Describe your automation. Get the n8n workflow, ready to import.

One n8n workflow per request, as files: workflow.json to import, SETUP.md with every step from import to switching it on, check-report.md showing the n8n release it was imported into and the checks it passed, and diagram.png. For a fix you also get what-changed.md. Each workflow includes two rounds of changes.

It will import cleanly: we import every file into n8n ourselves and check that each node type and version exists, each connection resolves and each expression points at a real node. What we cannot do is run it with your accounts, because we never have them. So the first live run is yours. If a node fails on your real data, send the error message and you get the corrected file; that is what the two rounds of changes are for.

No. You import the file yourself (workflow menu, Import from File) and connect your own accounts inside your n8n. The setup guide lists each account and where to get its key. Please do not send us passwords or API keys.

Files are built for and checked in a current stable release of n8n; the check report names the exact one. They work on n8n Cloud and on self-hosted n8n. If you run an older version, say so in your request and the workflow is built for that version.

Yes. You get the agent node with a chat model, memory, the tools it may call and a structured output, with the system prompt written for your case. You connect your own chat model account in n8n, and its usage is billed by that provider to you, not through us.

Yes. In n8n select all nodes, copy, and paste them into the box together with the error message. The preview shows what is wrong and where. Before you paste: copied workflows contain the names of your credentials but not the secrets; an HTTP Request node that was imported from a cURL command can contain a key in its headers, so remove that first.

The builder inside n8n is good for a first draft and for edits while you sit in the editor, and it runs on your plan's credits. A chatbot gives you JSON that often names nodes or versions that do not exist, and it fails on import. We deliver a file that was imported into n8n and checked, with error handling, test data, a setup guide, and fixes when it meets your real data. If the built-in builder already does the job for you, use it.

One request is one workflow file of up to about 30 nodes, with two rounds of changes. Fixing a workflow you already have counts as one. A system that needs several workflows working together uses one per workflow, and the preview tells you when that is the case.

The preview takes about a minute. The file is usually ready within the hour. You get an email with the link when it is done.

You own it. Agencies and freelancers can hand it to their clients; nothing in the file mentions us. We use your request only to do the job, never for training or marketing. It is stored and processed in Germany after passing through our server in the US; our AI provider in the US reads it under terms that exclude training. Deleted after 30 days.

Ask for a change first, in one line. If it still misses, email us within 14 days and we refund your month.

Any time, from your account page, in two clicks. You keep access until the end of the paid month.

No. This is an independent service from 20Vision GmbH in Vienna. We are not affiliated with or endorsed by n8n. You need your own n8n to use the files.