How to Send a Zapier Webhook to OpenAI and Get a Summary Back
Catch any incoming payload in Zapier, pass it to the OpenAI Chat Completions API, and reuse the summary in later steps.
A Catch Hook trigger in Zapier gives you a public URL that accepts any POST request. Once data lands there you can hand it to OpenAI and turn long text into a tidy summary that the rest of your Zap can use. This guide wires that up end to end without writing a server.
What you need
- A Zapier account (the free plan includes Webhooks by Zapier)
- An OpenAI account with an API key from platform.openai.com
- A source that can POST JSON to a URL (a form, app, or curl)
Step 1: Create a Catch Hook trigger
Make a new Zap and pick Webhooks by Zapier as the trigger app. Choose the Catch Hook event. Zapier shows you a unique URL such as https://hooks.zapier.com/hooks/catch/123456/abcde. Copy it.
Step 2: Send a test payload
Zapier needs a sample to learn the field names. Fire a quick POST from your terminal so the trigger records a real example.
Step 3: Add an OpenAI action
Add an action step. You can use the built-in OpenAI app, but the most flexible option is Webhooks by Zapier with the POST event so you control the exact body. Point it at https://api.openai.com/v1/chat/completions and set the Authorization header.
{
"model": "gpt-5-mini",
"messages": [
{ "role": "system", "content": "Summarize the text in 3 bullet points." },
{ "role": "user", "content": "{{body}}" }
],
"temperature": 0.3
}Step 4: Map the reply into later steps
OpenAI returns the text under choices.0.message.content. Open the action output in Zapier and pick that field for any downstream step, such as appending a row to a sheet or sending a Slack message.
Result: a meeting transcript POSTed to your hook comes back as three clean bullet points within a couple of seconds, ready to drop into Slack or Notion.
Watch related tutorials
12:38
14:09
17:53
15:00
12:00
20:00