在n8n中使用任意LangChain模块(通过LangChain代码节点)
中级
这是一个AI领域的自动化工作流,包含 6 个节点。主要使用 Set, ManualTrigger, Code, LmChatOpenAi 等节点,结合人工智能技术实现智能自动化。 在n8n中使用任意LangChain模块(通过LangChain代码节点)
前置要求
- •OpenAI API Key
分类
工作流预览
可视化展示节点连接关系,支持缩放和平移
导出工作流
复制以下 JSON 配置到 n8n 导入,即可使用此工作流
{
"meta": {
"instanceId": "408f9fb9940c3cb18ffdef0e0150fe342d6e655c3a9fac21f0f644e8bedabcd9",
"templateCredsSetupCompleted": true
},
"nodes": [
{
"id": "b7e2de27-e52c-46d1-aaa9-a67c11c48a8f",
"name": "## 试试看!",
"type": "n8n-nodes-base.stickyNote",
"position": [
-420,
-60
],
"parameters": {
"width": 328.41313484548044,
"height": 211.30313955500145,
"content": "执行前请将 `YOUR_API_KEY` 替换为 searchapi.io 的 API 密钥"
},
"typeVersion": 1
},
{
"id": "fd2ac655-73fd-434a-bba4-e460af8dfa8a",
"name": "当点击\"执行工作流\"时",
"type": "n8n-nodes-base.manualTrigger",
"position": [
-820,
20
],
"parameters": {},
"typeVersion": 1
},
{
"id": "e1bd87f7-283b-496d-910d-b92d1cb19237",
"name": "便签",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1140,
-20
],
"parameters": {
"color": 7,
"height": 220.82906011310624,
"content": "## 关于"
},
"typeVersion": 1
},
{
"id": "a43bb1c5-dd90-4331-930c-128ef0ecb38a",
"name": "LangChain 代码",
"type": "@n8n/n8n-nodes-langchain.code",
"position": [
-380,
20
],
"parameters": {
"code": {
"execute": {
"code": "// IMPORTANT: add in your API key for searchapi.io below\nconst searchApiKey = \"<YOUR API KEY>\"\n\nconst { loadSummarizationChain } = require(\"langchain/chains\");\nconst { SearchApiLoader } = require(\"@n8n/n8n-nodes-langchain/node_modules/@langchain/community/document_loaders/web/searchapi.cjs\");\nconst { PromptTemplate } = require(\"@langchain/core/prompts\");\nconst { TokenTextSplitter } = require(\"langchain/text_splitter\");\nconst loader = new SearchApiLoader({\n engine: \"youtube_transcripts\",\n video_id: $input.item.json.videoId,\n apiKey: searchApiKey,\n});\n\nif (searchApiKey == \"<YOUR API KEY>\") {\n throw new Error(\"Please add your API key for searchapi.io to this node\")\n}\n\nconst docs = await loader.load();\n\nconst splitter = new TokenTextSplitter({\n chunkSize: 10000,\n chunkOverlap: 250,\n});\n\nconst docsSummary = await splitter.splitDocuments(docs);\n\nconst llmSummary = await this.getInputConnectionData('ai_languageModel', 0);\n\nconst summaryTemplate = `\nYou are an expert in summarizing YouTube videos.\nYour goal is to create a summary of a podcast.\nBelow you find the transcript of a podcast:\n--------\n{text}\n--------\n\nThe transcript of the podcast will also be used as the basis for a question and answer bot.\nProvide some examples questions and answers that could be asked about the podcast. Make these questions very specific.\n\nTotal output will be a summary of the video and a list of example questions the user could ask of the video.\n\nSUMMARY AND QUESTIONS:\n`;\n\nconst SUMMARY_PROMPT = PromptTemplate.fromTemplate(summaryTemplate);\n\nconst summaryRefineTemplate = `\nYou are an expert in summarizing YouTube videos.\nYour goal is to create a summary of a podcast.\nWe have provided an existing summary up to a certain point: {existing_answer}\n\nBelow you find the transcript of a podcast:\n--------\n{text}\n--------\n\nGiven the new context, refine the summary and example questions.\nThe transcript of the podcast will also be used as the basis for a question and answer bot.\nProvide some examples questions and answers that could be asked about the podcast. Make\nthese questions very specific.\nIf the context isn't useful, return the original summary and questions.\nTotal output will be a summary of the video and a list of example questions the user could ask of the video.\n\nSUMMARY AND QUESTIONS:\n`;\n\nconst SUMMARY_REFINE_PROMPT = PromptTemplate.fromTemplate(\n summaryRefineTemplate\n);\n\nconst summarizeChain = loadSummarizationChain(llmSummary, {\n type: \"refine\",\n verbose: true,\n questionPrompt: SUMMARY_PROMPT,\n refinePrompt: SUMMARY_REFINE_PROMPT,\n});\n\nconst summary = await summarizeChain.run(docsSummary);\n\nreturn [{json: { summary } } ];"
}
},
"inputs": {
"input": [
{
"type": "main",
"required": true,
"maxConnections": 1
},
{
"type": "ai_languageModel",
"required": true,
"maxConnections": 1
}
]
},
"outputs": {
"output": [
{
"type": "main"
}
]
}
},
"typeVersion": 1
},
{
"id": "a36440c5-402e-44e6-819c-2a19dc9e3e1e",
"name": "设置 YouTube 视频 ID",
"type": "n8n-nodes-base.set",
"position": [
-600,
20
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "c2dc2944-a7c7-44c3-a805-27a55baa452a",
"name": "videoId",
"type": "string",
"value": "OsMVtuuwOXc"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "02386530-9aef-4732-9972-5624b78431a6",
"name": "OpenAI 聊天模型",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
-340,
220
],
"parameters": {
"model": {
"__rl": true,
"mode": "list",
"value": "gpt-4o-mini"
},
"options": {}
},
"credentials": {
"openAiApi": {
"id": "8gccIjcuf3gvaoEr",
"name": "OpenAi account"
}
},
"typeVersion": 1.2
}
],
"pinData": {},
"connections": {
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "LangChain Code",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Set YouTube video ID": {
"main": [
[
{
"node": "LangChain Code",
"type": "main",
"index": 0
}
]
]
},
"When clicking \"Execute Workflow\"": {
"main": [
[
{
"node": "Set YouTube video ID",
"type": "main",
"index": 0
}
]
]
}
}
}常见问题
如何使用这个工作流?
复制上方的 JSON 配置代码,在您的 n8n 实例中创建新工作流并选择「从 JSON 导入」,粘贴配置后根据需要修改凭证设置即可。
这个工作流适合什么场景?
中级 - 人工智能
需要付费吗?
本工作流完全免费,您可以直接导入使用。但请注意,工作流中使用的第三方服务(如 OpenAI API)可能需要您自行付费。
相关工作流推荐
评估指标示例:RAG文档相关性
评估指标示例:RAG文档相关性
Set
Evaluation
Google Sheets
+13
26 节点David Roberts
工程
使用AI翻译音频
使用AI翻译音频
Set
Http Request
Manual Trigger
+3
12 节点David Roberts
人工智能
使用AI与Google表格对话
使用AI与Google表格对话
Set
Code
Filter
+8
19 节点David Roberts
人工智能
评估指标示例:分类
评估指标示例:分类
Set
Webhook
Evaluation
+6
13 节点David Roberts
工程
评估指标示例:检查工具是否被调用
评估指标示例:检查工具是否被调用
Set
Evaluation
Agent
+7
15 节点David Roberts
工程
评估指标示例:正确性(由AI判断)
评估指标示例:正确性(由AI判断)
Set
Evaluation
Agent
+6
13 节点David Roberts
工程