n8n Chat Webhook: Q&A from Uploaded PDFs with OpenAI

Ask questions about your uploaded PDFs—directly from an n8n website chat webhook

This n8n workflow creates a public chat webhook for your website where visitors can upload a PDF, TXT, or Markdown file and ask Q&A questions. Using OpenAI, the agent answers based on the uploaded document text and keeps that content in conversation memory for follow-up questions.

What this workflow does

• Receives chat messages via a public webhook, with streaming responses and file upload support.
• Detects attachments: checks whether the visitor attached a file and determines if it’s a PDF or a text/Markdown document.
• Extracts and prepares content: extracts text from the uploaded file (or flags it as unreadable/unsupported) and truncates it to the first 20,000 characters.
• Enables follow-up Q&A: stores the extracted document text as hidden messages in the chat conversation memory, so later turns can reference it.
• Answers with OpenAI: uses an OpenAI chat model to respond strictly from the document content, streaming the reply back to the chat. If only a file is sent, it can generate a short summary.

Use cases

• Let customers upload product manuals and ask questions about specific sections.
• Enable internal teams to upload policy documents and run Q&A during support conversations.
• Turn uploaded documentation (TXT/Markdown/PDF) into an interactive help assistant for your SaaS.

Technical details

• Nodes/logic: if, set, switch, sticky note, n8nn8n-nodes-langchainagent, and extract from file.
• Core behavior: file text extraction + 20,000-character truncation + OpenAI chat answering with conversation memory.
• Setup requirements: add an OpenAI API credential (or swap the chat model node), publish the workflow, copy the production chat webhook URL, and configure the chat widget with file uploads.