The HTTP node makes HTTP requests using cURL syntax. It can be a workflow starting point (no incoming edge) or receive input from upstream nodes.
Overview
| Property | Value |
|---|---|
| Inputs | 0 or 1 |
| Outputs | 1 |
| Output | $nodeLabel.status, $nodeLabel.body, $nodeLabel.headers |
Parameters
| Parameter | Type | Description |
|---|---|---|
label | string | Node identifier (camelCase) |
curl | string | cURL command (e.g. curl -X GET https://api.example.com) |
Response Format
{
"status": 200,
"headers": { "content-type": "application/json", ... },
"body": "response body or parsed JSON",
"request": { "method": "POST", "url": "...", "headers": {...} }
}Accessing Response
$httpNode.status– HTTP status code$httpNode.body– Response body (string or parsed JSON)$httpNode.body.fieldName– Access JSON fields$httpNode.headers– Response headers
Example
{
"type": "http",
"data": {
"label": "fetchApi",
"curl": "curl -X GET https://api.example.com/data"
}
}Use cURL syntax with -H for headers, -d for POST body, etc. Expressions can be used in the curl string for dynamic URLs.
Related
- Node Types – Overview of all node types
- Triggers – HTTP as workflow entry point
- Credentials Tab – Add API keys for authenticated requests