Resources · n8n integration

n8n Wikipedia nodeConfigure Wikipedia in n8n.

The n8n Wikipedia node hands an AI agent a lookup tool it can call on its own. It ships inside the n8n AI pack, exposes 0 parameters, and wires onto a single port. Drop it next to an agent and it pulls article content mid-conversation, with nothing to configure.

Try the Wikipedia node on n8n

Verified Trustpilot reviews · AI, automation & growth agency

Why automate

What does the n8n Wikipedia node do in a workflow?

Wikipedia is a tool sub-node, meaning a small block that an AI agent calls by itself when it needs a fact. It never sits in the main path of a workflow. You drop it on the canvas and wire it to the Tool port of a root node, the small connector that hangs under the agent. From there the agent decides when a question deserves a lookup, runs the search, and reads the article back into its own answer.

Take a support assistant answering questions in chat. The model knows your product, but it stumbles the moment someone asks about a standard, a company or a public figure it half remembers. Wire Wikipedia to the agent and that gap closes: the agent searches, quotes what it found, and stops improvising. You change nothing in the node itself, because there is nothing in it to change.

Second case, content drafting. A workflow takes a few keywords, asks the model to gather background, then writes a first draft. Without a lookup tool the background is whatever the model remembers. With one, the agent pulls article text first and drafts from it, which makes the output far easier to fact check afterwards.

Third case, a multi tool agent. Wikipedia rarely travels alone: it sits next to the tools that actually act, and the agent picks whichever fits the question. That is the point of the AI Agent node, and of AI Agent Tool when a whole agent becomes a tool for another agent. The chat model driving the choice comes from a provider node such as OpenAI or Anthropic.

When should you skip it? Whenever the model has no decision to make. If you already know the exact article you want, an agent adds a step and a bill for nothing: fetch the content in a regular workflow node and put it straight in the prompt. A tool earns its place when the choice of looking something up, and of what to look up, belongs to the model.

The limits follow from the same simplicity. The node is at version 1 and its panel is empty, so there is no field to narrow a search, pick a language edition or cap the length of what comes back. It searches Wikipedia and nothing else. And like every sub-node, it never runs alone: unplugged from a root node, it simply sits there. If the trade off between a tidy node and a configurable one matters to you, the n8n review covers how the platform handles that tension elsewhere.

Connections

What do you wire the Wikipedia node to?

n8n splits AI work in two: a root node receives the workflow items through its main input, and sub-nodes clip onto its ports, one type of sub-node per port.

Output (what it plugs into)

  • Toolai_tool
01ToolThis is the only connection the node has, and it points outward: Wikipedia plugs into the Tool port of a root node, which then exposes the search to its model as something it may call.

This is the only connection the node has, and it points outward: Wikipedia plugs into the Tool port of a root node, which then exposes the search to its model as something it may call.

Key parameters

  • Required: a sub-node has no main input and never executes on its own, so an unwired Wikipedia node does nothing at all.
  • AI Agent: the usual host, an agent that reasons over a question and picks its tools.
  • AI Agent Tool: for nesting, when a full agent is itself offered as a tool.
  • MCP Server Trigger: exposes the tool to an outside MCP client instead of a local agent.
Use cases
a chat assistant that answers general knowledge questions without leaving the conversation, or an agent shared over MCP so several clients reuse the same lookup.
Parameters

Wikipedia

The Wikipedia node has 0 parameters. For each one: the node as you configure it in n8n, what the parameter changes, and our field notes.

Need help

Need help automating Wikipedia with n8n?

A person reads every message.

FAQ

Wikipedia and n8n, the common questions

01Is the Wikipedia node free in n8n?
Yes. It belongs to the n8n AI pack, which ships with n8n itself, so there is nothing to install and no extra cost on the n8n side. That holds on n8n Cloud, the hosted offer run by n8n, and on a self-hosted instance under the Community Edition and its Sustainable Use license. One cost does exist, but it sits elsewhere: the agent driving the tool needs a chat model, and the model provider bills its own API calls under its own terms. n8n adds nothing on top of that. The search itself costs you nothing to run.
02What do you need to make the n8n Wikipedia node work?
Nothing, and that is worth stating plainly. The node has no credential, no Authentication selector and no parameters at all, so there is no account to connect and no panel to fill in. Setup mistakes on this node are impossible because there is no setting to get wrong. What it does need is context: a root node such as an AI Agent to host it, wired through the Tool port, and a chat model sub-node attached to that agent. The API key lives on the model sub-node, not here. Drop it, wire it, done.
03What are the limits of the Wikipedia node?
The node sits at version 1 and its panel is empty, which cuts both ways. Nothing to configure also means nothing to tune: no field narrows the search, selects a language edition or limits how much text comes back. What the agent gets is what the search returns. The node also does one job only, searching Wikipedia, so anything outside that encyclopedia needs a different tool. Finally it never runs alone. A Wikipedia node left unconnected on the canvas is inert, since a sub-node has no main input and only executes when a root node calls it.
04What is the minimum to wire, and which sub-node should you start with?
Three pieces. An AI Agent as the root node, a chat model sub-node clipped to it so the agent can reason, and Wikipedia on the Tool port. Start with a single tool rather than a shelf of them: an agent given one clear option picks it reliably, and you can tell from the execution log whether the search fired. Add the rest once that works. For the chat model, pick whichever provider you already hold a key for, since the key lives on that sub-node. Then extend the same agent with the tools that act on your own systems.
05n8n or Make for an AI agent with tools?
It comes down to hosting, data and cost shape rather than features. Make is hosted only, with no self-hosting option, and bills per operation, so a chatty agent that calls a tool several times per conversation makes its own volume. n8n runs on Cloud too, but it can also live on your own Docker or npm install, which keeps the conversation data and the model keys on infrastructure you control. Workflows behave the same either way. If self-hosting is not on the table and per-operation pricing suits your volume, Make stays a fair choice.
Hack'celeration Lab

Get our weekly integration tips.

No spam. Unsubscribe anytime.

Same platform

Other docs you might need on this stack.

Browse all integrations