StarNet GitHub: Download, Install From Source and Run Locally

Share this post

StarNet GitHub is where the free pixel-art AI agent station lives — and with dozens of forks floating around, it’s worth knowing which repo is the real one and how to get it running. Below I walk through the official repository, the downloads, running it from source, and connecting free local models. My full walkthrough of the app is in the video.

Key takeaways

The official repo is github.com/androoAGI/starnet (1.2K+ stars at the time of writing).
Grab installers from the starnet-releases page, or run from source with Node.js.
Run free local models through Ollama on 127.0.0.1:11434.
The code is MIT; the StarNet name, logo and art are not licensed for forks.

The official StarNet GitHub repository

Search GitHub and you’ll find lots of copies of StarNet — they’re forks. The original is androoAGI/starnet, and the official site is starnetos.com. Install from the original or its releases page, not a random fork, because a fork can contain anything.

StarNet at a glance
What it is	A local-first desktop agent harness with a pixel-art space station interface
Price	Free — MIT-licensed open source (code only)
Platforms	Windows 10/11 (64-bit) and macOS (Apple Silicon and Intel)
Models	OpenRouter key, sign-in with Anthropic, OpenAI or Google accounts, or local models through Ollama
Official site	starnetos.com
Source code	github.com/androoAGI/starnet

How to install StarNet from GitHub

The easy way: download the installer for your platform from the androoAGI/starnet-releases releases page — Windows 10/11 64-bit or macOS (Apple Silicon or Intel).

From source: you need Node.js 18+ and Git. Then run:

git clone https://github.com/androoAGI/starnet.git
cd starnet
node sidecar/index.js

Then open http://localhost:8787 in your browser.

Building the desktop app yourself also needs Rust and the Tauri prerequisites:

npm ci
npm run desktop:dev
npm run desktop:build

Want this working in your business, not just bookmarked? Running open-source AI agents on your own machine is exactly what we build together inside the AI Profit Boardroom — 3,400+ members, four live calls a week, and a full one-hour DeepSeek Harness course in the classroom.

Prefer it mapped 1-on-1 first? Book a free strategy session and we’ll plan it for your exact situation.

Join the AI Profit Boardroom →Book a Free Strategy Session →

Running StarNet with free local models

StarNet supports Ollama out of the box. Install Ollama, pull a model — the README uses ollama pull qwen3:8b — and choose the OLLAMA provider when you connect your agent’s brain. StarNet talks to Ollama on 127.0.0.1:11434, so everything stays on your machine.

Prefer a cloud model? Paste an OpenRouter key, or sign in with Anthropic, OpenAI or Google. Keys are stored by your operating system.

What’s in the StarNet GitHub project

Concurrent multi-agent runs, each with its own workspace, transcript and memory
Task Briefs, recipes, reusable skills and cron scheduling
Telegram, Discord, Slack, Signal and Matrix integrations
A Night Shift mode for running work while you’re away
Real file deliverables in an OUTBOX
MCP connector support, voice push-to-talk, and spend ledgers with budgets

Spend ledgers and budgets matter: every agent run makes real model calls, so set a budget before you summon a big crew.

StarNet licence and safety tips

The code is MIT-licensed, so you can use and modify it freely. The StarNet branding, logo and artwork are not — forks can’t reuse them. Because agents can get access to the web and your local files, start with “ask for approval” mode and only grant the permissions each agent needs.

If you’re reporting a bug or requesting a feature, the GitHub issues tab is the place — and it’s worth reading open issues before you install, to see what’s currently broken.

StarNet GitHub troubleshooting

localhost:8787 won’t load — check the sidecar is still running in your terminal and that nothing else is using port 8787.
Node errors on start — confirm node -v shows 18 or higher.
Ollama models don’t appear — make sure Ollama is running and you’ve pulled a model, then pick the OLLAMA provider again.
Desktop build fails — install Rust and the Tauri prerequisites for your platform first.
On Linux — it isn’t a supported public release target, so expect rough edges if you build it yourself.

If you’re stuck, search the repo’s issues for your error message before opening a new one — someone has often hit it already.

Release build or source build?

Option	Pros	Cons
Installer from starnet-releases	One click, signed builds, automatic updates path	Less control
Run from source (node sidecar)	See the code, try the latest changes, runs in the browser	Needs Node and Git; you update it yourself
Build the desktop app yourself	Full control over the build	Needs Rust and Tauri; slowest to set up

For most people the installer is the right call. Run from source only if you want to read the code or contribute. Either way, check you’re on the original androoAGI repo, not a fork.

What to build with StarNet once it’s running

Installing it is the easy part; the value comes from giving your agents real, repeatable jobs. Three I’d start with:

A daily brief agent — connect your email and calendar, and have it summarise your day every morning.
A research agent — give it web access and a niche, and let it gather the week’s news into a file in the OUTBOX.
A content crew — summon a researcher and a marketer, and hand off from one to the other along a hallway.

Use Night Shift for jobs that can run while you sleep, and cron scheduling for anything that repeats. Keep each agent’s permissions tight, and check the spend ledger after the first week so you know what your setup actually costs.

Getting started after StarNet GitHub

Once it’s installed, my StarNet AI agent guide walks through creating your first agent, adding apps and skills, and recruiting a crew. Start with the official StarNet GitHub repo, run a local model, and you’ve got a free agent station on your own machine.

A quick word on forks: GitHub shows many StarNet repos with identical descriptions because forking copies the README. That’s normal for a popular open-source project, but it means the description alone tells you nothing. Check the owner name, look for the “forked from” line under the repo title, and compare the latest commit date with the original. Downloading the official releases is the simplest way to be sure you’re running the real thing, and starring the original repo is the best way to follow new releases as they ship.

FAQ: starnet github

Where is the official StarNet GitHub repo?

github.com/androoAGI/starnet. Most other StarNet repos are forks of it.

How do I install StarNet from GitHub?

Download an installer from the starnet-releases page, or clone the repo and run node sidecar/index.js, then open localhost:8787.

What do I need to run StarNet from source?

Node.js 18+ and Git; building the desktop app also needs Rust and the Tauri prerequisites.

Can StarNet use local models?

Yes — through Ollama on 127.0.0.1:11434.

What licence is StarNet under?

MIT for the code. The StarNet name, logo and artwork are not licensed for forks.

Two ways I can help from here. Join the AI Profit Boardroom — it’s where your open-source agent setup gets built with 3,400+ members doing the same.

Or grab a free strategy session and bring your questions.

Join the AI Profit Boardroom →Book a Free Strategy Session →

About Julian Goldie

I’m Julian Goldie — founder of Goldie Agency, best-selling author, and an AI educator with 428K+ YouTube subscribers. I’ve spent 10+ years in SEO and online business and I test every AI agent on my own work before I recommend it. Join the AI Profit Boardroom for the daily builds, or book a free strategy session to talk through yours.