# nod (n8n-io/nod)

> A personal agent app that runs on your own computer with its own browser. It handles real errands on real websites, like chasing invoices, prepping calls and cancelling trials, and asks before it acts. It is made by the n8n team.

- GitHub: https://github.com/n8n-io/nod
- GitHub.Locker page: https://github.locker/n8n-io/nod
- Stars: 2.3K
- Language: TypeScript
- License: Apache-2.0
- Last push: 2026-11-17

## Questions and answers

Written by GitHub.Locker from the README and the repository; check the README below for details.

**Is it just a chat app, or does it actually do things?**

It acts. It has its own computer and browser, so it can finish multi-step jobs like logins and forms instead of only replying. You can talk to it in the app, on Telegram, or on WhatsApp.

**How do I get it running?**

Install the app, connect a model provider and your apps, then ask for something in the app, on Telegram, or on WhatsApp. Setup takes about ten minutes.

**Do I need an API key or a paid model subscription?**

You pick a model provider such as OpenAI, Anthropic, Gemini or OpenRouter, or a local server, and that provider bills you directly. You can also sign in with ChatGPT Plus or Pro instead of using an API key.

**Does it work when my laptop is closed?**

Not yet. Scheduled tasks only run while the computer running Nod is on, so a Mac mini or your own server keeps it always on. A hosted version is planned but not released.

**Which computers does it run on?**

The desktop app targets macOS 14 or later, and the README mentions Apple silicon. Running from source works on macOS or Linux with Node.js 24.21 or later, pnpm and Docker.

**Is it really free?**

Yes, the app is free and open source under Apache 2.0. You pay only your model provider, and with API models Nod shows what each task cost and stops at your monthly limit.

**Do I need n8n to use Nod?**

No. Nod is its own app made by the team behind n8n, and it is not the n8n AI Assistant.

**Does it remember things about me?**

Yes. It keeps notes learned from your messages, and you can see, edit or forget each one along with the message it came from. It also learns your writing style from sent emails, without names or numbers.

**Does it ask before doing things for me?**

Yes. The server checks each app action and risky website step against your request and rules, and anything they do not cover waits for your approval. Passwords, codes and cards are entered privately through Vault.

**Which model should I pick for safety?**

In 240 test decisions GPT-6.1 Sol and Claude Opus 5.5 got every one right, while Claude Haiku 5.5 let 8 unsafe actions through. The README recommends a capable model.


---

## README (checked against GitHub 2026-11-18)

Source: https://github.com/n8n-io/nod/blob/main/README.md

<p align="center">
  <a href="https://nod.n8n.io#gh-dark-mode-only"><img src="docs/src/assets/readme/hero-dark.webp" alt="n8n presents Nod: open source personal agent with its own computer" width="100%" /></a>
  <a href="https://nod.n8n.io#gh-light-mode-only"><img src="docs/src/assets/readme/hero.webp" alt="n8n presents Nod: open source personal agent with its own computer" width="100%" /></a>
</p>

<p align="center">
  <strong>It chases invoices, preps your calls and cancels trials, on real websites.</strong><br />
  It runs on your machine with the model you choose, and asks before it acts for you.
</p>

<p align="center">
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-arm64.dmg#gh-dark-mode-only"><img src="docs/src/assets/readme/button-mac-dark.webp" alt="Download for Mac" width="219" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-arm64.dmg#gh-light-mode-only"><img src="docs/src/assets/readme/button-mac.webp" alt="Download for Mac" width="219" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-Setup-x64.exe#gh-dark-mode-only"><img src="docs/src/assets/readme/button-windows-dark.webp" alt="Download for Windows" width="259" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-Setup-x64.exe#gh-light-mode-only"><img src="docs/src/assets/readme/button-windows.webp" alt="Download for Windows" width="259" /></a>
  <a href="https://nod.n8n.io/docs/#gh-dark-mode-only"><img src="docs/src/assets/readme/button-docs-dark.webp" alt="Read the docs" width="190" /></a>
  <a href="https://nod.n8n.io/docs/#gh-light-mode-only"><img src="docs/src/assets/readme/button-docs.webp" alt="Read the docs" width="190" /></a>
</p>

<p align="center">
  <sub>macOS 14+ (Apple silicon or <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-x64.dmg">Intel</a>) · Windows 11 preview · <a href="https://nod.n8n.io/docs/get-started/install-desktop/#linux">Linux x64 preview</a> · Free and open source · ChatGPT Plus or Pro, or an AI key</sub>
</p>

<p align="center">
  <a href="https://github.com/n8n-io/nod/releases/latest"><img alt="Latest release" src="https://img.shields.io/github/v/release/n8n-io/nod?style=flat-square&labelColor=191919&color=ff91ac&label=release" /></a>
  <a href="https://github.com/n8n-io/nod"><img alt="GitHub stars" src="https://img.shields.io/github/stars/n8n-io/nod?style=flat-square&labelColor=191919&color=ff91ac&logo=github" /></a>
  <a href="LICENSE"><img alt="License: Apache 2.0" src="https://img.shields.io/badge/license-Apache_2.0-ff91ac?style=flat-square&labelColor=191919" /></a>
  <a href="https://www.bestpractices.dev/projects/16526"><img alt="OpenSSF Best Practices" src="https://img.shields.io/cii/level/16526?style=flat-square&labelColor=191919&color=ff91ac&label=openssf" /></a>
  <a href="https://discord.gg/n8n"><img alt="Discord" src="https://img.shields.io/badge/discord-join_us-ff91ac?style=flat-square&labelColor=191919&logo=discord&logoColor=white" /></a>
  <img alt="Status: research preview" src="https://img.shields.io/badge/status-research_preview-ff91ac?style=flat-square&labelColor=191919" />
</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-machine-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-machine-dark.webp" alt="On your machine, on your word" width="586" /></a>
  <a href="docs/src/assets/readme/heading-machine.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-machine.webp" alt="On your machine, on your word" width="586" /></a>
</p>

<p align="center">It gets on with the work across Gmail, Calendar, Slack, Notion, and the web,<br />and asks before it speaks for you.</p>

[![Walk into every call prepared. Nod creates a one-page call brief from your messages, calendar, and the client's latest news.](docs/src/assets/readme/call-brief.webp)](https://nod.n8n.io/docs/use/workspace/)

<p align="center">Every morning, Nod reads your threads, your calendar, and the client's latest news,<br />then writes a one-page brief for each call.</p>

[![Waits for your Nod. Nod drafts a Slack reply from your pricing sheet, asks for approval, and sends it after you say yes.](docs/src/assets/readme/approval.webp)](https://nod.n8n.io/docs/trust/approvals/)

<p align="center">It drafts the reply you owe from your own pricing sheet,<br />and sends it only after you say yes.</p>

[![Hear about it when it matters. A scheduled price check reports a change from $1,200 to $1,500.](docs/src/assets/readme/price-watch.webp)](https://nod.n8n.io/docs/use/schedules/)

<p align="center">It checks a page on your schedule and tells you only when something changes.</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-hard-ones-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-hard-ones-dark.webp" alt="Hand it the hard ones" width="421" /></a>
  <a href="docs/src/assets/readme/heading-hard-ones.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-hard-ones.webp" alt="Hand it the hard ones" width="421" /></a>
</p>

<p align="center">Jobs that eat an afternoon of tabs, logins, and forms.<br />Nod has its own computer and browser, so it finishes them.</p>

<p align="center">
  <img src="docs/src/assets/readme/use-cases.webp" alt="Six errands Nod handles: chase an invoice until it's paid, get your money back from a subscription, turn client requests into Linear tasks, report your weekly numbers, file month-end paperwork, and book the team offsite." width="100%" />
</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-memory-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-memory-dark.webp" alt="A memory like an elephant" width="508" /></a>
  <a href="docs/src/assets/readme/heading-memory.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-memory.webp" alt="A memory like an elephant" width="508" /></a>
</p>

<p align="center">Say once that Anna, not Bob, handles invoices. It remembers.</p>

<table>
  <tr>
    <td width="50%"><strong>Every note, visible.</strong><br />See, edit, or forget each one, with the message it came from.</td>
    <td width="50%"><strong>Drafts that sound like you.</strong><br />Learns your style from sent emails, with no names or numbers.</td>
  </tr>
  <tr>
    <td><strong>Tell it once.</strong><br />Say something changed, and it uses the new fact from then on.</td>
    <td><strong>It learns how to do a task.</strong><br />Reuses the steps, and fixes them when a site changes.</td>
  </tr>
</table>

<p align="center">
  It learns about you only from your own messages, never from the pages or emails it reads.<br />
  <a href="https://nod.n8n.io/docs/how-it-works/memory/">How memory works</a>
</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-rules-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-rules-dark.webp" alt="Your machine. Your model. Your rules." width="688" /></a>
  <a href="docs/src/assets/readme/heading-rules.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-rules.webp" alt="Your machine. Your model. Your rules." width="688" /></a>
</p>

- **Runs on your computer or server.** Files, memory, and saved logins stay there. Your model provider gets what each task needs.
- **Uses the model you choose.** OpenAI, Anthropic, Gemini, OpenRouter, Bedrock, Vertex AI, Azure, 20+ more, or your own server. Or sign in with ChatGPT Plus or Pro.
- **Follows your rules.** The server checks each app action, and each website step that can change something, against your request and rules. Anything they don't cover waits for your yes.
- **Keeps secrets out of the model.** Vault enters passwords, codes, and cards privately on the website.
- **Shows what it did.** Every action and sign-in is in Usage, with costs and a monthly limit on API models.

<p align="center">
  <a href="https://nod.n8n.io/docs/trust/">How trust works</a> · <a href="https://nod.n8n.io/docs/trust/privacy/">Where your data goes</a> · <a href="https://nod.n8n.io/docs/trust/security-model/">Security model</a>
</p>

<br />

<a name="benchmarks"></a>

<p align="center">
  <a href="docs/src/assets/readme/heading-measured-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-measured-dark.webp" alt="Measured, not promised" width="476" /></a>
  <a href="docs/src/assets/readme/heading-measured.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-measured.webp" alt="Measured, not promised" width="476" /></a>
</p>

<p align="center">We give Nod real errands in test apps and websites, then check what actually changed.</p>

<table align="center">
  <tr>
    <td align="center" valign="top" width="25%"><h1>98%</h1><strong>tasks finished</strong><br /><sub>226 errands, 3 runs each</sub></td>
    <td align="center" valign="top" width="25%"><h1>0</h1><strong>unsafe actions</strong><br /><sub>in those 678 runs</sub></td>
    <td align="center" valign="top" width="25%"><h1>93.2%</h1><strong>memory recall</strong><br /><sub>LongMemEval-S</sub></td>
    <td align="center" valign="top" width="25%"><h1>0 of 216</h1><strong>false memories</strong><br /><sub>WhisperBench attacks</sub></td>
  </tr>
</table>

<p align="center">
  <a href="https://nod.n8n.io/docs/how-it-works/benchmarks/">How we measured this</a> · <a href="https://nod.n8n.io/docs/contribute/evals/">Run the evals yourself</a><br />
  <sub>GPT-6.1 Sol, October 2026, in simulated accounts and websites. Test results, not a guarantee.</sub>
</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-hood-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-hood-dark.webp" alt="Under the hood" width="328" /></a>
  <a href="docs/src/assets/readme/heading-hood.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-hood.webp" alt="Under the hood" width="328" /></a>
</p>

<p align="center">One Node process and a few Docker containers, all on the machine that runs Nod.</p>

```mermaid
flowchart TB
  you(["You<br/>app · Telegram · WhatsApp"]):::pink
  server["Nod server<br/>Pi runtime · trust check<br/>Vault · memory · schedules"]:::pink
  you --> server
  server --> workspace["Linux workspace<br/>sandbox container"]
  server --> browser["Chromium<br/>browser container"]
  server --> db[("PostgreSQL")]
  server -- "your key" --> model["Your model<br/>provider"]
  server --> apps["Your apps<br/>Gmail · Slack · MCP"]
  classDef pink fill:#ff91ac,stroke:#191919,color:#191919
```

- **Policy lives in the server, not the prompt.** The trust check, Vault, payments, and spending limits run in the server.
- **Untrusted by default.** App, web, and browser results reach the model marked as data, not instructions.
- **Keys stay out of the sandbox.** Provider keys never enter the workspace or the browser.
- **Built with** TypeScript on Node.js 24, Fastify, React, PostgreSQL with pg-boss, the [Pi](https://github.com/earendil-works/pi) agent runtime, agent-browser, Composio, MCP, and Electron.

<p align="center">
  <a href="https://nod.n8n.io/docs/how-it-works/architecture/">Architecture</a> · <a href="https://nod.n8n.io/docs/how-it-works/agent-runtime/">Agent runtime</a> · <a href="https://nod.n8n.io/docs/trust/security-model/">Security model</a>
</p>

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-next-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-next-dark.webp" alt="Next on the list" width="316" /></a>
  <a href="docs/src/assets/readme/heading-next.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-next.webp" alt="Next on the list" width="316" /></a>
</p>

- **Managed hosting.** An always-on Nod, hosted for you.
- **Make phone calls.** For the errands that still need a voice.
- **Nod for your team.** Share one Nod, with its apps and memory.
- **Make it your own.** Add your own skills and tools without touching the core.

<br />

<p align="center">
  <a href="docs/src/assets/readme/heading-answers-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-answers-dark.webp" alt="Straight answers" width="355" /></a>
  <a href="docs/src/assets/readme/heading-answers.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-answers.webp" alt="Straight answers" width="355" /></a>
</p>

<details>
<summary><strong>What does it cost?</strong></summary>
<br />

Nod is free and open source. You bring your own model provider, which bills you directly. With an API model selected, Nod [shows what each task cost](https://nod.n8n.io/docs/use/spending/) and stops at your monthly limit.

</details>

<details>
<summary><strong>Do I need n8n?</strong></summary>
<br />

No. Nod is its own app, made by the team behind [n8n](https://n8n.io). It is not the n8n AI Assistant.

</details>

<details>
<summary><strong>Where does my data go?</strong></summary>
<br />

Your files, memory, and saved logins stay on the machine running Nod. Your model provider gets what each task needs, and app connections go through a connection service that n8n hosts by default. Nod also shares basic usage data with n8n, on by default for new users. Turn off **Share usage data** in **Settings → Privacy & data** to stop new sends. [Your data and privacy](https://nod.n8n.io/docs/trust/privacy/) lists exactly what is sent.

</details>

<details>
<summary><strong>Does it work when my laptop is closed?</strong></summary>
<br />

Not yet. [Schedules](https://nod.n8n.io/docs/use/schedules/) run while the computer that runs Nod is on. Run it on a [Mac mini](https://nod.n8n.io/docs/deploy/mac-mini/) or [your own server](https://nod.n8n.io/docs/deploy/self-host/) to keep it always on. A hosted version is next on the list.

</details>

<details>
<summary><strong>Can it make mistakes?</strong></summary>
<br />

Yes. Its [safety check](https://nod.n8n.io/docs/trust/) is a model's judgement, and Nod is a research preview. That is why it asks before anything you didn't ask for, and why everything it did is on record.

Pick a capable model. In 240 test decisions, GPT-6.1 Sol and Claude Opus 5.5 got every one right. Claude Haiku 5.5 let 8 unsafe actions through.

</details>

<br />

<a name="get-nod"></a>

<p align="center">
  <a href="docs/src/assets/readme/heading-get-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-get-dark.webp" alt="Get Nod" width="209" /></a>
  <a href="docs/src/assets/readme/heading-get.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-get.webp" alt="Get Nod" width="209" /></a>
</p>

<p align="center">Install the app, connect a model and your apps, then ask for something<br />in the app, on Telegram, or on WhatsApp. Setup takes about ten minutes.</p>

<p align="center">
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-arm64.dmg#gh-dark-mode-only"><img src="docs/src/assets/readme/button-mac-dark.webp" alt="Download for Mac" width="219" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-arm64.dmg#gh-light-mode-only"><img src="docs/src/assets/readme/button-mac.webp" alt="Download for Mac" width="219" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-Setup-x64.exe#gh-dark-mode-only"><img src="docs/src/assets/readme/button-windows-dark.webp" alt="Download for Windows" width="259" /></a>
  <a href="https://github.com/n8n-io/nod/releases/latest/download/nod-Setup-x64.exe#gh-light-mode-only"><img src="docs/src/assets/readme/button-windows.webp" alt="Download for Windows" width="259" /></a>
  <a href="https://nod.n8n.io/docs/#gh-dark-mode-only"><img src="docs/src/assets/readme/button-docs-dark.webp" alt="Read the docs" width="190" /></a>
  <a href="https://nod.n8n.io/docs/#gh-light-mode-only"><img src="docs/src/assets/readme/button-docs.webp" alt="Read the docs" width="190" /></a>
</p>

<p align="center">
  <a href="https://nod.n8n.io/docs/get-started/install-desktop/">Install guide</a> · <a href="https://nod.n8n.io/docs/connect/models/">Connect a model</a> · <a href="https://nod.n8n.io/docs/connect/apps/">Connect your apps</a> · <a href="https://nod.n8n.io/docs/deploy/self-host/">Host your own server</a>
</p>

<a name="run-from-source"></a>

<details>
<summary><strong>Run from source</strong></summary>
<br />

On macOS or Linux, install Node.js 24.21 or later, pnpm, and Docker. Then run:

```sh
git clone https://github.com/n8n-io/nod.git
cd nod
corepack enable
pnpm start
```

See [Install from source](https://nod.n8n.io/docs/get-started/install-from-source/) for the Linux Vault key and access link, and [`docs/`](docs/README.md) for the docs of this revision.

</details>

<br />

<a name="build-with-us"></a>

<p align="center">
  <a href="docs/src/assets/readme/heading-build-dark.webp#gh-dark-mode-only"><img src="docs/src/assets/readme/heading-build-dark.webp" alt="Build with us" width="280" /></a>
  <a href="docs/src/assets/readme/heading-build.webp#gh-light-mode-only"><img src="docs/src/assets/readme/heading-build.webp" alt="Build with us" width="280" /></a>
</p>

<p align="center">Nod is open source, from the team behind <a href="https://n8n.io">n8n</a>.<br />Read <a href="CONTRIBUTING.md">CONTRIBUTING</a> to send your first change.</p>

<p align="center"><a href="https://nod.n8n.io/docs/contribute/apps/">Creating your own connectors</a><br />Use MCP, an HTTP API, or a CLI. Add an app folder, run the required tests, and submit a PR.<br />The guide includes an AI builder skill, approval requirements, and shareable app pages.<br /><a href="https://nod.n8n.io/docs/connect/community-apps/">Browse custom connections</a></p>

<p align="center">
  <a href="https://github.com/n8n-io/nod/issues/new/choose">Report a bug</a> · <a href="https://github.com/n8n-io/nod/discussions">Share an idea</a> · <a href="https://discord.gg/n8n">Discord</a> · <a href="https://community.n8n.io/c/n8n-labs/48">Forum</a> · <a href="SECURITY.md">Security</a>
</p>

<p align="center">
  <sub>Copyright 2026 n8n GmbH · <a href="LICENSE">Apache License 2.0</a> · <a href="NOTICE">Notice</a><br />Bundled third-party assets keep their own terms: <a href="public/fonts/Geist-OFL.txt">Geist</a>, <a href="public/logos/SOURCES.md">provider logos</a>,<br />and the desktop runtime components listed in the app's licenses/COMPONENTS.md.</sub>
</p>

