Resources »

Solutions Hub »

Integrations »

OpenAI

Introduction

In this guide we'll walk through the process of linking Cerb to OpenAI. You'll be able to use OpenAI's full API from automations in Cerb.

Configuration

OpenAI

Log into OpenAI: https://platform.openai.com/settings/

Create a project

Click on Projects in the left sidebar.

Click the green +Create button in the top right.

Name your project Cerb and click the green Create button in the popup.

Create an API key

Click on API keys in the left sidebar.

Click on the green +Create new secret key button in the top right.

Name the key Cerb Automations and select the Cerb project.

Click the Create secret key button.

Copy the secret key to your clipboard for the next step.

Cerb

In Cerb, navigate to Search » Connected Services » (+) and select OpenAI from the library.

Paste your API key from above.

Click the Create button in the popup.

Examples

Create chat completion

https://platform.openai.com/docs/api-reference/chat

start:
 http.request/chat:
   output: http_response
   inputs:
     method: POST
     url: https://api.openai.com/v1/chat/completions
     headers:
       Content-Type: application/json
     authentication: cerb:connected_account:openai
     body:
       model: gpt-4o
       messages:
        0:
          role: developer
          content: you are a helpful assistant
        1:
          role: user
          content: hello
   on_success:
     set:
        response@json: {{http_response.body}}
        http_response@json: null

Generate text embeddings

https://platform.openai.com/docs/api-reference/embeddings

start:
 http.request/embeddings:
   output: http_response
   inputs:
     method: POST
     url: https://api.openai.com/v1/embeddings
     headers:
       Content-Type: application/json
     authentication: cerb:connected_account:openai
     body:
       input: Cerb automates helpdesk inboxes and workflows. It has evolved continuously since 2002 based on the feedback of thousands of teams; from solo founders to 1,000+ person enterprises managing millions of customer requests.
       model: text-embedding-3-small
   on_success:
     set:
        response@json: {{http_response.body}}
        http_response@json: null

Related Resources