한 채팅 지원 고객 인증
Set
Form
Redis
+
Set
Form
Redis
19 노드Jimleuk
지원
시각화 참조 라이브러리에서 n8n 노드를 탐색
可视化 참조 라이브러리에서 n8n 노드를 탐색
If
Ftp
Set
+
If
Ftp
Set
113 노드I versus AI
기타
컨텍스트 혼합 RAG AI 콘텐츠
Google Drive에서 Supabase 상황 벡터 데이터베이스로 동기화, RAG 애플리케이션 사용
If
Set
Code
+
If
Set
Code
76 노드Michael Taleb
AI RAG
AI 어시스턴트와 Airtable 대화 및 데이터 분석
AI스마트어시스턴트与Airtable对话及데이터분석
If
Set
Merge
+
If
Set
Merge
41 노드Mark Shcherbakov
엔지니어링
워크플로우 정보
난이도
고급
노드 수16
카테고리1
노드 유형12
저자
Jimleuk
@jimleukFreelance consultant based in the UK specialising in AI-powered automations. I work with select clients tackling their most challenging projects. For business enquiries, send me an email at hello@jimle.uk LinkedIn: https://www.linkedin.com/in/jimleuk/ X/Twitter: https://x.com/jimle_uk
외부 링크
n8n.io에서 보기 →
이 워크플로우 공유