What hardware do I need?
The kit supports Nvidia GPUs (via Docker profile), Mac/Apple Silicon (CPU-only or with external Ollama), and CPU-only environments for all other setups.
Self-hosted AI starter kit (by the n8n team) MCP Server
@cedricdcc
Basic information
About Self-hosted AI starter kit (by the n8n team) MCP Server
pydantic agent workflow using mcp servers to test out advanced retrieval of linked data
Self-hosted AI starter kit (by the n8n team) MCP Server config
No standard config provided
This server doesn't expose a parseable MCP config block in its README. See the repository for install instructions.
RepositorySelf-hosted AI starter kit (by the n8n team) MCP Server tools
Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Self-hosted AI starter kit (by the n8n team)?
The Self-hosted AI starter kit is an open Docker Compose template that quickly bootstraps a fully featured Local AI and Low Code development environment. It combines the self-hosted n8n platform with curated AI products including Ollama, Open WebUI, Flowise, Qdrant, and PostgreSQL. This version by coleam00 adds Open WebUI and Flowise improvements and includes a pre-configured local RAG AI Agent workflow.
How to use Self-hosted AI starter kit (by the n8n team)?
Clone the repository, navigate to the local-ai-packaged directory, and run docker compose with the appropriate profile (--profile gpu-nvidia, --profile cpu, or without a profile for Mac). Then open http://localhost:5678/ to set up n8n and http://localhost:3000/ to set up Open WebUI. Configure credentials for Ollama, Postgres, Qdrant, and optionally Google Drive, then activate the included workflow and paste the n8n_pipe.py code into Open WebUI’s Functions.
Key features of Self-hosted AI starter kit (by the n8n team)
- Includes self-hosted n8n with over 400 integrations and AI components
- Bundles Ollama for running local LLMs like Llama3.1
- Provides Open WebUI for interacting with models and n8n agents
- Includes Flowise for no/low code AI agent building
- Ships with Qdrant vector store and PostgreSQL database
- Pre-loaded local RAG AI Agent workflow auto-installed on first run
Use cases of Self-hosted AI starter kit (by the n8n team)
- Building and testing self-hosted AI workflows with local language models
- Creating a private ChatGPT-like interface for interacting with n8n agents
- Developing proof-of-concept AI applications that keep data entirely on-premises
- Experimenting with RAG pipelines using local vector stores and LLMs
Frequently asked questions
How do I upgrade the starter kit?
Run `docker compose --profile <profile> pull` followed by `docker compose create && docker compose --profile <profile> up`, using the same profile as your initial installation.
Where are local files stored?
A shared folder is mounted to the n8n container at `/data/shared`. Nodes like Read/Write Files from Disk, Local File Trigger, and Execute Command can access files there.
How do I connect to Ollama from n8n?
For Docker-based setups, use `http://ollama:11434` as the Ollama URL. For Mac users running Ollama natively, use `http://host.docker.internal:11434/`.
Is this production-ready?
The starter kit is designed to help get started with self-hosted AI workflows for proof-of-concept projects and is not fully optimized for production environments, though it uses robust components.
More Developer Tools MCP servers
Foremerge
naw103Catch intent conflicts before code conflicts. The open-source coordination protocol for coding agents, built above Git.
50
Meticulous
Meticulous
LastPing
LastPingKnow when your AI agents, cron jobs and CI pipelines fail, stall or go quiet. Your agent sets up monitors from a plain description, tells you what is broken and why, reads run history and traces, and writes notes on inci
DocBin
DocBinGive an agent a place to put large output. Instead of returning 40k characters of a build log into the conversation, it posts the file to DocBin and hands back one short URL. Docs render with syntax highlighting and mark
discoverGPT
Joe MonastierodiscoverGPT is visualAI's MCP gateway for AI commerce: one connection gives Claude 40 tools spanning product search, data-quality review, AI-discovery feeds, and virtual try-on, all backed by a single canonical product c
Comments