Generate an SEO PDF report from HTML with Gotenberg and Claude Opus 4.6

What this workflow does

This workflow takes any website URL, extracts its HTML content, and uses an AI Agent (Claude Opus 4.6) to perform a comprehensive SEO analysis. The AI evaluates the page structure, meta tags, heading hierarchy, link profile, image optimization, and more — then generates a beautifully formatted HTML report. Finally, it converts the report into a PDF using Gotenberg, a free and open-source HTML-to-PDF engine.

Workflow steps:

1. Form submission — pass the URL you want to analyze
2. HTML extraction — fetches the full HTML content from the URL
3. AI SEO analysis — Claude Opus 4.6 analyzes the HTML and generates a detailed SEO report in HTML format
4. File conversion — converts the HTML output into a file (index.html) for Gotenberg
5. PDF generation — sends the file to Gotenberg and returns the final PDF

Setup Guide

Gotenberg — Choose one of 3 options:

Option 1 — Demo URL (testing only):
Use https://demo.gotenberg.dev as the URL in the HTTP Request node. This is a public instance with rate limits — do not use in production.

Option 2 — Docker Compose (self-hosted n8n):
Add Gotenberg to the same docker-compose.yml where your n8n service is defined:

services:
  # ... your n8n service ...

  gotenberg:
    image: gotenberg/gotenberg:8
    restart: always

Run docker compose up -d to restart your stack. Gotenberg will be available at http://gotenberg:3000 from inside your n8n container.

Option 3 — Google Cloud Run (n8n Cloud or no Docker access):
Deploy gotenberg/gotenberg:8 as a Cloud Run service via the Google Cloud Console. Set the container port to 3000, memory to 1 GiB, and use the generated URL as your endpoint.

📖 Full Gotenberg docs: gotenberg.dev/docs

AI Model

This workflow uses Claude Opus 4.6 via the Anthropic API. You can swap it for OpenAI, Google, or Ollama — just replace the Chat Model node.

Requirements

• Anthropic API key (or alternative LLM provider)
• Gotenberg instance (demo URL included for quick testing)
• No other external services or paid tools required

Feel free to contact me via LinkedIn if you have any questions! 👋🏻