TL;DR:
I’ve forked and extended @coleam00’s excellent local-ai-packaged to build a VPS-optimized, beginner-friendly installer for n8n and a full AI/automation stack.
This version focuses on self-hosted deployment, automatic HTTPS, and an interactive setup wizard — ideal for spinning up a private LLM-powered environment on any cloud server.
GitHub – kossakovsky/n8n-installer
What it includes (and sets up)
The installer launches a complete, containerized automation and LLM environment, built specifically for VPS use:
| Service | Purpose |
|---|---|
n8n |
Workflow automation (runs in queue mode, multi-worker) |
| Supabase | Database, auth, storage, REST API, and vector support |
| Flowise | Visual builder for LLM chains and agents |
| Open WebUI | Chat-style interface for LLMs |
| Qdrant | High-performance vector DB |
| Weaviate | Scalable vector search engine |
| Langfuse | LLM observability and feedback tracking |
| Letta | Lightweight orchestrator for local LLM flows |
| Crawl4ai | Web crawler for RAG pipelines |
| Neo4j | Graph database (optional) |
| Redis, Postgres | Core infra for queues and persistence |
| Grafana | Monitoring dashboards |
| Prometheus | Metrics backend |
| Caddy | Automatic HTTPS reverse proxy |
| SearXNG | Private metasearch engine |
All services run in a secure private Docker network and are routed via HTTPS. You choose what to install — it’s modular.
Who is this for?
AI builders deploying to cloud servers
Devs working with GPT, embeddings, vector pipelines
Indie hackers launching MVPs with agents and automations
Self-hosters who want full control
Anyone tired of maintaining 12 different compose files
This stack is built for VPS/server use only — not intended for local/desktop environments.
How to install
Prerequisite:
You must own a registered domain (e.g., yourdomain.com) and configure a DNS A-record (or wildcard *.yourdomain.com) pointing to your VPS IP address before running the installer.
SSH into your VPS and run:
git clone https://github.com/kossakovsky/n8n-installer && cd n8n-installer && bash install.sh
You’ll be asked for your domain and a few other basics. The wizard will auto-generate secure secrets, configure services, and set up HTTPS.
Installation time: 5–10 minutes, depending on your server specs.
Keep in mind that DNS changes may take time to propagate.
Depending on your domain registrar and DNS provider, it can take anywhere from a few minutes to up to 24 hours for your new A-record to become active globally.
How to update
To pull the latest version and refresh services:
bash update.sh
Any new components will appear in the wizard automatically.
Bonus: 300+ ready-to-import workflows
You can optionally import 300+ real-world n8n workflows during setup — helpful if you’re exploring or just want to save time.
Technical details
VPS requirements (RAM, CPU, disk), OS recommendations, volume mappings, and other setup notes are listed in the GitHub README.
Credits
This project builds on the excellent work by @coleam00.
Huge thanks to Cole Medin for creating and maintaining local-ai-packaged.
My fork focuses on:
- VPS-first setup experience
- Interactive service wizard
- Auto HTTPS with Caddy
- Simplified updates and config
- 300+ n8n workflows available during install
GitHub
https://github.com/kossakovsky/n8n-installer
PRs and feedback are welcome!
Final thoughts
This project started as a shortcut for myself and turned into a reusable, production-ready stack for anyone building AI-powered automations with n8n.
Let me know what you think — and if it helps you, a 
on GitHub is always appreciated 