Getting started with n8n's AI nodes: the AI Agent node, chains and memory

Open the node panel of a recent n8n and the AI section can be intimidating: agents, chains, models, memories, parsers, retrievers, tools. The good news is that it all reduces to one mental model. Master it and every AI workflow — including every template we sell — becomes readable at a glance.

The mental model: root nodes and sub-nodes

Classic n8n nodes pass data left to right through main connections. AI nodes add a second dimension: cluster nodes. A root node (Agent, Basic LLM Chain, Q&A Chain) sits in the main flow like any node — but its capabilities plug in underneath, as sub-nodes with typed connections:

• ai_languageModel — the brain: OpenAI Chat Model, Anthropic Chat Model…
• ai_memory — conversation state: Window Buffer Memory…
• ai_tool — things an agent may call: HTTP Request Tool, workflows, vector stores
• ai_outputParser — structure enforcement: Structured Output Parser
• ai_embedding, ai_vectorStore, ai_retriever, ai_document, ai_textSplitter — the RAG family

The key consequence: capabilities are composable and swappable. The model is not welded to the logic; swap the model sub-node and everything else stands.

The three root nodes you'll actually use

Basic LLM Chain — one prompt, one completion, no autonomy. This is the workhorse: classification, summarization, extraction, drafting. Add a Structured Output Parser and set hasOutputParser and the model must return your JSON schema — enums for categories, integers for scores — so Switch and IF nodes downstream can route on solid values.

AI Agent — a model with tools and, usually, memory. The agent reads the conversation, decides whether to call a tool, reads the result, iterates. Use it when the flow genuinely depends on what the user says: guided interviews, assistants that fetch data. Two rules keep agents reliable: narrow tools with excellent descriptions (the description is what the model reads to decide), and a system message that spells out the protocol step by step. Our guided questionnaire bot is a complete worked example — 8 questions, one recording tool, mandatory tool call after each confirmed answer.

Retrieval Q&A Chain — RAG without agent unpredictability: retrieval always happens, then one completion grounded in the retrieved chunks. The right default for "chat with my documents".

The n8n AI Agent node under the hood

Open the JSON export of any agent workflow and you'll find the AI Agent node declared by its full type, @n8n/n8n-nodes-langchain.agent:

{
  "type": "@n8n/n8n-nodes-langchain.agent",
  "typeVersion": 1.9,
  "parameters": {
    "promptType": "auto",
    "hasOutputParser": true,
    "options": {
      "systemMessage": "You are the audit assistant. Ask one question at a time…"
    }
  }
}

The parameters worth knowing before you edit one by hand:

• promptType — auto takes the incoming chatInput field (what the Chat Trigger produces); define lets you write the user prompt yourself with expressions. Agents fed by a webhook or an IMAP trigger almost always need define.
• hasOutputParser — a flag, not a connection. Setting it to true tells the Agent node to expect a parser sub-node on its ai_outputParser port and to append the format instructions to the prompt. The classic import error — "Output parser missing" — means hasOutputParser is true in the JSON but no Structured Output Parser is actually connected underneath. The reverse is silent: a connected parser with hasOutputParser: false is simply ignored, and you get free text where you expected JSON.
• options.systemMessage — the agent's protocol. For anything beyond a demo, write it long: role, tool-usage rules, output language, stop conditions.
• maxIterations — the loop budget (think tool call + reasoning per iteration). The default of 10 is generous; lower it on agents whose tools are expensive.

Since typeVersion 1.7+, the Agent is a Tools Agent by default: it uses the model's native function-calling to pick tools, which is why tool descriptions matter more than anything else you configure. A concrete pairing that shows hasOutputParser and a tool working together: an agent that interviews a user, calls a record_answer tool after each confirmed answer, and returns a final {status, missing_fields[]} JSON via the Structured Output Parser — that's word for word the architecture of our guided questionnaire bot.

Memory, in thirty seconds

LLM calls are stateless; memory sub-nodes replay recent exchanges into the prompt. Window Buffer Memory keeps the last N messages per session — the Chat Trigger supplies the session ID automatically. That's all a questionnaire or support assistant needs. Skip memory entirely for one-shot chains: it costs tokens and adds nothing.

Your first AI workflow, in six steps

A concrete starter — an email summarizer you can build in fifteen minutes:

1. Manual Trigger (you'll wire IMAP later — see automating email triage).
2. Set node: a text field with a pasted email.
3. Basic LLM Chain: prompt = "Summarize in one sentence and list requested actions", pass {{ $json.text }}.
4. OpenAI Chat Model sub-node: gpt-4o-mini, attached via ai_languageModel.
5. Execute, read the output.
6. Add a Structured Output Parser ({summary, actions[]}) and re-run: same call, now machine-readable.

Those six steps contain 80% of production AI automation: prepare clean input, prompt precisely, force structure, route on the result.

Mistakes to skip

• Reaching for an Agent when a Chain suffices — you trade determinism for nothing.
• Sending raw, unbounded input (whole PDFs, full threads) — truncate; tokens are the cost driver.
• Parsing free-text model output with string operations — that's the parser's job.
• One giant system prompt for every case — small, dedicated prompts per task age far better.

From here, the natural next steps are the RAG stack with Supabase or dissecting the 12 documented workflows on this site — each one is a node-by-node annotated example of exactly these patterns.