Generate SEO-optimized blog content with Gemini, Scrapeless and Pinecone RAG

This workflow contains community nodes that are only compatible with the self-hosted version of n8n.

How it works

This advanced automation builds a fully autonomous SEO blog writer using n8n, Scrapeless, LLMs, and Pinecone vector database. It’s powered by a Retrieval-Augmented Generation (RAG) system that collects high-performing blog content, stores it in a vector store, and then generates new blog posts based on that knowledge—endlessly.

Part 1: Build a Knowledge Base from Popular Blogs

- Scrape existing articles from a well-established writer (in this case, Mark Manson) using the Scrapeless node. - Extract content from blog pages and store it in Pinecone, a powerful vector database that supports similarity search. - Use Gemini Embedding 001 or any other supported embedding model to encode blog content into vectors. - Result: You’ll have a searchable vector store of expert-level content, ready to be used for content generation and intelligent search.

Part 2: SERP Analysis & AI Blog Generation

- Use Scrapeless' SERP node to fetch search results based on your keyword and search intent. - Send the results to an LLM (like Gemini, OpenRouter, or OpenAI) to generate a keyword analysis report in Markdown → then converted to HTML. - Extract long-tail keywords, search intent insights, and content angles from this report. - Feed everything into another LLM with access to your Pinecone-stored knowledge base, and generate a fully SEO-optimized blog post.

Set up steps