n8n Local Deployment

Applicability and prerequisites

OS: Linux (Ubuntu/Debian/CentOS, etc.) or macOS

Suggested resources: 2 GB+ RAM, 10 GB+ free disk space

Shell access and the ability to reach the public npm registry

Local-only deployment

n8n 1.x requires Node.js 18.x or 20.x (LTS recommended)

Preparation

• Log in to cometapi. Click "ADD API key" in the API keys to get your token key: sk-xxxxx;

Configuration

1.Install Node.js (recommended: nvm)

If you see “n8n: command not found” later, restart the terminal or ensure the nvm-managed npm global bin path is on your PATH.

2. Install and start n8n (local)

Start in the foreground:

Default URL: http://localhost:5678/

If the port is in use:

Optional: keep it running in the background (still local):

Open http://localhost:5678/ in your browser.

On first run, follow the wizard to create the Owner account (email + password) and sign in.

4. Install the CometAPI community node

Go to Settings (bottom-left) > Community nodes

Click “Install a community node”

In “npm Package Name”, enter: @cometapi-dev/n8n-nodes-cometapi

Check the risk acknowledgment box

Click “Install”

After installation, you can search for “CometAPI” in the node panel.

If you don’t see the community nodes section:

Make sure you are logged in as the Owner

Check Settings > Security to ensure community node installation is allowed

5. Create a workflow and configure the API

API Configuration

Click “+ New” at the top to create a new workflow.

Click the “+” on the right side of the canvas and search for “CometAPI” to add the node.

You can also search “CometAPI” from the left node list and drag it onto the canvas.

Full steps

Add a trigger node: search for “Manual Trigger” or “When clicking ‘Execute workflow’”.

Add the “CometAPI” node and connect the trigger to it.

Configure credentials (first time only):

In the CometAPI node, under “Credential to connect with”, choose “Create new”.

Paste the token you created in the CometAPI console into “CometAPI Key”.

Save. The Base URL defaults to https://api.cometapi.com and typically needs no change.

Configure model and messages:

Model: enter a supported model, e.g., gpt-4o or gpt-4o-mini.

Messages:

Role: User, Content: e.g., “Hello”

Optional Role: System, Content: e.g., “You are a useful assistant”

Optional parameters (as needed):

Maximum Number of Tokens (e.g., 4096)

Sampling Temperature (e.g., 0.7)

Top P, Presence/Frequency Penalty

Stream (enable/disable streaming)

Test execution:

Select the CometAPI node and click “Execute step”, or click “Execute workflow” at the bottom.

The OUTPUT panel on the right should show a JSON response (including choices, usage, etc.).

6. Troubleshooting

node-gyp/native build errors

Install build tools (build-essential/make/g++/python3) as shown in step 1.

npm global permission errors (EACCES)

Use nvm for a user-level Node.js install (no sudo needed).

Port conflicts (cannot open http://localhost:5678/)

Set a different port via N8N_PORT before starting.

Community nodes section or install button not available

Corporate network/proxy causes npm install failures

Configure a proxy if needed: npm config set proxy http://user:pass@host:port

CometAPI request returns 401/403

Verify your API Key and model access permissions.

Empty responses or token limits

Increase Maximum Number of Tokens; adjust Temperature/Top P as needed.

n8n stops after closing the terminal

Use pm2 to run n8n in the background (still local).

You have now installed n8n locally on Linux via npm, installed and configured the CometAPI community node, and can run workflows that call models such as gpt-4o to receive responses.

Configuration#

1.Install Node.js (recommended: nvm)#

2. Install and start n8n (local)#

3. First access and sign in#

4. Install the CometAPI community node#

5. Create a workflow and configure the API#