What You'll Set Up
By the end of this guide, your AI agent will be able to manage your n8n workflow automation platform. It can list and inspect workflows, trigger executions, monitor active runs, and help you orchestrate complex automations — all through a conversation instead of the n8n editor.
This is particularly useful for teams running dozens of automated workflows. Your agent can quickly check which workflows ran successfully, re-trigger failed ones, and give you a high-level view of your automation infrastructure without opening the n8n dashboard.
Before You Begin
Make sure you have:
- An n8n instance — Either a self-hosted installation or an n8n Cloud account
- Admin access to your n8n instance — You'll need to generate an API key
- A Pipeworks workspace — Sign up at pipeworks.ai if you haven't already
Your n8n instance must be accessible over the internet for Pipeworks to connect to it. If you're self-hosting, make sure your instance has a public URL or is reachable through a tunnel.
Steps
Get your n8n API key and host URL
Log into your n8n instance. Click on your user icon in the bottom-left corner and go to Settings. Select API from the settings menu.
Click Create an API Key (or copy an existing one if you've already created one). This key gives Pipeworks access to manage your workflows and executions.
Also note your n8n host URL. For n8n Cloud, this is something like https://your-instance.app.n8n.cloud. For self-hosted, it's whatever URL you use to access your n8n dashboard (e.g., https://n8n.yourcompany.com).
Open your Pipeworks console
Log into your Pipeworks workspace and go to the Integrations page. Find n8n in the list and click on it.
Enter your credentials
Fill in both fields:
- API Key — The API key you generated in step 1
- Host URL — Your n8n instance URL (e.g.,
https://your-instance.app.n8n.cloud)
Click Connect to save. Pipeworks encrypts your credentials with AES-256-GCM before storing them. The plaintext is never saved to disk and is only decrypted when your integration container starts.
Start the integration
Click Start to launch the n8n integration container. This takes a few seconds. Once the status shows Running, your agent is ready to work with your workflows.
Test the connection
Use the Test button in your Pipeworks console to verify the connection. This sends a simple API call to your n8n instance to confirm your API key and host URL are correct.
What Your Agent Can Do
Once connected, your AI agent has access to n8n tools for:
- Workflow management — List, inspect, activate, and deactivate workflows
- Trigger executions — Start any workflow on demand through your agent
- Monitor runs — View execution history, check success/failure status, and review output data
- Orchestrate automations — Coordinate multiple workflows as part of a larger process
- Execution details — Drill into specific runs to see what happened at each step
Use tool profiles to control what your agent can do. Set the profile to read-only if you only want your agent to monitor workflow status and execution history without triggering or modifying any workflows.
Connecting from Your AI Agent
Add the Pipeworks tools endpoint to your agent's configuration. Here's an example for Claude Desktop:
{
"mcpServers": {
"pipeworks-n8n": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://app.pipeworks.ai/tools/n8n"]
}
}
}
Claude Desktop will prompt you to sign in through Pipeworks when you first connect. No access key needed.
Troubleshooting
Connection test fails? Verify that your n8n host URL is correct and accessible from the internet. If you're self-hosting, check that your instance isn't behind a firewall blocking external connections.
"Unauthorized" errors? Make sure your API key is still valid. If you regenerated the key in n8n, update it in your Pipeworks console.
Workflows not showing up? Confirm that the API key belongs to a user with access to the workflows you expect to see. In n8n, API keys inherit the permissions of the user who created them.
Agent can't find tools? Verify your Pipeworks session is authenticated and that the integration status shows "Running" in your Pipeworks console.