To use the HTTP Request node in n8n, add it to your workflow, pick a method, enter the URL and choose authentication, then switch on query parameters, headers or a body if the API needs them. You can also import a curl command. n8n recommends a Predefined Credential Type where one exists, and the node returns only the response body unless you change that.
When there's no ready-made node
The HTTP Request node makes a REST API call to any service. It's how you reach an app n8n doesn't have a node for, or an endpoint the existing node doesn't cover. You can set it up by filling in the parameters, or by importing a curl command, which is the fastest route when the API's docs give you a curl example.
Setting up a request
- Add the HTTP Request node to the workflow, or import a curl command.
- Select the Method and enter the URL.
- Choose Authentication, picking a Predefined Credential Type if one exists for the service.
- Turn on Send Query Parameters, Send Headers or Send Body as needed, and choose a body content type.
- Use Add Option for batching, redirects, response format or pagination.
- Run the node and check the output.
The parts of a request
| Part | Your choices |
|---|---|
| Method | DELETE, GET, HEAD, OPTIONS, PATCH, POST or PUT |
| Authentication | A Predefined Credential Type, or generic: Basic, Custom, Digest, Header, OAuth1 API, OAuth2 API or Query auth |
| Query parameters and headers | Using Fields Below or Using JSON |
| Body | Form URLencoded, Form-Data, JSON, n8n Binary File or Raw |
| Response Format | Autodetect (default), File, JSON or Text |
n8n recommends the Predefined Credential Type option when it's available, because it's easier to manage than generic credentials. If n8n already knows the service, use its credential and let n8n handle the details.
Defaults that catch people out
The n8n Docs for the HTTP Request node list a few defaults worth knowing before your first run. The one that causes the most head-scratching is that the node returns only the response body. If you need the status code or headers, for example to check a rate limit, switch on Include Response Headers and Status to get the full response.
The node also only succeeds on a 2xx code. Anything else, such as a 404, is treated as an error and stops the workflow. If you'd rather handle that yourself, perhaps to log missing records and carry on, switch on Never Error so the node returns success whatever the code.
- SSL: by default the node only downloads the response if certificate validation succeeds. Ignore SSL Issues overrides that, which is sometimes needed for internal tools.
- Redirects: followed by default, with a Max Redirects setting.
Big results and busy APIs
When results are too big for one call, the Pagination option fetches them across several requests. n8n's advice is to inspect the API's data first, so you know how it signals the next page before you set it up.
For APIs with strict limits, the Batching option sets Items per Batch and a Batch Interval in milliseconds. That lets you pace requests, for example a few items every second, rather than sending hundreds at once and getting blocked. If the push-or-poll question is new to you, my explainer on APIs versus webhooks covers it.
Before you switch the workflow on
- Run the node once and check the output matches what the API's docs describe.
- Decide whether a non-2xx response should stop the workflow or carry on with Never Error.
- Switch off Include Response Headers and Status if you only needed it for testing.
- Set Batching if the node will run over many items.
- Leave SSL validation on unless you've a clear reason to ignore it.
Similar tools elsewhere
Zapier lists a Webhooks by Zapier integration for connecting apps through webhooks. Make has scenario settings such as Cycles per run, which cap how many cycles a scenario execution can run. If you're comparing platforms, my roundup of n8n alternatives goes wider.
My advice for your first API call
Copy the curl example from the API's docs and import it, so the method, URL and headers arrive filled in. Use a Predefined Credential Type where you can, switch on Include Response Headers and Status while you're testing, and add Batching before you point the node at a long list. If you'd rather hand the build to someone, hiring a Zapier, Make or n8n expert is worth considering.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.
Frequently asked questions
Can I import a curl command?
Yes, as an alternative to filling in the parameters.
Which auth should I use?
n8n recommends a Predefined Credential Type when one is available.
How do I see the status code?
Turn on Include Response Headers and Status.
How do I stop the workflow failing on a 404?
Turn on Never Error.
Does it handle paginated APIs?
Yes, through the Pagination option.


