AI Website Scraping Agent with OpenAI

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": "Tool - Website Agent",
  "nodes": [
    {
      "parameters": {
        "promptType": "define",
        "text": "={{ $json.query }}",
        "options": {
          "systemMessage": "=### System Prompt for n8n AI Agent: Website Analyzer\n\nYou are an AI Agent designed to analyze web pages.  \nWhen the user says something similiar to, \"analyze the following page\" and provides a URL, follow these steps:\n\n1. **Scrape the web page at the provided URL using the available Tool.**\n2. **Extract and return the following information:**\n   - **Title:** The main title of the page.\n   - **Description:** The meta description or a brief summary if no meta description is available.\n   - **Summary:** A concise summary of the main content of the page.\n   - **N-gram Analysis:**  \n     - **Guidelines for N-gram Analysis:**\n       - Only analyze the main textual content of the page (ignore navigation, ads, and boilerplate).\n       - Remove all common stop words (such as \"the\", \"and\", \"is\", etc.) from the analysis.\n       - Normalize text by converting to lowercase and removing punctuation.\n       - For each n-gram type, list the top 10 most frequent items.\n       - Present results as lists, ordered by frequency (most frequent first).\n       - For bigrams and trigrams, do not include n-grams that contain only stop words.\n       - If the page has very little text, return as many n-grams as possible and note the limitation.\n\n     - **Unigram:** List of the most frequent single words (excluding stop words).\n     - **Bigram:** List of the most frequent two-word combinations (excluding those with only stop words).\n     - **Trigram:** List of the most frequent three-word combinations (excluding those with only stop words).\n\n**Output your results in a clear, structured format.**  \nIf any information is missing or cannot be found, state this explicitly.\n\n---\n\n#### Example Output\n\n**Title:** [Page Title]  \n**Description:** [Meta Description or brief summary]  \n**Summary:** [Concise summary of main content]  \n**N-gram Analysis:**  \n- **Unigram:** [word1, word2, ...]  \n- **Bigram:** [word1 word2, word2 word3, ...]  \n- **Trigram:** [word1 word2 word3, word2 word3 word4, ...]"
        }
      },
      "type": "@n8n/n8n-nodes-langchain.agent",
      "typeVersion": 2,
      "position": [
        220,
        0
      ],
      "id": "3c621351-ad1f-4652-a2c8-0fe0db3a094a",
      "name": "Scrape Agent"
    },
    {
      "parameters": {
        "model": {
          "__rl": true,
          "mode": "list",
          "value": "gpt-4.1-mini"
        },
        "options": {}
      },
      "type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
      "typeVersion": 1.2,
      "position": [
        200,
        200
      ],
      "id": "e01072a7-f096-4ea2-b238-da6362e9fd3d",
      "name": "GPT-4.1-mini",
      "credentials": {
        "openAiApi": {
          "name": "<your credential>"
        }
      }
    },
    {
      "parameters": {
        "inputSource": "passthrough"
      },
      "type": "n8n-nodes-base.executeWorkflowTrigger",
      "typeVersion": 1.1,
      "position": [
        0,
        0
      ],
      "id": "4d03f030-1eef-4c0c-8055-2f0f11096f0b",
      "name": "When Executed by Another Workflow"
    },
    {
      "parameters": {
        "description": "Call this tool to scrape websites by providing a URL and return the prepared data.",
        "workflowId": {
          "__rl": true,
          "value": "RkFq0paD5YLLof2K",
          "mode": "list",
          "cachedResultName": "Tool - Serper Crawl URL"
        },
        "workflowInputs": {
          "mappingMode": "defineBelow",
          "value": {},
          "matchingColumns": [],
          "schema": [],
          "attemptToConvertTypes": false,
          "convertFieldsToString": false
        }
      },
      "type": "@n8n/n8n-nodes-langchain.toolWorkflow",
      "typeVersion": 2.2,
      "position": [
        420,
        220
      ],
      "id": "af549453-a56e-4ca6-bffe-a1b074234d07",
      "name": "Call Serper"
    }
  ],
  "connections": {
    "GPT-4.1-mini": {
      "ai_languageModel": [
        [
          {
            "node": "Scrape Agent",
            "type": "ai_languageModel",
            "index": 0
          }
        ]
      ]
    },
    "When Executed by Another Workflow": {
      "main": [
        [
          {
            "node": "Scrape Agent",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Call Serper": {
      "ai_tool": [
        [
          {
            "node": "Scrape Agent",
            "type": "ai_tool",
            "index": 0
          }
        ]
      ]
    }
  },
  "active": false,
  "settings": {
    "executionOrder": "v1"
  },
  "versionId": "4d9d630a-3d6f-4a6b-83ea-8607cec1d98b",
  "meta": {
    "templateCredsSetupCompleted": true
  },
  "id": "4DiQ75K8cdqHbobH",
  "tags": []
}

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.

• openAiApi

About this workflow

Tool - Website Agent. Uses agent, lmChatOpenAi, executeWorkflowTrigger, toolWorkflow. Event-driven trigger; 4 nodes.

Source: https://github.com/Marvomatic/n8n-templates/blob/main/seo-data-analyst/tool_website_agent.json — original creator credit. Request a take-down →

More AI & RAG workflows → · Browse all categories →

Related workflows

Workflows that share integrations, category, or trigger type with this one. All free to copy and import.