Skip to main content

Rasa Open Source

With over 25 million downloads, Rasa Open Source is the most popular open source framework for building chat and voice-based AI assistants.

Integrating Rasa Open Source with your Airy Core instance enables you to leverage its conversational AI capabilities on all your instance's channels. When Rasa Open Source is integrated with an Airy Core instance, Airy Core will forward the messages to the Rasa Open Source installation, which will respond to the Airy Core API according to its conversation flow.

Airy Core includes an internal Rasa connector that you can install and configure on your Airy Core instance. This page will guide through the installation and configuration process.

note

This guide outlines the steps to connect your Airy Core instance to Rasa Open Source through Airy Core's Rasa connector.

Check out our blog post if you are interested in building an integration from scratch between your Airy Core instance and Rasa X.

What you will learn in this guide
  • Installing Rasa Open Source and creating a Rasa project
  • Setting up your credentials on your Rasa project
  • How to connect Rasa to your Airy Core instance through Airy Core's Rasa connector
Prerequisites

Step 1: Install Rasa Open Source

The first step is to install Rasa Open Source: follow the instructions from the Rasa documentation. Make sure your environment is properly set up if the installation fails.

Once Rasa Open Source is successfully installed, create a Rasa project.

Step 2: Set up your credentials in your project

Connecting Rasa Open Source to Airy Core requires one configuration value: a Rest webhook URL, which is made of your instance's URL appended with /webhooks/rest/webhook (for example: https://123.eu.ngrok.io/webhooks/rest/webhook).

In the Rasa project you created in the previous step, add your Rest webhook URL in the credentials.yml file, below rest.

For example:

rest:
webhook_url: "https://123.eu.ngrok.io/webhooks/rest/webhook"

Congratulations! You are now ready to connect Rasa Open Source to your Airy Core instance 🎉


Connecting this component to your Airy Core instance can be done in 2 ways:

We cover both options in this section: keep reading for connection via API request, scroll down to the next section for connecting it via the Airy Control Center UI.

Step 2 (option 1): Installation and configuration via API request

First, you need to install the component on your instance: send an API request to the Components Install endpoint to install the Rasa Open Source on your Airy instance.

The request body should be:

{
"name": "rasa-connector"
}

Once the installation is successful (indicated by a 200 response status code), you can configure the component using the Components Update endpoint.

Configuring the Rasa connector on your Airy Core instance requires your instance's URL, without any trailing slashes at the end (Airy Core's Rasa connector automatically appends /webhooks/rest/webhook to the URL you provide).

{
"components": [
{
"name": "rasa-connector",
"enabled": true,
"data": {
//your instance's URL without trailing slashes
//(example: https://123.eu.ngrok.io)
"rasaWebhookUrl": "yourInstanceURL"
}
}
]
}

The request is considered successful if the component's name (in this case:rasa-connector) is returned in the response.

{
"components": {
"rasa-connector": true
}
}

The Rasa Open Source is now installed and configured on your Airy Core instance.

Step 2 (option 2): Installation and configuration via the Airy Control Center UI

On the Airy Control Center UI, navigate to the Catalog and select Rasa. Click on the Install button.

Once the installation is completed, navigate to the Control Center's Connectors page and select Rasa: this will open the Rasa connector configuration page.

Configuring the Rasa connector on your Airy Core instance requires your instance's URL, without any trailing slashes at the end (Airy Core's Rasa connector automatically appends /webhooks/rest/webhook to the URL you provide).

On the configuration page, paste your instance's URL in the respective input and save.

The Rasa Open Source is now installed and configured on your Airy Core instance.

Step 3: Rasa's connection with Airy

Start your rasa server from your Rasa project you created in step 1:

rasa run --enable-api

If this command fails, make sure you've set up your virtual environment properly and start again.

Once the server runs, write a message to one of your channels: Airy Core will forward it to your Rasa Open Source installation, which will respond to the Airy Core API according to its conversation flow.

The screenshot below displays an Airy Live Chat Plugin from an instance connected to Rasa Open Source through Airy Core's Rasa connector. It shows an example where a message sent to Airy Core is automatically responded to by a Rasa Open Source installation.

Airy connection with Rasa