Free n8n workflow library
Free n8n Workflow Library
Download, study, and customize twelve practical n8n workflows built to solve real problems without creating an enormous automation stack.
These starter workflows use n8n, local AI, optional paid APIs, structured inputs, human review checkpoints, and useful outputs you can inspect and improve.
Why this exists
Small Workflows Are Easier to Understand
The best first n8n workflow is usually not a giant assistant or an agent that tries to run your entire life.
It is one small process with a clear input, a useful output, and a human reviewing the result.
This library was created to help people learn that pattern through practical examples they can import, inspect, test, and customize.
Shared architecture
The Pattern Behind the Library
Most workflows intentionally use a shared beginner-friendly architecture: manual or structured input, prompt preparation, a local Ollama model, response parsing and formatting, human review, and Google Docs or another useful destination.
The workflows are intentionally similar in structure. That is part of the lesson. Once you understand the pattern, you can change the input, prompt, validation rules, and output destination to solve many different problems.
Most workflows currently use Google Docs as an easy review destination. You can replace that final step with Markdown files, Obsidian, Notion, a database, email, or another destination.
Workflow downloads
Free templates you can import, inspect, and customize.
Productivity
Daily Action Brief Builder
Messy notes do not become a usable plan.
Turn scattered working notes into a structured daily action brief with priorities, next steps, parked ideas, and a human review note.
Content and SEO
Search Intent Blog Outline Builder
A topic is too loose to become a useful article.
Turn a topic, reader, search intent notes, competitor observations, and a unique angle into a structured article outline.
Research
RSS Research Digest
RSS feeds contain more updates than a person can reasonably review.
Collect feed items, apply a topic lens, summarize useful updates, identify must-read items, and create a practical research digest.
Prompt Engineering
Prompt Starter Library Builder
Useful prompts disappear inside old chat conversations.
Turn rough tasks and prompt ideas into a categorized, reusable prompt starter library with variables, use cases, and quality notes.
Content
YouTube Transcript Cleaner
Raw transcripts are difficult to read, search, quote, or repurpose.
Clean a rough transcript while preserving its meaning, then extract useful takeaways, review flags, and repurposing ideas.
Operations
SOP Generator
A repeatable process exists in someone's head but is not documented.
Turn rough process notes into a structured standard operating procedure with steps, prerequisites, quality checks, risks, and completion criteria.
Research
Research Collector
Saved links, snippets, and notes become scattered and difficult to reuse.
Organize research material into sources, themes, findings, assumptions, follow-up questions, next actions, and tags.
Content Repurposing
Social Post Generator
One useful article or workflow takes too long to rewrite for multiple platforms.
Turn one source idea into platform-specific drafts, hooks, calls to action, hashtag ideas, thread concepts, carousel outlines, and review notes.
Communication
Professional Email Builder
Rough email notes take too long to turn into a clear professional message.
Turn bullet points and recipient context into subject options, primary and alternate drafts, a follow-up draft, boundaries, and review flags.
Knowledge Management
Markdown Knowledge Base Builder
Knowledge becomes scattered across documents, chats, apps, and notes.
Turn scattered notes into Markdown-ready knowledge content with frontmatter, folder suggestions, internal-link ideas, decisions, tasks, questions, and cleanup notes.
Support Operations
AI Support Ticket Triage
Messy support requests are easy to misread, misroute, or answer too quickly.
Turn one messy support ticket into a structured triage brief with urgency, category, risk flags, confidence, routing notes, and a human-reviewed reply draft.
Content and SEO
Paid API SEO Keyword Research
AI can suggest topics, but it should not guess search volume, CPC, or keyword difficulty.
Use OpenRouter for seed expansion, DataForSEO for keyword metrics, cleanup logic for deduping, and human review before choosing what to write.
Repository contents
More Than a JSON Download
Each workflow repository is designed to help readers understand and customize the system, not just import a file and hope for the best.
Requirements
What You Need
- n8n
- Ollama running locally
- A compatible local model, such as llama3.1:8b
- Google Docs credentials for workflows using Google Docs
- Willingness to review AI output before using it
The workflows are built for learning and customization. Exact node versions, model names, and credential setup may vary depending on your environment.
New to the pieces? Start with Local AI for Beginners, the Ollama tutorial, How to Build Your First AI Workflow, or the Daily Action Brief walkthrough.
How to use the library
A Simple Way to Start
- 01Pick one workflow that solves a problem you already have.
- 02Open the GitHub repository.
- 03Read the requirements and installation guide.
- 04Import the workflow into n8n.
- 05Connect your own credentials.
- 06Confirm the Ollama model name.
- 07Run the included example.
- 08Review the output.
- 09Customize one part at a time.
Do not import all twelve and attempt to automate everything at once. Start with one workflow, understand it, and then expand.
Quick answers
Common questions before you import a workflow.
Are these workflows really free?
Yes. The workflow exports and supporting repository documentation are available under the MIT license.
Do I need to be a developer to use them?
Not much. Some workflows include readable JavaScript nodes, but beginners can import and run them without rewriting the code. Understanding the basic n8n interface will still help.
Do I need a paid AI API?
Most workflows do not. The beginner workflows use Ollama and a local language model for the AI step. GP-WF-012 is different: it is a paid-API example that requires your own OpenRouter and DataForSEO accounts.
Are the workflows fully local?
The AI reasoning runs locally through Ollama. Most workflows currently save their final review copy to Google Docs, so that output step is not fully local. Readers can replace it with a local destination.
Will these workflows work on Windows, macOS, and Linux?
The workflow logic is platform-independent, but Ollama, Docker, networking, file paths, and local n8n setup can vary by operating system. I would not promise identical setup steps on every platform.
Can I customize the workflows?
Yes. Readers are encouraged to replace prompts, models, input fields, output destinations, and review rules.
Do the workflow files contain credentials?
No. The public exports should not contain private credentials, OAuth tokens, API keys, or personal account information.
Does the Markdown workflow write straight to Obsidian?
No. GP-WF-010 generates Markdown-ready content and currently saves the review copy to Google Docs. It does not automatically write directly into an Obsidian vault or local Markdown folder.
The library also includes a paid-API example for builders who are ready to connect real keyword data. The n8n SEO keyword research workflow shows how OpenRouter seed expansion, DataForSEO keyword metrics, cleanup logic, and human review fit together.
Before you import a workflow or rebuild it for your own process, you can also run the idea through the free AI Workflow Preflight Checker to see whether the trigger, input, review step, and output are clear.
If you are not sure which workflow to try first, start with these beginner AI workflow examples. They show what each workflow does, what the input should be, where review belongs, and which free n8n build fits the idea.
If you want the lowest-friction way to try these workflows before deciding whether to self-host, n8n Cloud is the simplest starting point. You can still learn the workflow logic here, then decide later whether local hosting makes sense.
As you rebuild these workflows, it helps to know when a regular workflow is enough and when an agentic pattern is worth exploring. Start with this plain-language guide to bots, workflows, AI agents, and agent harnesses.
If you want to share your own workflows, do not stop at the export file. This guide shows how to package an n8n workflow template so the next person understands setup, sample data, and review points.
If one of these examples becomes a workflow you rely on, this local n8n vs VPS vs n8n Cloud guide can help you decide whether to keep testing locally or give the workflow a more reliable home.
When a workflow starts sending messages, creating records, or writing to external tools, reliability means more than a green execution. This guide on preventing duplicate actions in n8n workflows shows how to use event IDs, dedupe logs, and safe retry boundaries.
Once you have a starter workflow working, the next layer is reliability. The guide to n8n error handling for AI workflows shows how to add retries, review gates, and logs to the workflows you decide to keep using.
If you decide to modify one of these workflows to use a local Ollama model, start by testing the connection in layers. The Ollama with n8n troubleshooting guide shows what to check before you assume the workflow itself is broken.
As you modify these starter workflows, one of the most useful upgrades is a review gate. The guide to human approval in n8n AI workflows shows how to pause before an AI output is sent, published, saved, or acted on.
Disclosure: Some outbound links on GetPrompting are affiliate links, including this n8n link. If you sign up through one, I may earn a commission at no extra cost to you.
If you are comparing automation platforms before committing to n8n, read this n8n vs Zapier vs Make comparison first. It explains when n8n is worth learning and when a simpler tool may be enough.
And before you turn a starter workflow into an AI Agent, read this n8n AI Agent vs regular nodes guide. It will help you decide whether the workflow really needs flexible tool selection or just a clearer set of normal steps.