Web Research Automation with Structured Data and Screenshots using Firecrawl and OpenAI

Who’s it for

This workflow is designed for researchers, content creators, and AI agents who need to quickly scrape structured web data and capture full-page screenshots for further use. It’s especially useful for automating competitive research, news monitoring, and content curation.

How it works

The workflow uses the Firecrawl API integrated with n8n to perform web searches and return results in structured formats (Markdown and screenshots). It includes:

A search agent that transforms natural language queries into Firecrawl-compatible search strings. HTTP requests to retrieve results from specific sites (e.g., YouTube, news outlets) or across the web. Automatic capture of full-page screenshots alongside structured text. Integration with the OpenAI Chat Model for enhanced query handling.

How to set up

Import this workflow into your n8n instance. Add and configure your Firecrawl API credentials. Add your OpenAI credentials for natural language query parsing. Trigger the workflow via the included chat input or modify it to run on schedule.

Requirements

A Firecrawl account with an active API key. n8n self-hosted or cloud instance. OpenAI account if you want to enhance search queries.

How to customize the workflow

Update the search queries to focus on your preferred sites or keywords. Adjust the number of results with the limit parameter. Extend the workflow to store screenshots in Google Drive, Notion, or your database. Replace the chat trigger with any other event trigger (webhook, schedule, etc.).