Ragfeed Webpage Chunking n8n Workflow (Vector-Ready RAG)

Turn any webpage URL into vector-ready text chunks with Ragfeed (n8n)

This Ragfeed Webpage Chunking n8n Workflow lets you manually send a webpage URL to the Ragfeed scraping API and receive token-bounded content chunks back—already split into one n8n item per chunk for easy embedding and storage in your vector database.

What this workflow does

• Runs on demand using a manual trigger.
• Accepts two key inputs: the target webpage URL and a maximum chunk size (in tokens).
• Scrapes the page via Ragfeed by sending a POST request to https://api.ragfeed.de/v2/scrape.
• Processes Ragfeed’s response: it takes the returned data.chunks array and splits it so you get one n8n item per chunk.
• Outputs chunked text in a format that’s ready for embeddings and independent storage.

Use cases

• RAG ingestion pipelines: generate embedding inputs from a webpage while respecting token limits.
• Vector database population: store each chunk as a separate record for retrieval-optimized search.
• Ops for SaaS teams: automate the first step of webpage-to-RAG ingestion with a simple manual run.
• Content testing: quickly try different maxTokens values to find the best chunk granularity.

Technical details

• n8n nodes used: Manual Trigger, Set, Split Out, Sticky Note, HTTP Request.
• Ragfeed API endpoint: POST /v2/scrape at https://api.ragfeed.de/v2/scrape.
• Authentication: use an n8n Header Auth credential to send Authorization: Bearer <your Ragfeed API key>.
• Config to update: the webpage URL and the maxTokens chunk size.