Turn a list of web pages into clean markdown and a full link list. Great for marketing and SEO teams that need readable text for research, content planning, or AI models. It saves manual copy work and keeps the data consistent across pages.

The flow starts with a manual test run. URLs come from an example array or your own database under a field named Page. The list is split so each URL is handled on its own. A limit of 40 items controls memory use. Batches of 10 go to the Firecrawl API through an HTTP Request, which returns markdown, page metadata, and links. A short wait of 45 seconds is used to respect 10 requests per minute. The mapped fields are prepared so you can write the results into your system of record.

To get going, add your Firecrawl API key to the Authorization header and keep the Page field name for URLs. Adjust the 40 item limit and the 10 item batch size to fit your server. Expect faster content prep, fewer errors from copy and paste, and consistent structure for AI pipelines. Common uses include building a knowledge base, link audits for SEO, and feeding LLM prompts with clean context.

• Manual test run to start the process safely
• Example Page array ready for quick input or swap with your database
• Split out URLs so each page is processed as a single item
• Limit to 40 items per run to manage server memory
• Process 10 URLs at a time with batch control
• Wait 45 seconds to respect 10 requests per minute API limits
• HTTP Request to Firecrawl returns markdown, metadata, and links
• Field mapping collects title, description, markdown, and all page links
• Placeholder nodes mark where to connect your input and output systems

• Reduce manual copy work from hours to minutes
• Automate up to 90% of page processing
• Handle 40 URLs per cycle and 10 requests per minute safely
• Improve data quality by removing HTML and keeping structure
• Feed clean text and links to AI and SEO tools without extra cleaning

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with Firecrawl. See the Tools Required section above for links to create accounts with these services.
3. Create your Firecrawl API key in your Firecrawl account and keep it secure.
4. Open the HTTP Request node named Retrieve Page Markdown and Links. In the Credential to connect with dropdown, click Create new credential. Choose HTTP Header Auth, then add header name Authorization and value Bearer YOUR_API_KEY. Save the credential with a clear name.
5. Verify the HTTP Request node uses POST and points to the Firecrawl scrape endpoint. Confirm the JSON body includes the url field and formats such as markdown.
6. Open the Set node named Example fields from data source if you want a quick test. Keep the array under the Page field or replace it with your URLs.
7. If you have your own URL source, connect it to the flow where the Get urls from own data source and Connect to your own data source placeholders are. Keep the field name Page so the Split out page URLs node can read it.
8. Open the Limit node and set the max items for each cycle. Use 40 by default to avoid memory issues, or lower it if your server is small.
9. Open the SplitInBatches node and set batch size to 10. This controls how many URLs are sent before waiting.
10. Open the Wait node and keep the 45 seconds pause. Increase the wait time if you see rate limit errors, or reduce it only if your API plan allows more requests.
11. Run a test with the manual trigger. Check the Markdown data and Links node output for title, description, markdown, and links. Confirm the count matches the number of URLs processed.
12. Replace the output placeholder with your chosen database or storage node when ready. Map the fields from Markdown data and Links. If you connect a new app, double click that node, choose Create new credential, and follow the on screen steps.
13. Troubleshoot common issues: if you get 429 errors, lower the batch size or increase the wait; if memory errors appear, lower the 40 item limit; if responses are empty, check the target URL and your API key.