# My ChatGPT

## **Introduction**

The **My GPT integration** allows Autom Mate users to connect their **custom-trained GPT models** into their automation workflows seamlessly. This integration enables users to leverage **specialized AI bots** tailored to their specific needs without requiring any additional coding. **My GPT** can also integrate with external platforms such as **Microsoft Teams** or other supported messaging tools, enhancing user experiences with AI-driven interactions.

***

## **What is My GPT?**

My GPT integrates with **OpenAI’s GPT models**, allowing users to customize and deploy AI bots directly within Autom Mate. By connecting to the OpenAI platform, users can create and manage conversations using their GPT models and **automate various interactions**.

{% hint style="info" %}

### **Key Features**

* **Custom GPT Integration**: Use custom GPT models trained on the OpenAI platform.
* **Flexible Thread Management**: Support for both **static** and **dynamic conversation threads** using Thread IDs.
* **Multi-platform Integration**: Autom Mate integrates custom GPTs with external platforms like **MS Teams**.
* **Dynamic Message Handling**: Automate AI-driven responses based on user inputs.
  {% endhint %}

***

#### **Autom Mate Integration with My GPT**

Autom Mate seamlessly integrates with custom GPT models, allowing users to leverage their own AI creations in automation workflows. With **My GPT**, users can train and use their GPT models to interact with Autom Mate actions, facilitating **intelligent conversations and responses** within their processes.

{% hint style="info" %}

### **Key Integration Features**

* **Thread Management**: Use **Custom Thread ID** for static conversations or **Generate Thread ID** for dynamic conversations.
* **Message Handling**: Automate message sending and response handling with GPT.
* **Session Continuity**: Maintain **consistent conversations** across sessions using **Thread IDs**.
* **AI Integration**: Utilize **AI bots** in processes without writing any code.
  {% endhint %}

{% hint style="success" %}

### **Benefits of Autom Mate's My GPT Integration**

* **No Additional Coding**: Easily create and integrate AI bots into workflows without programming skills.
* **Customizable AI Conversations**: Tailor AI interactions based on specific business needs.
* **Time Savings**: Automate repetitive tasks or queries with AI-driven responses.
* **Seamless Workflow Integration**: Integrate AI into any Autom Mate process for **dynamic and intelligent automation**.
  {% endhint %}

{% hint style="danger" %}

### Attention

It is essential to properly configure the **Thread Scope** when working with GPT models in Autom Mate. The **Thread Scope** ensures session management remains intact, whether you are using a **static Custom Thread ID** for ongoing sessions or a **dynamic Generate Thread ID** for temporary interactions. Mismanaging the Thread ID can cause a loss of conversation context, leading to irrelevant or incomplete responses from the GPT model. Additionally, proper configuration of the **Assistant ID** and credentials in the [**Vault** ](/userguide/product-features/management/vault.md)is crucial. Without correct **API key setup**, Autom Mate will not be able to communicate with the GPT model. Moreover, users must handle the conversation thread carefully—deleting or resetting conversations may cause the **Thread ID** to change, breaking the flow continuity.

Incorrect usage of these elements may result in the automation failing or producing inaccurate responses. Ensure that you follow the correct action sequence—**Thread Scope**, **Message**, and **Run**—to avoid issues in workflow execution. Misplacing these actions, especially running a flow without a valid **Assistant ID** or **Thread ID**, will disrupt the GPT interaction and cause errors.
{% endhint %}

## **Credentials Setup for My GPT**

Autom Mate provides secure management of credentials to connect GPT models through **API keys**. Follow these steps to set up credentials for My GPT:

#### **Step-by-Step Guide**

{% stepper %}
{% step %}

#### Go to the [**Vault** ](/userguide/product-features/management/vault.md)section under [**Management**](/userguide/product-features/management.md).

{% endstep %}

{% step %}

#### Navigate to the [**Application Credentials**](/userguide/product-features/management/vault.md#app-credential-management) section.

{% endstep %}

{% step %}

#### Click on the **"New App Credentials"** button located at the top-right.

{% endstep %}

{% step %}

#### From the list of applications, select **GPT**.

{% endstep %}

{% step %}

#### Define a **Connector Name** for the GPT integration.

{% endstep %}

{% step %}

#### Under **Authentication Type**, select **API-KEY**.

{% endstep %}

{% step %}

#### Enter your **OpenAI API Key** in the designated field.

{% endstep %}

{% step %}

#### Click **Connect and Create** to complete the process.

{% endstep %}
{% endstepper %}

***

## **How to Use My GPT Actions**

<figure><img src="/files/wKpQNg7yOa7oBqnRlKvN" alt=""><figcaption></figcaption></figure>

Autom Mate provides **three main actions** for integrating custom GPT models into automation workflows: **Thread Scope**, **Message**, and **Run**. Each action plays a crucial role in managing and interacting with GPT-based conversations.

***

### **Thread Scope**

<figure><img src="/files/n8QtGEn4m3HXfadPJoae" alt=""><figcaption></figcaption></figure>

**Description and Purpose**

The **Thread Scope** action manages conversation threads in Autom Mate, ensuring that GPT conversations are either **static or dynamic** across multiple interactions. It helps Autom Mate identify and maintain the same conversation session, whether the thread is predefined or generated dynamically.

#### **Usage**

{% stepper %}
{% step %}
**Drag and drop the Thread Scope action** into the Autom Mate flow designer.

{% endstep %}

{% step %}
**Select the Thread ID** option:

* **Custom Thread ID**: Use a predefined, **static thread ID** from the GPT logs for persistent sessions.
* **Generate Thread ID**: Automatically generate a **new thread ID** for dynamic sessions, where the conversation resets after each session.
  {% endstep %}

{% step %}
Define whether the session should remain **static or dynamic**, depending on the business needs.

{% endstep %}
{% endstepper %}

{% hint style="success" %}

#### **Inputs**

**Thread ID**: Either select **Custom Thread ID** from the OpenAI GPT platform or let Autom Mate generate one dynamically.
{% endhint %}

{% hint style="danger" %}

#### **Action Output**

A **maintained thread** for either static or dynamic GPT conversations. This thread ensures that messages in the same scope are part of the **same conversation**.

#### Sample Output

`thread_i8uryPGXOYu9VlTjZPBWTIC9`
{% endhint %}

***

### **Message**

<figure><img src="/files/nsD2p4jrUfVYuuG0Xqje" alt=""><figcaption></figcaption></figure>

#### **Description and Purpose**

The **Message** action sends a **user-defined input message** to the custom GPT model. This action is crucial for generating GPT responses based on the conversation thread.

#### **Usage**

{% stepper %}
{% step %}

#### **Drag and drop the Message action** into the Autom Mate flow after the **Thread Scope** action.

{% endstep %}

{% step %}

#### **Define the message content**: Enter the text or question that you want to send to the custom GPT model for processing.

{% endstep %}

{% step %}

#### Ensure that the **Thread Scope** action is set up correctly to ensure the conversation remains within the intended context.

{% endstep %}
{% endstepper %}

{% hint style="success" %}

#### **Inputs**

* **Message Content**: The **text input** or question that will be sent to the GPT model for generating a response.
  {% endhint %}

{% hint style="danger" %}

#### **Action Output**

The message is sent to the GPT model based on the thread defined in the **Thread Scope**. The output will depend on the next action, which handles the response from the GPT model.
{% endhint %}

***

### **Run**

<figure><img src="/files/kAe0XqFhD57svamTjsk0" alt=""><figcaption></figcaption></figure>

#### **Description and Purpose**

The **Run** action executes the conversation with the custom GPT model by sending the defined message and receiving the **AI-generated response**. This is where the input from the **Message** action is processed by GPT, and the response is returned for use in further actions.

#### **Usage**

{% stepper %}
{% step %}
**Drag and drop the Run action** into the Autom Mate flow.

{% endstep %}

{% step %}
**Enter the Assistant ID**: Copy the **Assistant ID** from the OpenAI platform and paste it into the **Run** action.

{% endstep %}

{% step %}
**Define the response handling**: The response from the GPT can be saved as a **string variable** for use in subsequent actions.

{% endstep %}

{% step %}
Execute the flow to **send the message**, process it through GPT, and retrieve the response.

{% endstep %}
{% endstepper %}

{% hint style="success" %}

#### **Inputs**

* **Assistant ID**: The ID for your **custom GPT model** obtained from the OpenAI playground.
* **Thread ID**: Use the **Thread ID** defined in the **Thread Scope** action to maintain session continuity.
* **Message**: The input message that the GPT model will process, provided by the **Message** action.
  {% endhint %}

{% hint style="danger" %}

#### **Action Output**

The **Run** action returns the **GPT model's response** as a string variable. This response can be further utilized in Autom Mate workflows, such as storing the result, triggering additional actions, or integrating with external platforms.

#### Sample Output

`The Excel Read File Action in Autom Mate is utilized to retrieve data from an Excel file, facilitating various tasks such as data analysis, reporting, or processing within workflows. ### Key Features: -`` `**`Purpose`**`: To extract data conveniently from an Excel file for further use. ### Steps to Use: 1.`` `**`Specify the File Path`**`: Provide the location of the Excel file that you want to read. 2.`` `**`Open Mode`**`: Choose between "Read only" or "Read Write." 3.`` `**`Select Data Format and Sheet`**`: Define the data format and specify which sheet to read from. ### Inputs: - A variable representing the data table from the Excel sheet is necessary for executing the action【26:14†source】. If you have more questions or need additional details, feel free to let me know!`
{% endhint %}

***

## Additional Tips

How can I manage conversation threads effectively when integrating custom GPT models into Autom Mate workflows?

<table data-card-size="large" data-view="cards" data-full-width="false"><thead><tr><th align="center"></th><th></th><th></th><th></th></tr></thead><tbody><tr><td align="center">Use Cases</td><td></td><td></td><td></td></tr><tr><td align="center"><strong>FAQ</strong></td><td><a href="/pages/Gb52FoaUWMi1KLlBAHKn#how-can-i-manage-conversation-threads-effectively-when-integrating-custom-gpt-models-into-autom-mate">How can I manage conversation threads effectively when integrating custom GPT models into Autom Mate workflows?</a></td><td><a href="/pages/Gb52FoaUWMi1KLlBAHKn#what-steps-are-necessary-to-set-up-credentials-for-integrating-custom-gpt-models-with-autom-mate">What steps are necessary to set up credentials for integrating custom GPT models with Autom Mate</a></td><td><a href="/pages/Gb52FoaUWMi1KLlBAHKn#how-can-i-automate-ai-driven-responses-in-autom-mate-using-custom-gpt-models">How can I automate AI-driven responses in Autom Mate using custom GPT models?</a></td></tr></tbody></table>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.autommate.com/userguide/libraries/integration/my-chatgpt.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.