Built-In Integrations

• What Are Kuali's Built-In Integrations?
• Chat GPT Integrations
• Claude Integrations
• Microsoft Teams Integrations
• Slack Integrations

What Are Kuali's Built-In Integrations?

You can connect Kuali to other systems faster with Built-In Integrations — preconfigured, ready-to-use integrations available right in your system settings.  System administrators can configure popular integrations “out of the box” with minimal setup — and share them across apps for consistent, secure use.

System administrators will notice an Integrations section under Spaces & Settings → Integrations (System Settings if you are not an enterprise customer) - this is below the Custom Integrations option. This is where you’ll find the growing list of built-in integrations that utilize commonly used AI models or institutional systems, including:

• ChatGPT
• Claude
• Microsoft Teams
• Slack

These built-in integrations will display for System Administrators in the Build/System space which is the default space for the Kuali system and once integrations are configured can be used across all products in Kuali.

Note: App administrators will see these integration options in the form builder only after they have been configured and made available for use - see below how to configure and share.

Editing a Built-In Integration

Kuali’s built-in integrations use a bring-your-own-key model, so you can easily connect and make use of the accounts you already have.  When a System Admin initially edit the built-in integration via the pencil icon you'll see the below disclaimer to enter the API key or token for that service before you can configure. Most of these integrations require an API Key for the service you're setting up. For example, to use one of the Chat GPT integrations you'll need to add a valid OpenAI API key, including the word "Bearer" and a space before the API key.

Note: Microsoft Teams does not require a key so there will be no place to enter one in this screen - simply share with the desired apps where you want to utilize and then configure in the form.

Once you've established the key (if needed) it will open the integration where you can update the API key via Details, share the integration with forms via Data Sharing, and see which forms are actively using the integration via Apps. so you can configure as needed - below we outline how each type of built-in integration can be configured.

Once the key is configured you can then share the integration to the desired apps so app administrators can utilize and add to their forms (see below sections for more detail on how to configure in the product and in the Kuali form):

ChatGPT Integrations

One of the AI models Kuali supports in built-in integrations is ChatGPT (via OpenAI). 

Chose a ChatGPT Integration

There are four different Chat GPT Integrations:

Chat GPT Pro

This is the most capable, general purpose Chat GPT AI model. It's best for complex form input processing and document analysis. It's the default choice when quality matters. This is a good integration to start with.

Chat GPT Reasoning

This is the deep reasoning Chat GPT AI model for complex decisions. Use it for multi-step logic, policy compliance evaluation, and high-stakes scenarios requiring audit trails.

Chat GPT Turbo

This is the fastest, most cost-effective Chat GPT AI model. Best for low complexity tasks that require the fastest response possible.

Chat GPT Custom

The custom integration for advanced users who have a deeper understanding of how AI models work. It provides additional parameters that can be used to change the behavior of the AI.

Integrate Chat GPT into a form

OpenAI is another advanced AI integration. It allows the form builder to set the model, input, instructions, max output tokens and reasoning effort.

Note that under the covers the integration is using the OpenAI Responses API. https://platform.openai.com/docs/api-reference/responses/create

Quick Start

1. Create a new app (make sure the app has access to the integration)
2. Add a Text gadget to the form
3. Add a Data Lookup (Single Item) gadget
4. Change the Data Source to Chat GPT Pro
5. Change the Input to come from the Text gadget 
6. Enter “tell me a joke” in the Text gadget
7. A joke should appear in the Data Lookup gadget.

Standard Configuration

The Pro, Reasoning and Turbo integrations expose 5 settings.

Input

The Input setting is the end-user request that is sent to the AI. It's typically connected to another gadget on the form.

Instructions

The Instructions setting allows you to give instructions. It's the system prompt that can be sent to the AI model. It allows you to tell the model how to handle the input and variables. This setting is optional.

Variable 1, Variable 2 and Variable 3

These settings allow you to include additional data from the form. This data is added to the instructions. You can reference them in the instructions. These are optional.

Custom Configuration

The Chat GPT Custom integration exposes three additional settings.

Model

This allows you to select the specific model you want to work with. Try setting the fixed value to gpt-5.  This is required. Here’s the list of available models:https://platform.openai.com/docs/models

Max Output Tokens

This controls the number of tokens used in the response. A smaller number limits how large the response will be. This is required for this integration. Must be greater or equal than 1. 1024 is the default.

Reasoning Effort

This setting controls how much effort the model should make. Supported values are minimal, low, medium, high. Reducing reasoning effort can result in faster responses and less tokens used.

Claude Integrations

The Claude integrations enable you to integration Anthropic's AI functionality into your forms.

Choose a Claude Integration

There are four different Claude integrations:

Claude Pro

This is the most capable, general purpose Claude AI model. It's best for complex form input processing and document analysis. The default choice when quality matters. This is a good integration to start with.

Claude Thinking

This is the deep reasoning Claude AI model for complex decisions. Use for multi-step logic, policy compliance evaluation, and high-stakes scenarios requiring audit trails.

Claude Turbo

This is the fastest, most cost-effective Claude AI model. It's best for low complexity tasks that require the fastest response possible.

Claude Custom

The custom integration for advanced users who have a deeper understanding of how AI models work. It provides additional parameters that can be used to change the behavior of the AI.

Integrate Claude into a form

Quick Start

1. Create a new app (make sure the app has access to the integration)
2. Add a Text gadget to the form
3. Add a Data Lookup (Single Item) gadget
4. Change the Data Source to Claude Pro
5. Change the Input to come from the Text gadget 
6. Enter “tell me a joke” in the Text gadget
7. A joke should appear in the Data Lookup gadget.

Standard Configuration

The Pro, Thinking and Turbo integrations expose 5 settings.

Input

The Input setting is the end-user request that is sent to the AI. It's typically connected to another gadget on the form.

Instructions

The Instructions setting allows you to give instructions. It's the system prompt that can be sent to the AI model. It allows you to tell the model how to handle the input and variables. This setting is optional.

Variable 1, Variable 2 and Variable 3

These settings allow you to include additional data from the form. You can reference this data in the instructions. These are optional.

Custom Configuration

The Claude Custom integration exposes three additional settings.

Model

Allows you select the specific AI model you want to work with. This is required. Here’s the list of available models: https://docs.claude.com/en/docs/about-claude/models/overview

Max Tokens

This controls the number of tokens used in the response. A smaller number limits how large the response will be. This is required for this integration. Must be greater or equal than 1. 1024 is the default.

Temperature

Temperature controls the randomness of the response. The valid range is 0.0 to 1.0. Use a temperature closer to 0.0 for analytical/multiple choice and closer to 1.0 for creative and generative tasks. Note that even with 0.0 the results are not fully deterministic.

Quick Claude Form Integration

1. Create a new app
2. Add a Data Lookup (Single Item) gadget
3. Change to Data Source to Claude (it’s under the Kuali Data section)
4. Set Model to fixed value: claude-sonnet-4-5-20250929
5. Add a text gadget to the form
6. Set Input to link to the text gadget on the form
7. Set Max Tokens to 1024
8. Enter “tell me a joke” in the text gadget
9. A joke should appear in the Data Lookup Gadget.

Several settings are available to control the behavior of the Claude integration.

Model

This allows you to select the specific model you want to work with. Try setting the fixed value to claude-sonnet-4-5-20250929  This is required. Here’s the list of available models: https://docs.claude.com/en/docs/about-claude/models/overview

Input

This is the user prompt that is sent to the AI model. Note how it can be connected to a field on a form. Try adding a text gadget and connecting input to that gadget. This is required.

Max Tokens

This controls the number of tokens used in the response. A smaller number limits how large the response will be. This is required for this integration. Must be greater or equal than 1. 1024 is the default.

Instructions

This is the system prompt that can be sent to the AI model. It allows you to tell the model how to handle the input and variables (see below).

Temperature

Temperature controls the randomness of the response. The valid range is 0.0 to 1.0. Use a temperature closer to 0.0 for analytical/multiple choice and closer to 1.0 for creative and generative tasks. Note that even with 0.0 the results are not fully deterministic.

Variables 1-5

Five variables that are made available. A variable can be connected to other gadgets on the form. This allows you to submit additional form data as part of the AI request.

Microsoft Teams Integrations

The Microsoft Teams integration can be used in a workflow to send a message to a Microsoft Teams channel.  The Teams integration doesn’t require an API Key (it just needs to be shared with apps), however, Teams requires setup on the Teams side. 

Setup in Teams

To use this integration you'll need to configure a Teams channel to expose a Webhook URL that can be used to send messages to the channel.

1. Go to your Teams channel:
2. Click the three dots (...) → "Workflows" (not "Connectors")
3. Search for “Send “webhook alerts to a channel”
4. Click on it and set it up:
5. Give your workflow a name
6. Select the team and channel
7. Click "Add workflow"
8. Once created, you'll get a webhook URL that looks like:

https://prod-XX.region.logic.azure.com/workflows/XXXXXXXXX/triggers/manual/paths/invoke?api-version=2016-06-01&sp=%2Ftriggers%2Fmanual%2Frun&sv=1.0&sig=XXXXXXXXX

Setup in Kuali Workflow

To use this integration do the following:

1. Add an Integration step to the workflow
2. Choose Microsoft Teams as the integration
3. Set the HTTP Post URL to the Webhook URL you got from Microsoft Teams
4. Set the Message to what you want sent to the channel
5. You can use the three variables to include additional data from the form in the message. This data will be appended to the end of the Message.

 

Slack Integrations

The Slack integration allows you to send messages to a slack channel from a Kuali workflow. Below outlines the steps required on the Slack side and how to add as an integration step in a Kuali form workflow:

Setup in Slack

Step 1: Create a Slack App

• Go to https://api.slack.com/apps
• Click "Create New App"
• Choose "From scratch"
• Give your app a name and select the workspace where you want to develop it
• Click "Create App"

Step 2: Configure OAuth Scopes

• To post messages, your app needs the correct OAuth scopes Send or schedule a message | Slack:
• In your app settings, go to "OAuth & Permissions" in the sidebar
• Scroll down to "Scopes"
• Under "Bot Token Scopes", add these permissions:
  • chat:write - This scope lets your app post messages Send or schedule a message | Slack
  • channels:read - This scope lets your app retrieve a list of public channels Send or schedule a message | Slack (optional, but useful)

Step 3: Install App to Workspace

• Still in "OAuth & Permissions", scroll to the top
• Click "Install to Workspace"
• Review the permissions and click "Allow"
• You'll be redirected back to the OAuth & Permissions page

Step 4: Get Your Bot Token

• After installation, you'll see a "Bot User OAuth Token" that starts with xoxb-. Copy this token - you'll need it for API calls.

Step 5: Add Your Bot to the Channel

• The most common cause is that your bot isn't a member of the channel:
  • For Public Channels:
    • Go to the channel in Slack (e.g., #general)
    • Type /invite @YourBotName (replace with your actual bot name)
    • Or click the channel name → "Integrations" → "Add apps" → Add your bot
  • For Private Channels:
    • You MUST manually invite the bot - it can't post to private channels without being a member

Setup in Kuali Workflow

To use this integration do the following:

1. Add an Integration step to the workflow
2. Choose Slack as the integration
3. Set the Channel or Channel ID to the info you got from your Slack setup
4. Set the Message to what you want sent to the channel
5. You can use the three variables to include additional data from the form in the message. This data will be appended to the end of the Message.