Rasa App

Rasa is a framework for developing AI-powered, industrial-grade chatbots. It is an open-source alternative to the Dialogflow app. With the Rasa Chatbot app, you can set up and train your chatbot to respond to any query it's been trained for and configure it to act as an Omnichannel agent.

You must have the Omnichannel feature enabled and have agents and managers assigned to receive and send Omnichannel messages.

Rasa app operation modes

The Rasa app operates in two modes: Synchronous and Asynchronous modes.

Synchronous mode

The app uses Rasa's REST API to exchange messages in synchronous mode. To enable this mode, configure the Rasa bot's credentials.yml file and enable the REST API.

Asynchronous/callback mode

In asynchronous mode, the app utilizes callbacks to receive messages from Rasa. It is helpful if your app uses features like Reminders and External events, as they only work in this mode. To configure Rasa for this mode, you need to include the following configuration in the credentials.yml file of your Rasa server:

callback:
  url: "<your-callback-url"

To get your app callback URL, navigate to the Rasa App Info > Details > APIs on your workspace.

Install Rasa app

To install the Rasa App,

1. Go to Administration > Apps > Marketplace.

2. Search for the Rasa app.

3. Click Install and accept the needed permissions.

To set up Rasa server, visit the Rasa installation guide.

Configure Rasa app

1. Go to Administration > Workspace > Users and create a new user with the following roles:

   1. bot

   2. livechat-agent

2. Go to Administration > Workspace > Settings > Omnichannel > Routing and enable Assign new conversations to bot agent.

3. Open the installed Rasa app. On the Rasa App Info screen, go to Settings.

4. Update the required fields:

   • Bot Username: Set the username of the bot you created above. This user will assist with the conversations through Rasa integration.

   • Rasa Server URL: The URL of your Rasa server (for example, http://localhost:5005).

   • Service Unavailable Message: A message sent to your customer if the service is unavailable.

   • Close Chat Message: A message is sent to the customer when a conversation is closed.

   • Handover Message: A message sent to a customer upon handover.

   • Default Handover Department Name: The target department name where you want to transfer the customer upon handover.

   • Enable Callbacks: Enable callbacks to allow the app to use only callback messages. It is useful when you are using reminder messages in your RASA bot.

   • Hide Quick Replies: If enabled, all quick replies will hide when customers click on any of them.

5. Click Save Changes.

To see the bot in action, go to the workspace’s Livechat and send one of the prompts you’ve trained the bot to recognize. The Rasa bot will reply with the appropriate response.

Add quick replies support to your Rasa bot

The Rasa app provides out-of-the-box support for quick replies to improve customer conversations. To add quick replies, you can follow the structure of the rich responses defined in Rasa.

Uninstall Rasa app

To delete the Rasa app from your workspace,

1. Go to Administration > Apps > Installed.

2. Click the kebab menu against the Rasa app.

3. Select Uninstall.

4. Click Yes to confirm the uninstallation.