Apify PDF & Office URL to RAG-Ready Markdown Chunks in n8n

Turn PDF and Office URLs into RAG-ready Markdown chunks—inside n8n

This n8n automation workflow takes PDF and Office document URLs, converts them to Markdown with the Apify Document to Markdown Actor, and outputs one n8n item per chunk with citation-friendly metadata for downstream RAG indexing.

What this workflow does

• Runs manually (Manual Trigger) so you can control when documents are processed.
• Lets you set your file URLs plus chunk size and chunk overlap parameters.
• Sends those URLs and chunking settings to the Apify Document to Markdown Actor using the Apify API, generating Markdown output and extracted tables.
• Splits the actor results into one output item per chunk, preserving chunk text along with metadata such as:
  • source URL
  • title
  • page range
  • nearest heading (for citation-friendly context)

Use cases

• RAG ingestion for document collections: index PDFs and Office files by turning them into consistent, chunked Markdown.
• Support knowledge bases: process policy, terms, or manuals stored as URLs to improve retrieval quality.
• SaaS ops and internal tooling: repeatedly re-chunk content with different chunk size/overlap to optimize search and citations.

Technical details

• Apify integration: uses an Apify API token with an HTTP Header Auth credential formatted as Authorization: Bearer <YOUR_APIFY_TOKEN>.
• Nodes / building blocks: set, code, sticky note, HTTP request, manual trigger.
• Configuration: replace the sample fileUrls with your public or pre-signed document URLs and tune chunkSize / chunkOverlap.