Bing Copilot Data Extraction and AI Summarization with Gemini - n8n Workflow

• Users needing reliable web scraping for dynamic content like Bing Copilot.


• Developers and analysts who require structured JSON data extracted from large text blocks.


• Teams looking for advanced AI summarization capabilities within their automated processes.


• Anyone seeking to leverage Bright Data and Google Gemini effectively in a single n8n workflow.

This sophisticated n8n workflow addresses the challenge of extracting actionable information from dynamic search results, specifically Bing Copilot. It starts by using the Bright Data Web Scraper API to scrape results based on a user-defined prompt (e.g., 'Top hotels in New York'). Since scraping jobs take time, the n8n workflow employs a robust polling mechanism using the n8n node Check Snapshot Status and an If node, ensuring it waits until the data is fully ready before proceeding. Once the data is downloaded, the workflow splits into two parallel AI processing paths using Google Gemini via dedicated n8n nodes. One path extracts structured data based on a JSON schema, and the other generates a concise summary. This powerful combination transforms raw web content into highly organized and digestible formats, proving the versatility of n8n templates for complex data automation.

This automation starts with the manual n8n trigger, initiating the entire process.

1. Initiate Scrape: The Perform a Bing Copilot Request n8n node sends a request to the Bright Data API to start the scraping job, receiving a snapshot_id.


2. Snapshot Tracking: The Set Snapshot Id n8n node captures this ID.


3. Status Polling: The Check Snapshot Status n8n node repeatedly queries Bright Data. The If node checks if the status is 'ready'. If not, the Wait for 30 seconds n8n node pauses the execution before trying again, creating a continuous polling loop vital for Bright Data jobs.


4. Error Handling & Download: Once ready, an error check ensures data integrity. Then, the Download Snapshot n8n node retrieves the full JSON data containing the Bing Copilot answer.


5. Parallel AI Processing: The data flows simultaneously to two branches:

To deploy this n8n workflow effectively, follow these steps:

1. Import: Copy the provided JSON code and import it directly into your n8n instance.


2. Bright Data Credentials: Configure a 'Header Auth' credential (named 'Header Auth account' in the template) for the Perform a Bing Copilot Request, Check Snapshot Status, and Download Snapshot n8n nodes. This credential requires your Bright Data API key.


3. Google Gemini Credentials: Configure the 'Google Gemini (PaLM) Api account' credential using your Google Gemini API key. This is required by the Google Gemini Chat Model and Google Gemini Chat Model1 n8n nodes.


4. Update Webhooks: Update the URLs in the Structured Data Webhook Notifier and Summary Webhook Notifier n8n nodes (currently set to a placeholder webhook.site address) to your desired endpoint for receiving the processed data.


5. Activate: Save and activate this powerful n8n workflow. Use the manual n8n trigger to test the process.

When clicking ‘Test workflow’ (n8n trigger): The manual starting point for the n8n workflow, ideal for testing and ad-hoc execution.
Perform a Bing Copilot Request (HTTP Request n8n node): Triggers the Bing Copilot search and scraping job via the Bright Data API. Key Configuration: Uses HTTP POST to datasets/v3/trigger, includes datasetid in query parameters, and sends the prompt ('Top hotels in New York') in the JSON body.
Check Snapshot Status (HTTP Request n8n node): Essential for the polling loop. It repeatedly checks the status of the Bright Data scraping job using the captured snapshotid.
If (n8n node): Controls the polling logic. If the Bright Data job status equals 'ready', the n8n workflow continues; otherwise, it is routed to the Wait node.
Download Snapshot (HTTP Request n8n node): Fetches the complete, scraped data from Bright Data after the job is confirmed 'ready'.
Google Gemini Chat Model / Google Gemini Chat Model1 (Langchain LLM Chat n8n nodes): Uses the Gemini 2.0 Flash Exp model to power both the structured data extraction and the summarization tasks within the n8n workflow.
Structured Output Parser (Langchain Output Parser n8n node): Defines the exact JSON schema that the AI must adhere to, ensuring predictable, high-quality data extraction.
Structured Data Extractor (Langchain Chain LLM n8n node): Uses Gemini to take the raw Bing Copilot text and structure it according to the specified schema, enhancing the utility of this n8n template.
Concise Summary Creator (Langchain Chain Summarization n8n node): Handles the complex task of summarization, leveraging the LangChain capabilities in n8n to condense large text inputs into readable summaries.
Structured Data Webhook Notifier / Summary Webhook Notifier (HTTP Request n8n nodes): The final action steps in this n8n workflow, sending the processed JSON and the summary to external endpoints.