Documentation

How to get the most out of Gydmation, from your first prompt to a deployed workflow.

Getting started

Type what you want automated into the composer, in plain English. Gydmation checks its template library first — if something close already exists, it customizes that; otherwise it generates from scratch. Either way, you get a real, validated n8n workflow.

Writing good prompts

The more specific you are about the trigger and the destination, the less back-and-forth you'll need.

Works well

"When a new row is added to my Google Sheet, post a summary to the #sales Slack channel."

Needs more detail

"Notify the team about new sales." — which sheet? which channel?

You don't need to know n8n's node names — just describe the trigger, the action, and where it should go.

Clarifying questions

If something in your prompt is genuinely ambiguous — an unnamed channel, an undefined threshold — Gydmation asks instead of guessing. Tap an option, or type your own answer, and it continues building with that in mind.

Why this matters: a silent guess on something like "notify the team" could mean email, Slack, or SMS — three very different workflows. One tap resolves it correctly the first time.

Editing a workflow

Open any workflow from your history and describe the change in plain English:

"Add a Gmail step at the end that emails me if it fails."

Gydmation edits the existing workflow rather than rebuilding it from scratch, and redeploys it if it's already connected to n8n.

Connecting n8n

  1. Click Connect n8n in the header
  2. Enter your n8n instance URL and an API key from your n8n account
  3. Once connected, every generated workflow can be deployed with one click

Don't have an n8n instance yet? You can still generate and review workflows — connect one whenever you're ready to deploy.

Deploying

Once connected, click Deploy on any workflow. It's sent straight to your n8n instance, ready to activate. Editing a deployed workflow and saving again updates the same workflow in n8n, rather than creating a duplicate.

Plans & billing

Free: 5 workflows a month, using your own API key.

Pro ($25/mo): 1,000 AI credits a month (roughly 200 workflows, based on typical usage) using our managed models — Claude Sonnet 5 and Claude Opus 5 — with no API key needed. Cancel anytime by emailing support@hacroo.com — you keep Pro access through the end of your current billing period.

Want to use a different model, like GPT-5.6 Terra? On any plan, you can add your own API key (BYOK) for that or any other supported model — those generations use your key directly and don't draw from your Pro credit balance at all.

Choosing a model

Switch models from the picker in the composer, any time — including mid-conversation. On Free, you'll need your own API key for whichever model you pick; Pro models work with no setup.

See the full AI Models page for every supported model ranked by accuracy, with guidance on which one to pick and when.