What is n8n?

n8n (pronounced "n-eight-n", short for "nodemation") is a workflow automation platform that lets you connect apps, databases, APIs and AI models on a visual canvas. Each step is a 'node' — you drag nodes onto a canvas, wire them together, and n8n executes the resulting workflow on a trigger (webhook, schedule, form submission, chat message, or manual click).

What sets it apart from typical no-code tools is that it never boxes you in: any node can be replaced with a JavaScript or Python Code node, and every field accepts expressions, so you can drop down to real code exactly where you need it and stay visual everywhere else. n8n ships with 500+ pre-built integrations (SaaS apps, databases, cloud storage, LLMs and vector stores) and can be extended with community or fully custom TypeScript nodes.

n8n is developed by n8n GmbH (Berlin) and is source-available: the full source code lives on GitHub, and most of the product is usable for free — either self-hosted or via the official managed n8n Cloud.

n8n is released under the Sustainable Use License (a fair-code license), not a traditional OSI open-source license. In practice this means: you may use n8n for free internally at any company of any size, self-host it, modify the source, and even embed it inside a product you sell — as long as you're not reselling access to n8n itself as a hosted automation service that competes with n8n Cloud.

A small subset of enterprise-only features (advanced permissions, SSO/SAML, log streaming, external secrets, etc.) require a paid Enterprise license key, but the core workflow engine, node library, and self-hosting path stay free forever.

Five ideas unlock everything in n8n:

A typical n8n workflow looks like: Trigger → Data shaping (Set/Edit Fields, Code) → Branching (IF/Switch) → App actions (HTTP Request, Slack, Google Sheets, database nodes) → Error handling (Error Trigger workflow, NoOp, Stop and Error) → optional sub-workflow calls (Execute Workflow) for reusable logic.

Workflows are stored as JSON, which is why templates are simply copy-paste-able — the entire canvas, node parameters, and connections travel in one file you can import via the UI or the API.

n8n has native, first-class support for building AI agents — not just calling an LLM API. The AI Agent node orchestrates a language model, a memory module (buffer, Postgres, Redis), and a set of Tools (any n8n node — or sub-workflow — exposed as a callable tool, plus built-ins like Wikipedia, SerpAPI, calculator, Code).

Under the hood, these nodes are built on LangChain, giving you agent types (Tools Agent, Conversational Agent, ReAct), output parsers, retrievers, and native vector-store nodes (Pinecone, Qdrant, Supabase, PGVector, Weaviate) for Retrieval-Augmented Generation (RAG). Combined with the Chat Trigger, you can ship a production chatbot or autonomous agent without writing a Python backend.

Self-hosting (Docker, a VPS, Kubernetes, or a hyperscaler) gives you unlimited workflow executions, full data residency, and the ability to install any community node — at the cost of owning upgrades, backups, TLS, and monitoring yourself.

n8n Cloud is the official managed service: automatic upgrades, backups, and support, priced per plan with workflow/execution limits. It's the fastest path to production and the only option with official SLAs and enterprise SSO out of the box.

n8n isn't the right tool for everything. Skip it (or pair it with something else) if:

Credentials (API keys, OAuth tokens, DB passwords) are encrypted at rest using an instance-wide encryption key (N8N_ENCRYPTION_KEY) and are never exposed in plain text in workflow JSON exports. n8n supports external secret managers (Vault, AWS/Azure/GCP secret stores) on Enterprise plans, role-based access control, and SSO/SAML/LDAP.

Best practices: rotate the encryption key carefully (it re-encrypts all credentials), run behind HTTPS with a reverse proxy, restrict the built-in Code node in shared/multi-tenant instances, and keep n8n updated — the team ships regular security patches.

Self-hosted n8n (Community edition) is free forever, with no execution limits — you only pay for your own infrastructure. n8n Cloud is priced per plan (Starter, Pro, Enterprise) based on active workflows, executions/month, and users; all plans include a free trial. Enterprise unlocks SSO, environments, and advanced permissions on both cloud and self-hosted (via license key).

Always check current numbers at n8n.io/pricing — plans and limits change periodically.