🔍🛠️ Perplexity研究员转HTML网页
高级
这是一个AI、Marketing领域的自动化工作流，包含 47 个节点。主要使用 If、Set、Webhook、Telegram、HttpRequest 等节点，结合人工智能技术实现智能自动化。 🔍 Perplexity研究转HTML：AI驱动的内容创作
前置要求
- •HTTP Webhook 端点（n8n 会自动生成）
- •Telegram Bot Token
- •可能需要目标 API 的认证凭证
- •OpenAI API Key
使用的节点 (47 个)
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
{
"id": "HnqGW0eq5asKfZxf",
"meta": {
"instanceId": "03907a25f048377a8789a4332f28148522ba31ee907fababf704f1d88130b1b6",
"templateCredsSetupCompleted": true
},
"name": "🔍🛠️Perplexity 研究员转 HTML 网页",
"tags": [],
"nodes": [
{
"id": "ad5d96c6-941a-4ab3-b349-10bae99e5988",
"name": "便签",
"type": "n8n-nodes-base.stickyNote",
"position": [
320,
1360
],
"parameters": {
"color": 3,
"width": 625.851492623043,
"height": 465.2493344282225,
"content": "## 基于 Perplexity 研究创建文章"
},
"typeVersion": 1
},
{
"id": "19b3ca66-5fd2-4d04-b25a-a17fb38642f8",
"name": "便签1",
"type": "n8n-nodes-base.stickyNote",
"position": [
1240,
1360
],
"parameters": {
"color": 4,
"width": 479.02028317328745,
"height": 464.14912719677955,
"content": "## 将文章转换为 HTML"
},
"typeVersion": 1
},
{
"id": "7fad54e8-5a50-42da-b38d-08f6912615ab",
"name": "gpt-4o-mini",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1380,
1660
],
"parameters": {
"model": "gpt-4o-mini-2024-07-18",
"options": {
"responseFormat": "text"
}
},
"credentials": {
"openAiApi": {
"id": "h597GY4ZJQD47RQd",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "5291869f-3ac6-4ce2-88f3-b572924b6082",
"name": "gpt-4o-mini1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1560,
1040
],
"parameters": {
"options": {
"topP": 1,
"timeout": 60000,
"maxTokens": -1,
"maxRetries": 2,
"temperature": 0,
"responseFormat": "text",
"presencePenalty": 0,
"frequencyPenalty": 0
}
},
"credentials": {
"openAiApi": {
"id": "h597GY4ZJQD47RQd",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "a232f6ca-ad4c-40fa-a641-f0dd83c8f18a",
"name": "结构化输出解析器1",
"type": "@n8n/n8n-nodes-langchain.outputParserStructured",
"position": [
640,
1660
],
"parameters": {
"schemaType": "manual",
"inputSchema": "{\n \"type\": \"object\",\n \"properties\": {\n \"article\": {\n \"type\": \"object\",\n \"required\": [\"category\", \"title\", \"metadata\", \"content\", \"hashtags\"],\n \"properties\": {\n \"category\": {\n \"type\": \"string\",\n \"description\": \"Article category\"\n },\n \"title\": {\n \"type\": \"string\",\n \"description\": \"Article title\"\n },\n \"metadata\": {\n \"type\": \"object\",\n \"properties\": {\n \"timePosted\": {\n \"type\": \"string\",\n \"description\": \"Time since article was posted\"\n },\n \"author\": {\n \"type\": \"string\",\n \"description\": \"Article author name\"\n },\n \"tag\": {\n \"type\": \"string\",\n \"description\": \"Article primary tag\"\n }\n },\n \"required\": [\"timePosted\", \"author\", \"tag\"]\n },\n \"content\": {\n \"type\": \"object\",\n \"properties\": {\n \"mainText\": {\n \"type\": \"string\",\n \"description\": \"Main article content\"\n },\n \"sections\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"object\",\n \"properties\": {\n \"title\": {\n \"type\": \"string\",\n \"description\": \"Section title\"\n },\n \"text\": {\n \"type\": \"string\",\n \"description\": \"Section content\"\n },\n \"quote\": {\n \"type\": \"string\",\n \"description\": \"Blockquote text\"\n }\n },\n \"required\": [\"title\", \"text\", \"quote\"]\n }\n }\n },\n \"required\": [\"mainText\", \"sections\"]\n },\n \"hashtags\": {\n \"type\": \"array\",\n \"items\": {\n \"type\": \"string\"\n },\n \"description\": \"Article hashtags\"\n }\n }\n }\n }\n}"
},
"typeVersion": 1.2
},
{
"id": "e7d1adac-88aa-4f76-92bf-bbac3aa6386a",
"name": "gpt-4o-mini2",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
420,
1660
],
"parameters": {
"options": {
"topP": 1,
"timeout": 60000,
"maxTokens": -1,
"maxRetries": 2,
"temperature": 0,
"responseFormat": "json_object",
"presencePenalty": 0,
"frequencyPenalty": 0
}
},
"credentials": {
"openAiApi": {
"id": "h597GY4ZJQD47RQd",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "156e51db-03f7-4099-afe8-6f0361c5b497",
"name": "Webhook",
"type": "n8n-nodes-base.webhook",
"position": [
160,
860
],
"webhookId": "6a8e3ae7-02ae-4663-a27a-07df448550ab",
"parameters": {
"path": "pblog",
"options": {},
"responseMode": "responseNode"
},
"typeVersion": 2
},
{
"id": "6dd3eba7-e779-4e4a-960e-c5a7b6b3a929",
"name": "响应 Webhook",
"type": "n8n-nodes-base.respondToWebhook",
"position": [
2820,
1480
],
"parameters": {
"options": {},
"respondWith": "text",
"responseBody": "={{ $json.text }}"
},
"typeVersion": 1.1
},
{
"id": "27ee681e-4259-4323-b4fe-629f99cb33d0",
"name": "Telegram",
"type": "n8n-nodes-base.telegram",
"position": [
2320,
880
],
"parameters": {
"text": "={{ $('Perplexity Topic Agent').item.json.output.slice(0, 300) }}",
"chatId": "={{ $json.telegram_chat_id }}",
"additionalFields": {
"parse_mode": "HTML",
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "BIE64nzfpGeesXUn",
"name": "Telegram account"
}
},
"typeVersion": 1.2
},
{
"id": "f437d40c-2bf6-43e2-b77b-e5c2cdc35055",
"name": "gpt-4o-mini5",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
2480,
1660
],
"parameters": {
"options": {
"topP": 1,
"timeout": 60000,
"maxTokens": -1,
"maxRetries": 2,
"temperature": 0,
"responseFormat": "text",
"presencePenalty": 0,
"frequencyPenalty": 0
}
},
"credentials": {
"openAiApi": {
"id": "h597GY4ZJQD47RQd",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "275bce4a-4252-41d4-bcba-174f0c51bf4a",
"name": "基础LLM链",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
2340,
1480
],
"parameters": {
"text": "=Create a modern, responsive single-line HTML document. Convert any markdown to Tailwind CSS classes. Replace markdown lists with proper HTML list elements. Remove all newline characters while preserving </br> tags in content. Enhance the layout with Tailwind CSS cards where appropriate. Use the following base structure, but improve the styling and responsiveness:\n\n<!DOCTYPE html>\n<html lang=\"en\">\n\n<head>\n <meta charset=\"UTF-8\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1.0\">\n <title>Comprehensive Overview of DeepSeek V3</title>\n <link href=\"https://cdn.jsdelivr.net/npm/tailwindcss@2.2.19/dist/tailwind.min.css\" rel=\"stylesheet\">\n</head>\n\n<body class=\"bg-gray-100 font-sans\">\n <div class=\"relative p-4\">\n <div class=\"max-w-3xl mx-auto text-sm\">\n <div class=\"mt-3 bg-white rounded-lg shadow-lg flex flex-col justify-between leading-normal\">\n <div class=\"p-6\">\n <h1 class=\"text-gray-900 font-bold text-4xl mb-4\">Comprehensive Overview of DeepSeek V3</h1>\n <div class=\"mb-4\">\n <p class=\"leading-8\"><strong>Time Posted:</strong> Just now</p>\n <p class=\"leading-8\"><strong>Author:</strong> AI Research Team</p>\n <p class=\"leading-8\"><strong>Tag:</strong> AI Models</p>\n </div>\n <p class=\"leading-8 my-4\"><strong>DeepSeek V3</strong> is a state-of-the-art AI model that leverages\n advanced architectures and techniques to deliver high performance across various applications.\n This overview covers its key concepts, practical applications, advantages, limitations, and best\n practices for implementation.</p>\n <section class=\"mb-6\">\n <h2 class=\"text-2xl font-bold my-3\">Key Concepts and Core Components</h2>\n <p class=\"leading-8 my-3\"><strong>1. Mixture-of-Experts (MoE) Architecture:</strong> DeepSeek V3\n employs a Mixture-of-Experts (MoE) architecture, which consists of multiple neural networks,\n each optimized for different tasks. This architecture allows for efficient processing by\n activating only a portion of the network for each task, reducing hardware costs.</p>\n <p class=\"leading-8 my-3\"><strong>2. Parameters:</strong> The model boasts a total of 671\n billion\n parameters, with 37 billion active parameters for each token during processing. The addition\n of\n the Multi-Token Prediction (MTP) module increases the total parameters to 685 billion,\n making it\n significantly larger than other models like Meta's Llama 3.1 (405B).</p>\n <p class=\"leading-8 my-3\"><strong>3. Multi-head Latent Attention (MLA):</strong> DeepSeek V3\n uses\n Multi-head Latent Attention (MLA) to extract key details from text multiple times, improving\n its\n accuracy.</p>\n <p class=\"leading-8 my-3\"><strong>4. Multi-Token Prediction (MTP):</strong> The model utilizes\n Multi-Token Prediction to generate several tokens at once, speeding up inference and\n enabling\n speculative decoding.</p>\n <blockquote\n class=\"italic leading-8 my-3 p-5 text-indigo-600 font-semibold bg-indigo-50 rounded-lg border-l-4 border-indigo-600\">\n DeepSeek V3 employs a Mixture-of-Experts architecture for efficient processing.</blockquote>\n </section>\n <section class=\"mb-6\">\n <h2 class=\"text-2xl font-bold my-3\">Practical Applications</h2>\n <ol class=\"list-decimal pl-5\">\n <li class=\"leading-8 my-3\"><strong>Translation, Coding, and Content Generation:</strong>\n DeepSeek V3 is designed for a wide range of tasks including translation, coding, content\n generation, and reasoning. It excels in English, Chinese, coding, and mathematics,\n rivaling leading commercial models like OpenAI's GPT-4.</li>\n <li class=\"leading-8 my-3\"><strong>Research and Development:</strong> The open-source nature\n of DeepSeek V3 fuels innovation, allowing researchers to experiment with and build upon\n its technology.</li>\n <li class=\"leading-8 my-3\"><strong>Commercial Applications:</strong> The licensing of\n DeepSeek V3 makes it permissible for commercial use, opening it up to numerous\n applications across different industries.</li>\n <li class=\"leading-8 my-3\"><strong>Democratization of AI:</strong> By making powerful AI\n accessible, DeepSeek V3 levels the playing field, allowing smaller organizations to\n compete with larger ones.</li>\n </ol>\n <blockquote\n class=\"italic leading-8 my-3 p-5 text-indigo-600 font-semibold bg-indigo-50 rounded-lg border-l-4 border-indigo-600\">\n DeepSeek V3 democratizes AI access for smaller organizations.</blockquote>\n </section>\n <section class=\"mb-6\">\n <h2 class=\"text-2xl font-bold my-3\">Advantages</h2>\n <ol class=\"list-decimal pl-5\">\n <li class=\"leading-8 my-3\"><strong>Speed and Efficiency:</strong> DeepSeek V3 processes\n information at a blistering 60 tokens per second, a threefold increase over its\n predecessor. It uses advanced inference capabilities, deploying 32 H800 GPUs for prefill\n and 320 H800 GPUs for decoding.</li>\n <li class=\"leading-8 my-3\"><strong>Cost-Effectiveness:</strong> The model was trained for a\n mere $5.5 million, a fraction of the estimated over $100 million invested by OpenAI in\n GPT-4. DeepSeek V3 offers significantly lower prices for its online services, with 1\n million tokens priced at just $1.1, currently offered at a promotional rate of $0.28.\n </li>\n <li class=\"leading-8 my-3\"><strong>Innovation in Inference:</strong> The model's advanced\n inference capabilities set the standard for future model deployment, making it a\n powerful tool in the digital realm.</li>\n </ol>\n <blockquote\n class=\"italic leading-8 my-3 p-5 text-indigo-600 font-semibold bg-indigo-50 rounded-lg border-l-4 border-indigo-600\">\n DeepSeek V3 processes information at 60 tokens per second.</blockquote>\n </section>\n <section class=\"mb-6\">\n <h2 class=\"text-2xl font-bold my-3\">Limitations</h2>\n <ol class=\"list-decimal pl-5\">\n <li class=\"leading-8 my-3\"><strong>Deployment Complexity:</strong> Deploying DeepSeek V3\n requires advanced hardware and a deployment strategy that separates the prefilling and\n decoding stages, which might be unachievable for small companies due to a lack of\n resources. The recommended deployment unit for DeepSeek V3 is relatively large, posing a\n burden for small-sized teams.</li>\n <li class=\"leading-8 my-3\"><strong>Potential for Further Enhancement:</strong> Although\n DeepSeek V3 has achieved an end-to-end generation speed of more than two times that of\n DeepSeek V2, there still remains potential for further enhancement with the development\n of more advanced hardware.</li>\n </ol>\n <blockquote\n class=\"italic leading-8 my-3 p-5 text-indigo-600 font-semibold bg-indigo-50 rounded-lg border-l-4 border-indigo-600\">\n Deployment of DeepSeek V3 may be complex for small companies.</blockquote>\n </section>\n <section class=\"mb-6\">\n <h2 class=\"text-2xl font-bold my-3\">Best Practices for Implementation</h2>\n <ol class=\"list-decimal pl-5\">\n <li class=\"leading-8 my-3\"><strong>Hardware Requirements:</strong> Ensure that the\n deployment environment has the necessary advanced hardware to handle the model's\n requirements, including multiple GPUs for prefill and decoding.</li>\n <li class=\"leading-8 my-3\"><strong>Deployment Strategy:</strong> Implement a deployment\n strategy that separates the prefilling and decoding stages to optimize performance and\n efficiency.</li>\n <li class=\"leading-8 my-3\"><strong>Monitoring and Optimization:</strong> Continuously\n monitor the model's performance and optimize it as needed to address any limitations and\n improve efficiency.</li>\n <li class=\"leading-8 my-3\"><strong>Community Engagement:</strong> Engage with the\n open-source community to leverage the collective knowledge and resources available,\n which can help in addressing any challenges and improving the model further.</li>\n </ol>\n <blockquote\n class=\"italic leading-8 my-3 p-5 text-indigo-600 font-semibold bg-indigo-50 rounded-lg border-l-4 border-indigo-600\">\n Engage with the open-source community for better implementation.</blockquote>\n </section>\n <p class=\"leading-8 my-6\"><strong>Hashtags:</strong> <span\n class=\"text-indigo-600\">#DeepSeekV3</span> <span class=\"text-indigo-600\">#AI</span> <span\n class=\"text-indigo-600\">#MachineLearning</span> <span\n class=\"text-indigo-600\">#OpenSource</span></p>\n </div>\n </div>\n </div>\n </div>\n</body>\n\n</html>\n\n-------\n\nRequirements:\n- Output must be a single line of HTML\n- Enhanced with modern Tailwind CSS styling\n- Proper HTML list structures\n- Responsive design\n- No newlines except </br> in content\n- No markdown formatting\n- Clean, readable layout\n- Properly formatted hashtags\n- No explanation or additional text in output\n- No code block markers or escape characters\n- Wnsure Metadata, Title and Content are included in HTML\n\nMetadata: {{ $('Article').item.json.article.metadata.toJsonString() }}\nTitle: {{ $json.title }}\nContent: {{ $json.html }}\n",
"promptType": "define"
},
"typeVersion": 1.4
},
{
"id": "cddd9324-8471-4dcb-a46b-836015db9833",
"name": "无操作1",
"type": "n8n-nodes-base.noOp",
"position": [
560,
1080
],
"parameters": {},
"typeVersion": 1
},
{
"id": "432a0ae9-451a-4830-b065-8b0593de92ea",
"name": "gpt-4o-mini3",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
1020,
1040
],
"parameters": {
"options": {
"topP": 1,
"timeout": 60000,
"maxTokens": -1,
"maxRetries": 2,
"temperature": 0,
"responseFormat": "text",
"presencePenalty": 0,
"frequencyPenalty": 0
}
},
"credentials": {
"openAiApi": {
"id": "h597GY4ZJQD47RQd",
"name": "OpenAi account"
}
},
"typeVersion": 1
},
{
"id": "55e00886-b6c1-4f7a-81ae-e8e0d4102cab",
"name": "便签4",
"type": "n8n-nodes-base.stickyNote",
"position": [
2200,
1360
],
"parameters": {
"color": 6,
"width": 531,
"height": 465,
"content": "## 使用 TailwindCSS 样式创建 HTML 页面"
},
"typeVersion": 1
},
{
"id": "1ed7f754-1279-4511-a085-6ed4e4c36de1",
"name": "便签2",
"type": "n8n-nodes-base.stickyNote",
"position": [
320,
760
],
"parameters": {
"width": 450.54438902818094,
"height": 489.5271576259337,
"content": "## 从 Get 请求解析主题"
},
"typeVersion": 1
},
{
"id": "e9dcb568-7f8d-40c5-94cb-6f25386436cf",
"name": "便签5",
"type": "n8n-nodes-base.stickyNote",
"position": [
820,
760
],
"parameters": {
"color": 5,
"width": 380,
"height": 488,
"content": "## 改进用户主题"
},
"typeVersion": 1
},
{
"id": "a7fdaddb-d6fc-4d45-85cc-a372cfb90327",
"name": "如果2",
"type": "n8n-nodes-base.if",
"position": [
2120,
1140
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "8e35de0a-ac16-4555-94f4-24e97bdf4b33",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "{{ $json.output }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "57d056b8-7e91-41e4-8b74-dce15847a09b",
"name": "提示词",
"type": "n8n-nodes-base.set",
"position": [
1300,
2080
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "efbe7563-8502-407e-bfa0-a4a26d8cddd4",
"name": "user",
"type": "string",
"value": "={{ $('Execute Workflow Trigger').item.json.topic }}"
},
{
"id": "05e0b629-bb9f-4010-96a8-10872764705a",
"name": "system",
"type": "string",
"value": "Assistant is a large language model. Assistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand. Assistant is constantly learning and improving, and its capabilities are constantly evolving. It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, Assistant is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics. Overall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist. "
}
]
}
},
"typeVersion": 3.4
},
{
"id": "8209cece-fde4-485f-81a1-2d24a6eac474",
"name": "执行工作流触发器",
"type": "n8n-nodes-base.executeWorkflowTrigger",
"position": [
420,
2180
],
"parameters": {},
"typeVersion": 1
},
{
"id": "445e4d15-c2b0-4152-a0f8-d6b93ad5bae6",
"name": "Telegram2",
"type": "n8n-nodes-base.telegram",
"position": [
860,
2180
],
"parameters": {
"text": "=<i>{{ $('Execute Workflow Trigger').item.json.topic }}</i>",
"chatId": "={{ $json.telegram_chat_id }}",
"additionalFields": {
"parse_mode": "HTML",
"appendAttribution": false
}
},
"credentials": {
"telegramApi": {
"id": "BIE64nzfpGeesXUn",
"name": "Telegram account"
}
},
"typeVersion": 1.2
},
{
"id": "57a5b3ce-5490-4d50-91cc-c36e508eee4d",
"name": "如果",
"type": "n8n-nodes-base.if",
"position": [
1080,
2180
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "7e2679dc-c898-415d-a693-c2c1e7259b6a",
"operator": {
"type": "string",
"operation": "notContains"
},
"leftValue": "={{ $('Execute Workflow Trigger').item.json.topic }}",
"rightValue": "undefined"
}
]
}
},
"typeVersion": 2.2
},
{
"id": "fdf827dc-96b1-4ed3-895b-2a0f5f4c41a3",
"name": "无操作，不执行任何操作",
"type": "n8n-nodes-base.noOp",
"position": [
1300,
2300
],
"parameters": {},
"typeVersion": 1
},
{
"id": "944aa564-f449-47a6-9d9c-c20a48946ab6",
"name": "便签6",
"type": "n8n-nodes-base.stickyNote",
"position": [
320,
1940
],
"parameters": {
"color": 5,
"width": 1614,
"height": 623,
"content": "## 🛠️perplexity_research_tool"
},
"typeVersion": 1
},
{
"id": "3806c079-8c08-48b7-a3ed-a26f6d86c67f",
"name": "Perplexity 主题代理",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1580,
860
],
"parameters": {
"text": "=Topic: {{ $json.text }}",
"options": {
"systemMessage": "Use the perplexity_research_tool to provide research on the users topic.\n\n"
},
"promptType": "define",
"hasOutputParser": true
},
"typeVersion": 1.6
},
{
"id": "cfc55dbb-78e6-47ef-bf55-810311bd37e8",
"name": "调用 Perplexity 研究员",
"type": "@n8n/n8n-nodes-langchain.toolWorkflow",
"position": [
1780,
1040
],
"parameters": {
"name": "perplexity_research_tool",
"fields": {
"values": [
{
"name": "topic",
"stringValue": "= {{ $json.text }}"
}
]
},
"workflowId": {
"__rl": true,
"mode": "id",
"value": "HnqGW0eq5asKfZxf"
},
"description": "调用此工具执行 Perplexity 研究",
"jsonSchemaExample": "{\n \"topic\": \"\"\n}"
},
"typeVersion": 1.2
},
{
"id": "5ca35a40-506d-4768-a65c-a331718040bc",
"name": "不执行任何操作",
"type": "n8n-nodes-base.noOp",
"position": [
2320,
1140
],
"parameters": {},
"typeVersion": 1
},
{
"id": "17028837-4706-43f3-8291-f150860caa4c",
"name": "无操作2",
"type": "n8n-nodes-base.noOp",
"position": [
1020,
1700
],
"parameters": {},
"typeVersion": 1
},
{
"id": "adebf1ad-62d9-4b79-b9a1-4a9395067803",
"name": "无操作3",
"type": "n8n-nodes-base.noOp",
"position": [
2000,
1700
],
"parameters": {},
"typeVersion": 1
},
{
"id": "fe19e472-3b2b-4c07-b957-fb2afc426998",
"name": "无操作4",
"type": "n8n-nodes-base.noOp",
"position": [
1260,
1080
],
"parameters": {},
"typeVersion": 1
},
{
"id": "41e23462-a7fa-42a8-adbc-83a662f63f0c",
"name": "便签7",
"type": "n8n-nodes-base.stickyNote",
"position": [
1460,
760
],
"parameters": {
"color": 3,
"width": 480,
"height": 488,
"content": "## 🤖执行 Perplexity 研究"
},
"typeVersion": 1
},
{
"id": "dcc3bd83-1f8c-4000-a832-c2c6e7c157ba",
"name": "获取主题",
"type": "n8n-nodes-base.set",
"position": [
380,
860
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "57f0eab2-ef1b-408c-82d5-a8c54c4084a6",
"name": "topic",
"type": "string",
"value": "={{ $json.query.topic }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "5572e5b1-0b4c-4e6d-b413-5592aab59571",
"name": "如果主题存在",
"type": "n8n-nodes-base.if",
"position": [
560,
860
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "2c565aa5-0d11-47fb-8621-6db592579fa8",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.topic }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "509ee61f-defb-41e8-84cf-70ac5a7448d0",
"name": "改进用户主题",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
880,
860
],
"parameters": {
"text": "=How would you improve the following prompt as of {{ $now }}, focusing on:\n\n1. Key Concepts & Definitions\n - Main terminology and foundational concepts\n - Technical background and context\n\n2. Core Components\n - Essential elements and their relationships\n - Critical processes and workflows\n\n3. Practical Applications\n - Real-world use cases\n - Implementation considerations\n\n4. Analysis & Insights\n - Advantages and limitations\n - Best practices and recommendations\n\nThe final output should be a maximum 2 sentence pure text prompt without any preamble or further explanation. The final output will be providced to Perplexity as a research prompt.\n\nPrompt to analyze: {{ $json.topic }}",
"promptType": "define"
},
"typeVersion": 1.4
},
{
"id": "69ee4c6a-f6ef-47a2-bd5c-ccaf49ec7c94",
"name": "如果主题",
"type": "n8n-nodes-base.if",
"position": [
1260,
860
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "329653d4-330f-4b41-96e7-4652c1448902",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.text }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "daa3027b-774d-44b1-b0a5-27008768c65d",
"name": "聊天 ID",
"type": "n8n-nodes-base.set",
"position": [
2120,
880
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "0aa8fcc9-26f4-485c-8fc1-a5c13d0dd279",
"name": "telegram_chat_id",
"type": "number",
"value": 1234567890
}
]
}
},
"typeVersion": 3.4
},
{
"id": "97f32ad1-f91e-4ccc-8248-d10da823b26a",
"name": "文章",
"type": "n8n-nodes-base.set",
"position": [
780,
1480
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "0eb5952b-c133-4b63-8102-d4b8ec7b9b5a",
"name": "article",
"type": "object",
"value": "={{ $json.output.article }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "e223dee3-c79f-421d-b2b8-2f3551a45f71",
"name": "提取 JSON",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
440,
1480
],
"parameters": {
"text": "=Extract a JSON object from this content: {{ $json.output }}",
"options": {},
"promptType": "define",
"hasOutputParser": true
},
"retryOnFail": true,
"typeVersion": 1.6
},
{
"id": "de8aafb6-b05d-4278-8719-9b3c266fcf3a",
"name": "如果文章",
"type": "n8n-nodes-base.if",
"position": [
1020,
1480
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "329653d4-330f-4b41-96e7-4652c1448902",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "{{ $json.article }}",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "f9450b58-3b81-4b61-8cbf-2cdf5a2f56a0",
"name": "创建 HTML 文章",
"type": "@n8n/n8n-nodes-langchain.agent",
"position": [
1360,
1480
],
"parameters": {
"text": "=Convert this verbatim into HTML: {{ $json.article.toJsonString() }}\n\n## Formatting Guidelines\n- HTML document must be single line document without tabs or line breaks\n- Use proper HTML tags throughout\n- Do not use these tags: <html> <body> <style> <head>\n- Use <h1> tag for main title\n- Use <h2> tags for secondary titles\n- Structure with <p> tags for paragraphs\n- Include appropriate spacing\n- Use <blockquote> for direct quotes\n- Maintain consistent formatting\n- Write in clear, professional tone\n- Break up long paragraphs\n- Use engaging subheadings\n- Include transitional phrases\n\nThe final JSON response should contain only the title and content fields, with the content including all HTML formatting.\n{\n\t\"title\": \"the title\",\n\t\"content\": \"the HTML\"\n}",
"agent": "conversationalAgent",
"options": {},
"promptType": "define"
},
"retryOnFail": true,
"typeVersion": 1.6
},
{
"id": "53cbaa6e-6508-48e3-9a5a-58f5bc111c2d",
"name": "如果 HTML",
"type": "n8n-nodes-base.if",
"position": [
1780,
1480
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "329653d4-330f-4b41-96e7-4652c1448902",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.output.parseJson().title }}",
"rightValue": ""
},
{
"id": "0a05f73a-2901-4157-8194-cb81d259ce71",
"operator": {
"type": "string",
"operation": "notEmpty",
"singleValue": true
},
"leftValue": "={{ $json.output.parseJson().content }}",
"rightValue": ""
},
{
"id": "b61c1d25-a010-42d3-9f9d-fa927c483bae",
"operator": {
"name": "filter.operator.equals",
"type": "string",
"operation": "equals"
},
"leftValue": "",
"rightValue": ""
}
]
}
},
"typeVersion": 2.2
},
{
"id": "33e4e2cd-be0c-4fc9-b705-b0e8aac496f9",
"name": "内容",
"type": "n8n-nodes-base.set",
"position": [
2000,
1480
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "af335333-acb8-4c9e-8184-d20cd03e08f6",
"name": "title",
"type": "string",
"value": "={{ $json.output.parseJson().title }}"
},
{
"id": "7fbd2264-c0e1-4bdc-b754-b0faa538879c",
"name": "content",
"type": "string",
"value": "={{ $json.output.parseJson().content }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "8bf36853-8a04-4a0b-8715-e03a8fc8359d",
"name": "聊天 ID1",
"type": "n8n-nodes-base.set",
"position": [
660,
2180
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "0aa8fcc9-26f4-485c-8fc1-a5c13d0dd279",
"name": "telegram_chat_id",
"type": "number",
"value": 1234567890
}
]
}
},
"typeVersion": 3.4
},
{
"id": "a3fe75d1-8db0-45cb-87f6-76fc27cb59f6",
"name": "便签3",
"type": "n8n-nodes-base.stickyNote",
"position": [
600,
2080
],
"parameters": {
"width": 420,
"height": 340,
"content": "## 可选"
},
"typeVersion": 1
},
{
"id": "22e9edbc-7aa6-4549-ae9f-2c31ad7d0542",
"name": "便签8",
"type": "n8n-nodes-base.stickyNote",
"position": [
2060,
760
],
"parameters": {
"width": 420,
"height": 340,
"content": "## 可选"
},
"typeVersion": 1
},
{
"id": "e62ff7d5-bd54-434c-b048-0dc7cd2c7f9b",
"name": "**结果：** 整理好的商家电子邮件数据库，准备用于外联",
"type": "n8n-nodes-base.set",
"position": [
1700,
2080
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "eb89464a-5919-4962-880c-3f5903e267de",
"name": "response",
"type": "string",
"value": "={{ $('Perplexity').item.json.choices[0].message.content }}"
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "c6ba0613-47c6-442f-99e8-0eaec8cacc20",
"name": "错误响应",
"type": "n8n-nodes-base.set",
"position": [
1700,
2300
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "eb89464a-5919-4962-880c-3f5903e267de",
"name": "response",
"type": "string",
"value": "=Error. No topic provided."
}
]
},
"includeOtherFields": true
},
"typeVersion": 3.4
},
{
"id": "30d8065c-55d8-4099-abb2-ddb01635129d",
"name": "Perplexity",
"type": "n8n-nodes-base.httpRequest",
"position": [
1500,
2080
],
"parameters": {
"url": "https://api.perplexity.ai/chat/completions",
"method": "POST",
"options": {},
"jsonBody": "={\n \"model\": \"llama-3.1-sonar-small-128k-online\",\n \"messages\": [\n {\n \"role\": \"system\",\n \"content\": \"{{ $json.system }}\"\n },\n {\n \"role\": \"user\",\n \"content\": \"{{ $json.user }}\"\n }\n ],\n \"max_tokens\": \"4000\",\n \"temperature\": 0.2,\n \"top_p\": 0.9,\n \"return_citations\": true,\n \"search_domain_filter\": [\n \"perplexity.ai\"\n ],\n \"return_images\": false,\n \"return_related_questions\": false,\n \"search_recency_filter\": \"month\",\n \"top_k\": 0,\n \"stream\": false,\n \"presence_penalty\": 0,\n \"frequency_penalty\": 1\n}",
"sendBody": true,
"specifyBody": "json",
"authentication": "genericCredentialType",
"genericAuthType": "httpHeaderAuth"
},
"credentials": {
"httpCustomAuth": {
"id": "vxjFugFpr4Od6gws",
"name": "Confluence REST API"
},
"httpHeaderAuth": {
"id": "wokWVLDQUDi0DC7I",
"name": "Perplexity"
}
},
"typeVersion": 4.2
}
],
"active": false,
"pinData": {},
"settings": {
"executionOrder": "v1"
},
"versionId": "9ebf0569-4d9d-4783-b797-e5df2a8e8415",
"connections": {
"If": {
"main": [
[
{
"node": "Prompts",
"type": "main",
"index": 0
}
],
[
{
"node": "No Operation, do nothing",
"type": "main",
"index": 0
}
]
]
},
"If2": {
"main": [
[
{
"node": "Extract JSON",
"type": "main",
"index": 0
}
],
[
{
"node": "Do Nothing",
"type": "main",
"index": 0
}
]
]
},
"Article": {
"main": [
[
{
"node": "If Article",
"type": "main",
"index": 0
}
]
]
},
"Chat Id": {
"main": [
[
{
"node": "Telegram",
"type": "main",
"index": 0
}
]
]
},
"If HTML": {
"main": [
[
{
"node": "Contents",
"type": "main",
"index": 0
}
],
[
{
"node": "Do Nothing3",
"type": "main",
"index": 0
}
]
]
},
"Prompts": {
"main": [
[
{
"node": "Perplexity",
"type": "main",
"index": 0
}
]
]
},
"Webhook": {
"main": [
[
{
"node": "Get Topic",
"type": "main",
"index": 0
}
]
]
},
"Chat Id1": {
"main": [
[
{
"node": "Telegram2",
"type": "main",
"index": 0
}
]
]
},
"Contents": {
"main": [
[
{
"node": "Basic LLM Chain",
"type": "main",
"index": 0
}
]
]
},
"If Topic": {
"main": [
[
{
"node": "Perplexity Topic Agent",
"type": "main",
"index": 0
}
],
[
{
"node": "Do Nothing4",
"type": "main",
"index": 0
}
]
]
},
"Get Topic": {
"main": [
[
{
"node": "If Topic Exists",
"type": "main",
"index": 0
}
]
]
},
"Telegram2": {
"main": [
[
{
"node": "If",
"type": "main",
"index": 0
}
]
]
},
"If Article": {
"main": [
[
{
"node": "Create HTML Article",
"type": "main",
"index": 0
}
],
[
{
"node": "Do Nothing2",
"type": "main",
"index": 0
}
]
]
},
"Perplexity": {
"main": [
[
{
"node": "Success Response",
"type": "main",
"index": 0
}
]
]
},
"gpt-4o-mini": {
"ai_languageModel": [
[
{
"node": "Create HTML Article",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Extract JSON": {
"main": [
[
{
"node": "Article",
"type": "main",
"index": 0
}
]
]
},
"gpt-4o-mini1": {
"ai_languageModel": [
[
{
"node": "Perplexity Topic Agent",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"gpt-4o-mini2": {
"ai_languageModel": [
[
{
"node": "Extract JSON",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"gpt-4o-mini3": {
"ai_languageModel": [
[
{
"node": "Improve Users Topic",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"gpt-4o-mini5": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Basic LLM Chain": {
"main": [
[
{
"node": "Respond to Webhook",
"type": "main",
"index": 0
}
]
]
},
"If Topic Exists": {
"main": [
[
{
"node": "Improve Users Topic",
"type": "main",
"index": 0
}
],
[
{
"node": "Do Nothing1",
"type": "main",
"index": 0
}
]
]
},
"Create HTML Article": {
"main": [
[
{
"node": "If HTML",
"type": "main",
"index": 0
}
]
]
},
"Improve Users Topic": {
"main": [
[
{
"node": "If Topic",
"type": "main",
"index": 0
}
]
]
},
"Perplexity Topic Agent": {
"main": [
[
{
"node": "If2",
"type": "main",
"index": 0
},
{
"node": "Chat Id",
"type": "main",
"index": 0
}
]
]
},
"Execute Workflow Trigger": {
"main": [
[
{
"node": "Chat Id1",
"type": "main",
"index": 0
}
]
]
},
"No Operation, do nothing": {
"main": [
[
{
"node": "Error Response",
"type": "main",
"index": 0
}
]
]
},
"Structured Output Parser1": {
"ai_outputParser": [
[
{
"node": "Extract JSON",
"type": "ai_outputParser",
"index": 0
}
]
]
},
"Call Perplexity Researcher": {
"ai_tool": [
[
{
"node": "Perplexity Topic Agent",
"type": "ai_tool",
"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？
这是一个高级难度的工作流，适用于AI、Marketing等场景。适合高级用户，包含 16+ 个节点的复杂工作流
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
ELEVEN LABS 多智能体系统
基于ElevenLabs、OpenAI和内容生成的语音驱动营销助手
If
Set
Webhook
+16
56 节点David Olusola
AI
AI邮件分诊与GPT-4警报系统及Telegram通知
AI邮件分诊与GPT-4警报系统及Telegram通知
If
Set
Gmail
+22
104 节点Peter Joslyn
Support
✨🩷 自动化社交媒体内容发布工厂 + 系统提示组合
✨🩷 自动化社交媒体内容发布工厂 + 系统提示组合
If
Set
Code
+20
100 节点Joseph LePage
Marketing
[模板] AI宠物店 v8
🐶 AI宠物店助手 - 集成GPT-4o、Google日历和WhatsApp/Instagram/Facebook
If
N8n
Set
+38
244 节点Amanda Benks
Sales
✍️🌄 您的首个WordPress内容创建器 - 快速入门
✍️🌄 您的首个WordPress + AI内容创建器 - 快速入门
If
Set
Code
+10
39 节点Joseph LePage
AI
✨🤖 自动化AI驱动的社交媒体内容工厂，适用于X + Facebook + Instagram + LinkedIn
✨🤖 使用AI自动化多平台社交媒体内容创建
If
Set
Code
+17
57 节点Joseph LePage
AI
工作流信息
难度等级
高级
节点数量47
分类2
节点类型14
作者
Joseph LePage
@joeAs an AI Automation consultant based in Canada, I partner with forward-thinking organizations to implement AI solutions that streamline operations and drive growth.
外部链接
在 n8n.io 上查看 →
分享此工作流