Heidelberg AICurriculum

100% AUTOMATED Blogging with AI Agents in n8n!

AI Foundations ·2026-01-14 ·2 min read

Summary written by us from the video's transcript. The video, and everything in it, is AI Foundations's work.

Learn how to set up a fully automated, human‑in‑the‑loop blogging workflow in n8n that generates SEO‑optimized posts using AI agents.

Takeaways

  • The workflow combines AI creativity with human oversight, letting you control topics, goals, and publishing actions.
  • Lerty provides real‑time status cues (researching, outlining) so you don’t need to monitor execution logs.
  • Section mapping ensures each blog segment stays within optimal word ranges for SEO and reader engagement.
  • Custom button payloads enable branching logic—revise, publish, or archive—without rewriting the core workflow.
  • A single OpenRouter API key lets you experiment with different language models for each step.

Overview of the Automated Blogging System

The workflow lets you input basic details about a blog—topic, audience, goal, CTA, and word count—and then runs through idea generation, research, outlining, writing, keyword optimization, and formatting without manual effort. It includes a human‑in‑the‑loop step for approval or adjustments before the content is finalized.

Key Tools Used

Three services power the workflow: **n8n** for orchestration, **Lerty** as a communication layer where agents are chatted with and status indicators appear, and **OpenRouter** to access any AI model via a single API key.

Configuring Company Context

An *end‑user config* node stores static information such as company name, description, brand voice, industry, and target keywords. All downstream agents read this context so the generated posts match the client’s tone and market.

Human‑in‑the‑Loop Interaction

The process starts with a slash command (e.g., “/start”) that triggers a human‑in‑the‑loop node. You fill out a form with the blog details, then click **Generate Blog**. The workflow pauses until you submit, giving you control over topics and objectives while the AI handles the heavy lifting.

Step‑by‑Step Generation Flow

1. **Brainstorming Assistant** – chat via Lerty to get fresh, industry‑specific blog ideas. 2. **Research Stage** – the workflow indicates “researching” in Lerty while it gathers sources. 3. **Section Mapper** – creates a structured outline with sections sized 250–750 words for optimal reader retention. 4. **Blog Writer Loop** – writes each section sequentially, tracking progress (e.g., “section 4 of 8”). 5. **Keyword Researcher** – inserts high‑target SEO keywords. 6. **Formatting** – assembles the final post and returns it in chat.

Extending with Additional Human Actions

After the post is generated you can present buttons such as **Revise**, **Send to WordPress**, or **Archive**. Each button sends a custom payload that a switch node can route to different sub‑workflows (e.g., a revision loop or publishing step). This makes the system adaptable to any publishing platform or review process.

Getting Started Quickly

A free template with all nodes pre‑configured is linked in the video description. Import it into n8n, adjust the system instructions and company config, sign up for a free Lerty account, and you can start generating blogs immediately.