n8n

n8n is a fair-code licensed, node-based workflow automation platform with native AI capabilities. With over 181K GitHub stars, it is the #1 JavaScript Rising Stars project of 2025, enabling technical teams to combine visual no-code building with custom JavaScript code across 400+ integrations.

automation workflow javascript ai-agents self-hosted low-code

Overview

n8n (pronounced “n-eight-n”, short for “nodemation”) was launched in 2019 and has grown into the leading open-source alternative to proprietary automation tools like Zapier. What distinguishes n8n is its hybrid approach: a drag-and-drop visual interface for rapid prototyping combined with full JavaScript code access in Function nodes for custom logic. In 2024-2025, n8n added native AI agent capabilities, making it a powerful platform for building LLM-powered automation workflows. Backed by Sequoia and Highland Capital ($55M Series B), n8n has surpassed 100M Docker pulls.

Key Features

Visual Workflow Builder — Drag-and-drop node-based interface for creating automation workflows
AI Agent Nodes — Native connectors to LLM providers (OpenAI, Anthropic, Hugging Face, Cohere) for agent-based automation
LangChain Integration — AI-native nodes for chaining LLMs, vector stores, and tools in workflows
400+ Integrations — GitHub, Google Sheets, Postgres, Slack, Supabase, and more
AI Workflow Builder — Text-to-workflow generation from natural language descriptions
Self-Hostable — Full data sovereignty via Docker, Kubernetes, or Node.js deployment
Fair-Code License — Free community edition with enterprise add-ons (RBAC, SSO)
Git Control — Version-controlled, testable workflows with role-based access

Architecture

n8n uses a node-based architecture where workflows are directed graphs of connected nodes:

graph LR A[Trigger: Webhook] --> B[AI Agent: LLM Call] B --> C[Action Node: Google Docs] B --> D[Tool Node: Search] D --> E[Output Node: Slack Post]

Infrastructure:

graph TD A[n8n Runtime] --> B[Postgres: State] A --> C[Redis: Cache] A --> D[Queue Mode: Workers]

Each node processes JSON data streams with support for parallel execution, branching logic, and error boundaries. The platform runs on Postgres for state persistence and Redis for caching and queue management.

Code Example

Example of an AI agent workflow using n8n's JavaScript Function node:

# Python script to trigger an n8n webhook-based AI workflow
import requests
 
webhook_url = "https://your-n8n.example.com/webhook/ai-agent"
 
payload = {
    "query": "Summarize the latest news about AI agents",
    "context": {
        "max_tokens": 500,
        "model": "gpt-4o"
    }
}
 
response = requests.post(webhook_url, json=payload)
result = response.json()
print(f"Agent response: {result['output']}")

Self-Hosting Options

Method	Description
Docker	`docker run -it –rm -p 5678:5678 n8nio/n8n`
Docker Compose	Full stack with Postgres + Redis for production
Kubernetes	Helm charts for scalable deployment
Node.js	`npx n8n` for quick local development
n8n Cloud	Managed hosting for convenience

Why #1 JS Rising Stars 2025

n8n topped the JavaScript Rising Stars 2025 rankings due to:

100M+ Docker pulls showing massive adoption
AI-native features bridging automation and LLM workflows
Self-hosting appeal amid data sovereignty concerns
Fair-code model offering openness without full open-source risks
10,000+ community workflow templates

Table of Contents

n8n