SeaLinkSeaLink
/
Back to integrations

Tool setup

n8n Integration Guide

Connect SeaLink models in n8n's OpenAI node to power your automation workflows. Works with both n8n Cloud and self-hosted instances.

AutomationOpenAI-compatible

Prerequisites

  • An n8n Cloud account or self-hosted n8n environment
  • n8n version ≥ 1.0 (includes OpenAI node by default)
  • A valid SeaLink API Key (starts with sk-sealink-)

Setup Steps

1

Open n8n Editor

Log into n8n (app.n8n.cloud or your self-hosted instance) and open the workflow editor.

2

Create an OpenAI credential

Click the gear icon (bottom-left) → 'Credentials' → 'Add Credential' → search 'OpenAI' → select 'OpenAI API'. If OpenAI isn't listed, install the n8n-nodes-openai community node. n8n 1.0 and above includes it out of the box.

3

Enter SeaLink connection info

Fill in the credential form: API Key: Paste your SeaLink API Key (starts with sk-sealink-) Base URL: https://test.sealink.io/v1 Leave all other fields empty and click 'Save'.

4

Use in workflows

Create a new workflow → add an 'OpenAI' node → select your SeaLink credential in the node settings → enter a model ID in the 'Model' field (e.g. gpt-4o-mini, claude-sonnet-4-6). One thing to note: n8n's Model field is a free-text input, not a dropdown. You'll need to type the model ID yourself.

5

Test

Add an OpenAI node to your workflow, select the 'Chat' operation, type in a test message, and click 'Test Step'. If you get a response back from the AI, your setup is complete.

Common Issues

OpenAI node shows 'Invalid API Key'

Check that the API Key is pasted correctly — make sure there are no extra spaces or line breaks before or after the key. Also confirm the Base URL ends with /v1, as this suffix is easy to miss.

Not sure which model ID to use

Visit SeaLink's /models page to browse all available models and their capability labels. Model IDs are case-sensitive, so we recommend copying them directly from the page to avoid typos.

'Model not found' error

This usually means one of two things: the model ID is misspelled, or the model isn't included in your current plan. Check the /models page for the exact spelling and availability.

Next Steps