OpenAI Assistant API: Set up integrations with any app through Albato

OpenAI Assistant API: Set up integrations with any app
4 min. read

With Albato, you can integrate AI assistants with over 600 apps, plus automate tasks, either with HTTP requests or by using the Albato App Integrator. You can automate lots of tasks, including creating bots in Telegram or online chats, managing data in Google Sheets, and integrating with task trackers and other apps.

Before you get started with Albato integrations

Before you get started with integrating AI assistants using Albato, make sure you follow a few important steps:

1. Create an account in OpenAI:

  • Sign up for a free OpenAI account. Once you're logged in, go to https:​//

2. Generate an API key:

  • If you don't have a key yet, just click on the "+ Create a new secret key" button to generate your unique API key. It's important to copy and save this key, as you'll need it to link your OpenAI account to Albato.


3. Create an assistant: Go to https:​// to find a list of all created assistants. Click +Create in the upper right corner to create a new one.

Ex (13).png

4. Set up an assistant:

  • Define the role of your assistant. For example, the Albato Support Assistant can answer questions about integrations, and recommend blog articles.

  • Give your assistant a name, and provide specific instructions that match its role. For example, the primary role is to provide accurate answers to Albato-related queries.

5. Select OpenAI model and tools

  • Select a model, such as gpt-4-0125-preview, for tools like file search.
  • In the Tools section, choose whether it will be a code interpreter or a search function. The search function uses downloaded files to answer user questions.

6. File upload:

  • Add files that the assistant can use as a knowledge base. For example, upload blog articles in JSON format, and dialog texts from online chats. This will help the assistant find the necessary information.

RR (7).png

How to set up OpenAI action: Message to assistant

OpnAI (2).png

1. Select an assistant:

  • Required field. Select an assistant from the pre-defined OpenAI list. You can customize this list in your OpenAI account.

2. Send a message to the assistant:

- Required field. Enter the text of the message you want to send to the assistant. The message can contain requests, commands, or questions that the assistant should respond to.

3. Thread ID:

  • Optional field. Identifies the thread within which the assistant maintains the context of the conversation. It keeps the previous context while continuing the dialogue.

4. Override instructions:

  • Optional field. You can use this to give the assistant specific instructions on how to behave during a conversation.

5. Additional instructions:

  • Optional field. You can use this to add extra instructions to the assistant's task.

Fields in the OpenAI response:

  • Response message ID
  • Response message
  • Assistant
  • Message to assistant
  • Thread ID
  • Override instructions
  • Additional instructions
  • Run ID
  • Usage - Prompt tokens
  • Usage - Completion tokens
  • Usage - Total tokens
  • Error text

Telegram and OpenAI assistant integration

Go to the Apps section in your Albato account. Click on the Add a connection button. Choose Telegram and OpenAI in the list, and connect the apps to Albato.

OpnAI (5).png

Create your bot by integrating the apps you need.

OpnAI (6).png

Use triggers in the form of incoming Telegram messages that will be sent to your bot. The next step is to send the received message to OpenAI. After receiving the response, return it to Telegram as well.

OpnAI (7).png

In the OpenAI assistant settings, you can choose: assistant, message, Thread ID, Override Instructions, and Additional Instructions.

OpnAI (8).png

Send the AI text back to the Telegram bot.

OpnAI (9).png

The automations are ready! Let's test the integration.

The Automation Log shows all the data Albato has received and sent.

OpnAI (10).png

You can customize automations with Albato tools.

Watch our video tutorial, to see the construction of this use case and discover how to elevate your integration while maintaining the dialog context through the thread_id transfer!