document-to-pdf

n8n community node that converts HTML, Markdown, and Word (DOCX) documents to PDF

Package Information

Released: 1 mo ago
Downloads: 0 weekly / 0 monthly
Latest Version: 0.3.0
Author: Enes Karaaslan

Available Nodes

Documentation generating…

Document to PDF icon
Document to PDF
Convert HTML, Markdown, or Word (DOCX) documents to PDF

Documentation

n8n-nodes-document-to-pdf

An n8n community node that converts HTML, Markdown, and Word (DOCX) documents to PDF — fully self-contained, no external API or credentials required.

  • HTML → PDF — full CSS fidelity via headless Chromium when available, with a pure-JavaScript fallback that needs no browser at all
  • Markdown → PDF — GitHub-flavored Markdown, rendered with clean default document styling (override with custom CSS)
  • Word (DOCX) → PDF — .docx converted via mammoth (headings, lists, tables, bold/italic, images), then rendered to PDF

Rendering engines

Engine Requires Best for
Chromium (Puppeteer) Chrome/Chromium installed locally, or a remote browser service (Steel, browserless) Pixel-perfect output: full CSS, backgrounds, web fonts, page headers/footers
Pure JavaScript (pdfmake) Nothing Markdown, DOCX, and simple HTML on hosts without a browser (e.g. stock n8n Docker image)

The default Auto mode uses Chromium when a browser is found (local install or a configured remote endpoint) and silently falls back to the pure-JavaScript engine otherwise — so the node works out of the box on an unmodified n8n Docker container.

Using a remote browser (Steel, browserless, …)

If you run a browser service next to n8n, set the Remote Browser WebSocket URL option (or the BROWSER_WS_ENDPOINT environment variable) to its CDP websocket endpoint and the node connects to it instead of launching a local browser — full Chromium fidelity with no changes to the n8n image, and it survives n8n updates. Example with Steel in the same docker-compose network:

services:
  n8n:
    image: docker.n8n.io/n8nio/n8n
    environment:
      - BROWSER_WS_ENDPOINT=ws://steel:3000
  steel:
    image: ghcr.io/steel-dev/steel-browser-api:latest

Use the compose service name as the host (ws://steel:3000), not localhost — inside the n8n container, localhost is the n8n container itself.

Pure-JavaScript engine limitations: no CSS stylesheets (semantic HTML + inline basics only), images must be data URIs, and the Custom CSS / Print Background / Scale / Page Ranges / header-footer options are ignored (page size, orientation, and margins are honored). Text uses embedded Roboto, covering Latin (incl. Turkish), Cyrillic, and Greek.

Installation

In n8n: Settings → Community Nodes → Install and enter:

n8n-nodes-document-to-pdf

Or on a self-hosted instance:

npm install n8n-nodes-document-to-pdf

Note: this package depends on Puppeteer, which downloads a Chromium build (~150 MB) on install. In Docker images you can skip the download and point the node at an existing Chrome/Chromium binary instead:

PUPPETEER_SKIP_DOWNLOAD=true npm install n8n-nodes-document-to-pdf
export PUPPETEER_EXECUTABLE_PATH=/usr/bin/chromium-browser

(Or set Chrome Executable Path in the node's options.)

Operations

Operation Input Output
HTML to PDF HTML as text (expressions supported) or a binary file PDF in a binary field
Markdown to PDF Markdown as text or a binary file PDF in a binary field
Word (DOCX) to PDF .docx file in a binary field PDF in a binary field

Legacy binary .doc files (Word 97–2003) are not supported — convert them to .docx first.

Options

  • Page Format (A3–A5, Letter, Legal, Ledger, Tabloid), Landscape, Margins, Scale, Page Ranges
  • Print Background — render background colors/images (on by default)
  • Header/Footer templates — Chromium print header/footer HTML with pageNumber / totalPages / date / title spans
  • Custom CSS — injected into the document before rendering (works for all three operations)
  • Wait for Network Idle — wait for external images/fonts to load before rendering; turn off for faster conversion of self-contained documents
  • Chrome Executable Path — use a system Chrome/Chromium instead of the bundled one (also honors PUPPETEER_EXECUTABLE_PATH)

Example: Markdown report → PDF → email

  1. Build your Markdown in a Set or Code node (e.g. {{ $json.report }})
  2. Document to PDF → operation Markdown to PDF, Markdown {{ $json.report }}
  3. Send Email → attach the data binary field

Troubleshooting

"Could not find Chrome" on n8n Docker

The official n8n Docker image ships without a browser, and community-node installation skips Puppeteer's Chromium download (which wouldn't run on the Alpine-based image anyway). Install Chromium in the container:

docker exec -u root -it <n8n-container> apk add --no-cache chromium

Or bake it into a custom image so it survives container recreation:

FROM docker.n8n.io/n8nio/n8n
USER root
RUN apk add --no-cache chromium
USER node

From v0.1.1 the node finds system Chromium automatically (/usr/bin/chromium-browser, /usr/bin/chromium, google-chrome, etc.). You can always point it somewhere explicitly with the Chrome Executable Path option or PUPPETEER_EXECUTABLE_PATH.

Compatibility

  • Requires Node.js ≥ 18.10
  • Tested against n8n 1.x
  • Works on self-hosted n8n (Docker, npm, desktop). n8n Cloud only allows verified community nodes, so this node is self-hosted only for now.

License

MIT

Discussion