Workflow Automation

Calling any API

In this guide, you will learn how to set up an HTTP Request action to call any API in Lindy.

Lindy supports 3,000+ integrations out of the box, but if you have custom APIs you want to call, you can use the HTTP Fetch action.

Setup

To get started, click the plus sign beside Lindies to start building your Lindy.

Click "Start from scratch" to create a new Lindy.

Add a Trigger

Click “select trigger”, then choose which trigger you wanted to use to initiate the workflow.

Add a HTTP Request

Configure the rest of your workflow (if you want additional steps prior to calling an API) or go directly to adding the Request action.

Click "Perform an action" to add the HTTP Request action.

‍

HTTP Request Configuration

In the HTTP Request settings, specify the Model you want to use for the action and the URL of the webpage or endpoint to fetch. (You can also reference URLs found in prior actions/outputs by using a prompt or referring to trigger/action output data. A guide on how to do this is available here.)

Lindy supports all standard HTTP Methods. You can specify the Fetch Method in the "Method" field and the content type of the body in the "Content Type" field. If Detect is selected, the content type will be detected as JSON or XML automatically. If JSON or XML is selected, the body will be parsed and validated before fetching.

You can also add custom headers to the HTTP Request action. Headers are additional key-value pairs sent with your request to provide information such as authentication, API versioning, or other metadata.

Common headers include:

Authorization: Used to send authentication tokens (e.g., Bearer tokens).
Content-Type: Specifies the format of the body (e.g., application/json).

If you supply the “Content-Type” header, it will override the content type previously specified.

For example, if you’re sending JSON data that requires authorization, set the Headers mode to “Set Manually” and set the key-value pairs to:

Name: Authorization
Value: Bearer YOUR_API_TOKEN
Name: Content-Type (optional, if not automatically detected or previously set)
Value: application/json (optional, if not automatically detected or previously set)

You can also use the Prompt AI mode in these Header fields if you want to more easily reference data from previous steps in your flow.

If you have any dynamic data you want Lindy to pass, add this as a prompt by setting the "Body" field to Prompt AI mode and instruct Lindy what dynamic data to pass.

In the example below, Lindy is sending data to a Zapier Webhook.

The Body of the request uses an "AI Prompt" that instructs Lindy to write a poem about the user information received from the Webhook trigger.

Lindy receives the information from the Webhook Trigger:

And sends the poem to the Zapier webhook using the HTTP Request.

Below, you can see the Raw Body received from Zapier:

Guides

Building a Slackbot: 101

Influencer Outreach