n8n data, guides and pitfalls.
What I learn building n8n automations with AI agents, written down so it can be used without talking to me: weekly figures, guides that answer one question each, and a list of known pitfalls.
Your first AI agent in n8n
A self-paced online course: you build an AI agent in n8n together with an AI assistant (Claude or ChatGPT). We give you a practice n8n for 14 days, so there is nothing to install.
Sign up for the course- Narrated lesson and step-by-step exercises
- Ready n8n with access to an AI model for 14 days
- Automatic exercise check and a certificate
n8n AI Automation Index
Eight questions about how AI agents build n8n workflows, computed weekly from n8n-mcp telemetry and a corpus of nearly a million workflows.
- Which LLM powers n8n AI agents: OpenAI, Anthropic, Gemini or Ollama?
- Which n8n nodes do AI agents use most, and which trigger nodes start their workflows?
- Which n8n node is growing fastest?
- Do AI-built n8n workflows handle errors?
- Do AI agents build n8n workflows from nodes or from code?
- How large are AI-built n8n workflows, and which fail validation?
- Which n8n nodes do AI agents search for?
- How do AI agents (Claude Code, Cursor) build n8n workflows through MCP: build or maintain?
Guides
One question, a full answer, updated instead of writing another post.
What is the n8n AI Agent and how does it work?
The n8n AI Agent is a node in which a language model receives instructions (a system message) and a set of tools, then decides on its own which tools to call to complete a task. Without tools it is just a chat. With tools, most often sub-workflows, the agent can read a document, check data in a spreadsheet and make a judgement call no plain rule can handle. Since August 2026 n8n also has a separate Agents category: agents with their own memory, channels and skills that live outside a single workflow.
Read the guideHow do you connect Claude Code to n8n through MCP, and what can the agent do then?
Claude Code connects to n8n through the n8n-mcp MCP server: one terminal command adds the server, and the agent gets documentation for every n8n node, configuration validation and workflow operations on your instance. From then on it can build, fix and test a workflow from a description in plain language. More than 190,000 people use n8n-mcp, and its telemetry shows agents spend four times more calls reading and checking than writing.
Read the guiden8n self-hosted or cloud: what does it cost and which should you choose?
n8n Cloud starts at €20 a month for 2,500 workflow executions, and the price grows with executions. Self-hosted n8n in the free Community Edition has no execution limit, costs what the server and maintenance time cost, and keeps data inside the company. For companies handling documents, personal data or larger volumes I recommend your own server. Cloud makes sense at the start, when nobody in the company wants to run a server.
Read the guideHow do you install n8n-skills, and why does an AI agent need skills for n8n?
n8n-skills is a set of 14 skills plus a skill router and a hooks layer that teach an AI agent the right patterns for building n8n workflows through the n8n-mcp server. You install it with one command in Claude Code, as zip files in Claude.ai and ChatGPT, or as a plugin in Codex. An agent without skills guesses at syntax and puts instructions in the wrong place; an agent with skills sticks to proven patterns and conventions.
Read the guideWhy is my n8n workflow not working, and how do I find out?
Most n8n failures fall into three groups: an agent or client can't connect to the instance, a node that used to work suddenly breaks, or the workflow goes silent and nobody finds out. Executions shows you what happened, but not what never ran. Without a separate error workflow, silence is the default state: nobody checks executions every day, so a failure waits until someone happens to open the tab.
Read the guiden8n vs Zapier vs Make: which should a company choose for automation?
Zapier counts tasks per successful step, Make counts operations per module, n8n counts executions regardless of how many steps a workflow has. That billing unit matters more to the bill than the integration list on the homepage. Zapier and Make win on onboarding speed and the number of ready integrations, n8n wins on volume, data control, and code where the ready-made blocks run out.
Read the guideWhere do you find n8n workflow templates, and how do you use them with an AI agent?
n8n.io publishes more than 12,000 workflow templates, and n8n-mcp gives AI agents access to 2,352 of them through search_templates, get_template and n8n_deploy_template. Telemetry shows agents reach for a template rarely, read documentation and build from scratch far more often, then revise the result about nine times. This guide covers when a template genuinely saves time and what you have to check before you activate one on your own instance.
Read the guideVoicebot, voice agent or AI receptionist: which to choose and what does it cost?
An off-the-shelf AI receptionist starts at $29 a month and can be live in a few hours. A custom voice agent on GPT-Live with Twilio telephony runs at about $0.065 per minute plus roughly $14 in fixed monthly fees, but someone has to build it. Below about 50 calls a month the ready-made service wins; above 500 a custom agent pays for itself within months. In both cases bookings, CRM lookups and quotes run in n8n, which the agent calls mid-conversation.
Read the guideHow do you secure self-hosted n8n on Docker?
Self-hosted n8n holds the keys to every system it automates, so secure it like a server full of company passwords. The minimum is ports 80 and 443 only, behind a reverse proxy with HTTPS, an owner account with 2FA, an encryption key backed up off the server, and authentication on every webhook. At the maximum level the editor is reachable only over VPN, and the only traffic from the internet is authenticated webhooks through a gateway with a WAF.
Read the guideHow do you build a company brain for your team's AI agents?
A company brain is memory your team's AI assistants write and read together: each person keeps private notes for their own assistant, while decisions, work in progress and how systems work go into a shared space that everyone's agents search before they start. It connects to ChatGPT, Claude and Codex through one MCP connector and runs on the company's own server. Unlike a knowledge base, it is written by agents during the work, not by people after the fact.
Read the guideMistakes AI agents make in n8n
Known n8n behaviours that n8n-mcp hands to AI agents as context before they build.
- Execute Workflow + Wait nodes cause data loss
- SplitInBatches + Wait nodes cause state loss
- Never use append on sheets with formula columns
- $env expressions blocked in Code nodes
- saveExecutionProgress fills disk space
- IF node uses branch true/false not sourceIndex
- Code node $input.all() vs $input.first()
- Per-item execution on large datasets
- Webhook node response must be sent explicitly
- Error Trigger workflow must be separate
- $getWorkflowStaticData global loses data in loops
Let's talk about AI in your business
Tell Alicja what takes up time in your business. She'll book you a free strategy consultation with me.
Prefer a calendar? Pick a time
or write: romuald@aiadvisors.pl · +48 695 263 884