HTTP Request

Type key: http_request
Group: Integrations
Category: integration
Tier: CORE

The HTTP Request node sends HTTP requests to any external API or service. It supports all standard HTTP methods, multiple authentication types, custom headers, query parameters, and request bodies in various formats.

Credentials (optional)

If the target API requires authentication, create a credential first in Voodflow → Credentials and select it via credential_id. The node supports API Key, Bearer Token, Basic Auth, and OAuth2 credential types.

→ How to create credentials

Handles

Handle	Direction	Description
input	Input	Receives data used for template resolution in URL, headers, and body
output	Output	Emits the parsed response body

Configuration

Field	Type	Default	Description
url	string	""	The target URL; supports {{tags}}
method	enum	GET	HTTP method: GET, POST, PUT, PATCH, DELETE, HEAD
credential_id	integer	null	Reference to a stored Credential
headers	array	[]	Custom HTTP headers (key-value pairs)
query_params	array	[]	URL query parameters (key-value pairs)
body_type	enum	json	Request body format: json, form, raw, input
body_data	array/string	[]	Request body data
timeout	integer	30	Request timeout in seconds
response_type	enum	auto	Response parsing: auto, json, text, binary

Authentication

The node resolves credentials in the following priority order:

1. System Credential (via credential_id) — recommended for production
2. Inline auth — configured directly in the node (for quick testing)

Supported Credential Types

Type	HTTP Mechanism
bearer_token / api_token	Authorization: Bearer <token>
basic_auth	Authorization: Basic <base64(user:pass)>
oauth2	Authorization: Bearer <oauth_access_token> (auto-refreshed)
api_key_header	Custom header (e.g. X-API-Key: <key>)
api_key_query	Query parameter (e.g. ?api_key=<key>)

Request Body Types

json

Body is serialized as JSON. Content-Type: application/json is set automatically.

{
    "model": "llm-model",
    "messages": [{ "role": "user", "content": "{{input.prompt}}" }]
}

form

Body is URL-encoded form data. Content-Type: application/x-www-form-urlencoded.

input

The body is taken directly from an input field of the execution context (resolved via {{tags}}).

raw

Raw string body. Set Content-Type manually in headers.

Response Handling

Mode	Behavior
auto	Attempts JSON parse; falls back to raw text
json	Strict JSON parse; fails if response is not valid JSON
text	Returns raw response body as string
binary	Returns base64-encoded binary data

Output Format

Successful response:

{
    "status": 200,
    "headers": {
        "content-type": "application/json",
        "x-request-id": "abc123"
    },
    "body": {
        "id": "chatcmpl-123",
        "choices": [{ "message": { "content": "Hello!" } }]
    }
}

On HTTP error (4xx / 5xx), the node returns ExecutionResult::failure() with the status code and body — use a Catch node to handle these gracefully.

Template Support

The url, header values, and body fields all support {{tags}}:

URL:    https://api.example.com/users/{{input.user_id}}/profile
Header: X-Tenant-ID: {{variables.tenant_id}}
Body:   { "query": "{{input.search_term}}" }

Examples

Call LLM Provider

URL: https://api.your-llm-provider.com/v1/chat/completions
Method: POST
Credential: LLM Provider (bearer_token)
Body Type: json
Body:
    {
        "model": "llm-model",
        "temperature": 0.7,
        "messages":
            [
                { "role": "system", "content": "You are a helpful assistant." },
                { "role": "user", "content": "{{input.user_message}}" },
            ],
    }

Fetch User from REST API

URL: https://api.example.com/users/{{input.id}}
Method: GET
Credential: Example API (api_token)

POST Form Data

URL: https://hooks.example.com/submit
Method: POST
Body Type: form
Body:
    name: { { input.name } }
    email: { { input.email } }

Notes

• The node uses Laravel's Http facade internally (powered by Guzzle)
• SSL certificate verification is enabled by default; disable only in development via PHP Code node
• Timeouts default to 30 seconds; increase for slow AI or third-party APIs
• Response headers are always included in the output for downstream use (e.g. reading pagination headers)