Agent Upstart

Cool things people are building with agents — and practical help getting the most out of them.

Playbooks · Everyday tasks

Nod: n8n's open-source personal agent

Written November 12, 2026

Nod is a personal agent that lives on your own computer, with its own browser, and finishes errands instead of handing you a to-do list: cancelling a subscription, gathering invoices, watching a price. It is free and open source, you bring your own model provider, and it asks before it does anything you did not ask for.

What it is, and what it is not

n8n's announcement is dated November 10, 2026 and releases Nod as a research preview from n8n Labs, a team inside n8n that works on ideas outside the core product. Nod is a standalone app, separate from n8n and from the agents you can build inside it. You do not need an n8n account, and there is no server or cloud service to set up first: you install it on your computer, connect your accounts, and ask. If you do use n8n, you can connect your instance so Nod can work with your workflows.

The jobs it takes

Nod works inside a private virtual machine on your computer, with its own browser, in connected apps such as Gmail, Calendar, Drive and Slack — and on websites that have no integration at all. The jobs n8n lists: cancel a subscription, work through the retention offers and ask for a refund; watch a price and speak up when it drops; find a hotel that fits a trip and book it once you say go; sign in to supplier portals each month, download the invoices and file them for your accountant; pull figures from a dashboard with no export and put them in a weekly report. Jobs can repeat on a schedule, several can run in parallel in the background, and you talk to Nod in one conversation, in the app or over Telegram or WhatsApp. It also reads your connected apps and offers jobs it could take over, rather than waiting for you to think of them.

Your provider, your bill

Nod is free and open source under Apache 2.0, and every rule it follows is in the code, in the public repository. You connect your own AI provider, with an API key or a ChatGPT plan, or run an open-source model on your own computer. You pay that provider for the AI you use. n8n is candid that some early testers found tasks cost more than expected, and the cost controls are concrete: every task shows what it cost and ends with a receipt, you can set a monthly spending limit, Nod asks once a task passes $5 (a figure you can change), and each schedule shows its estimated monthly cost.

The permission design

An agent that can send mail and spend money needs its refusals designed, not hoped for. Before Nod sends, pays, posts or deletes anything you did not ask for, a separate check confirms the action against your own words — a web page or an email cannot give Nod orders. Saved logins and cards are encrypted on your machine, and Nod enters them through private forms the model cannot read. Its data, memory and logins stay on the machine running it; your provider gets what each task needs. Every sign-in, card use and action is logged. n8n's own caveat belongs in the same paragraph: the permission check is itself a model and can make mistakes, which is why Nod asks when your words do not cover an action.

Limits

Related playbooks

Sources

Back to: Everyday tasks playbooks hub · Also: Cloakroom: hide your secrets from your agent · Infor's connector lets Claude act, with approval