Skip to content
All use cases

Use case · n8n

Mask PII before OpenAI in your n8n workflows.

Your n8n workflow sends a form entry, an email or a ticket to OpenAI, Anthropic or another model, and every email address, card number and key in it goes along. Put the Vrymdor™ AI Node one step before the AI step and the model reads [REDACTED:EMAIL] instead.

The problem

An AI step in n8n sends the whole item it is given. Nothing in between looks at what is inside.

With the AI Node

Add one step between your trigger and your AI step. Self hosted n8n uses the official community node from npm; n8n Cloud uses a ready workflow built from n8n’s own steps until n8n verifies the node.

Read the n8n setup guide

  1. In n8n, open Settings, Community Nodes, Install, and enter n8n-nodes-vrymdor.
  2. Add the credential Vrymdor AI Node API and paste your Node Host URL, Agent ID and Agent Secret.
  3. Add the Vrymdor AI Node node before your AI step, choose Remove PII and Secrets, and set Input to the text you send to the model.
  4. Press Execute step. The output shows [REDACTED:EMAIL] where the text had an email address. Map it into your AI step.

On n8n Cloud, import the n8n Cloud workflow from the setup guide, paste the URL from your dashboard into its Vrymdor token step, and keep your Agent ID and Agent Secret in its Custom Auth credential. The node needs an address n8n Cloud can reach; the hosted node (Pro and up) is the simplest.

Try it on a website form entry

The AI Node hides email addresses, phone numbers, card numbers, US Social Security numbers, private keys, and the API keys and tokens for OpenAI, Anthropic, Stripe, GitHub, AWS, Google and Slack, plus bearer tokens and JWTs. Each one becomes a label like [REDACTED:EMAIL], so the model still knows something was there.

Runs in your browserEdit the sample. What you type here never leaves your browser.
What the AI model reads3 masked
New lead from the website form Name: Jane Example Email: [REDACTED:EMAIL_1] Company: Acme Robotics Message: Please call me about the Pro plan. My card on file is [REDACTED:CREDIT_CARD_1], and our old integration used the key [REDACTED:OPENAI_KEY_1].

The name and the company stay as they are: names are not masked by default. Masked here: EMAIL 1, CREDIT_CARD 1, OPENAI_KEY 1.

This demo uses the Playground's Local engine: public, well known detectors set to the types the AI Node masks by default, in the AI Node's [REDACTED:TYPE] format. In your workflows, the Vrymdor™ AI Node does the masking.

What it does not cover

Plainly, so you know where you still need care.

Start free, upgrade when you grow.

Community is free for up to 2 agents and 10,000 requests a month, with a hard cap so nothing is ever billed. Pro is $29 a month for up to 15 agents and 100,000 requests, and Business is $99 a month for up to 50 agents and 500,000 requests, each with 14 days free. Enterprise is sized to your contract.

Questions, answered plainly

Does it work on n8n Cloud?
Yes. n8n Cloud only installs community nodes that n8n has verified, and that listing is still waiting for n8n’s approval. Until then, the ready n8n Cloud workflow does the same job with n8n’s own HTTP Request steps. Your node needs an internet address n8n Cloud can reach; the hosted node on Pro and up is the simplest.
Can the n8n AI Agent use it as a tool?
Yes. n8n also offers the Vrymdor AI Node node as an AI agent tool. Start n8n with N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=true, which n8n needs before it offers community nodes as tools.
Does masking stop my workflow?
No. Masking never stops a run: the text goes on with labels in place of the private details. Your own policy decides what is blocked, and a refusal shows as an error you can catch with On Error set to Continue.
Where does my text go?
To your AI Node only: your own machine on every plan, or the node Vrymdor hosts for you on Pro and up. The AI Node does not send your text to any AI provider to mask it. Then the masked text goes on to the model you chose.