> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rovax.io/llms.txt
> Use this file to discover all available pages before exploring further.

# HTTP Request Node

> Call any external REST API from within your workflow — GET, POST, PUT, PATCH, or DELETE, with custom headers and automatic retry.

## How It Works

The **HTTP Request** node connects the workflow to any external service that exposes a REST API. Execution is synchronous: the flow waits for the response before advancing to the next node.

What the node supports:

* Five HTTP methods: `GET`, `POST`, `PUT`, `PATCH`, and `DELETE`
* Custom headers with shortcuts for the most common presets (`Content-Type`, `Authorization`, `Accept`, `X-API-Key`)
* Request body in JSON, Form Data, or plain text — available only for `POST`, `PUT`, and `PATCH`
* Configurable timeout up to 30000ms (30 seconds)
* Automatic retry from 0 to 3 attempts with a configurable interval between them
* `{{ }}` templates in all fields — URL, headers, and body accept dynamic references to flow data

## Configuration Options

**Request tab:**

* **Method and URL** — choose the HTTP verb and enter the full URL; templates like `{{ $trigger.body.id }}` work directly in the URL
* **Body type** — `json` (textarea editor with highlighting), `form` (key/value pairs encoded as `application/x-www-form-urlencoded`), `text` (free-form), or `none`
* **Body** — editor with `{{ }}` template support — resolution happens at runtime before sending

**Headers tab:**

* Add headers via preset shortcuts or as custom fields
* For sensitive values, use `{{ $vars.token }}` instead of hardcoded strings

**Advanced tab:**

* **Timeout** — minimum 1000ms, maximum 30000ms
* **Retry** — 0, 1, 2, or 3 attempts; retry interval from 100ms to 10000ms

<Note>
  Avoid inserting tokens and passwords directly in the URL or body. Prefer workflow variables defined in a preceding Set Variable node or retrieved from secure integrations configured in the workspace.
</Note>

## Common Use Cases

The HTTP node is versatile enough to cover most no-code integrations:

* **Fetch data before acting** — make a GET to the ERP API with the order ID received via webhook and use the response in downstream nodes via `{{ $json.body.field }}`
* **Register events in external systems** — send a POST to a third-party CRM with contact data collected during the conversation, using `{{ $trigger.parameters.email }}` in the body
* **Chain dependent calls** — first GET to fetch authentication data, then POST to the main API using the token returned by the first node via `{{ node_auth.body.access_token }}`