are you getting lots of questions from your customers related to covered 19 in this video I will walk through the steps to help you build and implement a rapid-response virtual agent as the world responds to the code 19 crisis providing your customers with accurate timely information is crucial
to keeping people safe and informed you can use the template in this tutorial to set up a custom rapid response voice and chat virtual agent that integrates directly into your web hone social and message channels you can launch it in 23 different languages and it's available for any
organization including healthcare and public services government agencies and nonprofits as well as other industries that are severely impacted including travel financial services and retail let's take a look at how you can get started using the template your first step is to download the dialogue flow cup in 1918
template from github link in the description to create the agent navigate to dialogue flow to cloud to Google com log in and click create new agent give your agent a name and then click create now click on the gear icon on the right hand side of the new
agent under export and import select import from zip and drop the zip file that you just downloaded from github now you're ready to import and setup the backend for your agent and for that you deploy fullfillment go back to the github and download the dialogue flow covert 19
for ferment back in the dialog flow console click fulfillment in the left sidebar and toggle the switch to enable the in line editor you'll be asked to enable a billing account in Google Cloud console when you're ready navigate to cloud functions from the left panel under source code
select the zip upload and upload the zip file that you got from github now you should see three intents closure confirmed cases and debts that require setting up Google Maps API and bigquery API for metrics let's enable the bakery API first select the api's and services from the
left panel then click on enable api's and services search for bigquery API and enable it for the Maps API you will need to create credentials to do that click on credentials create credentials and then API key make sure to restrict the key to cloud functions API only now
back in our cloud function we need to set this API key as an environment variable for that edit the cloud function click on the environment variables and add the variable the name should be Google underscore Maps underscore API underscore key and then the value should be the API
key you just created and then just deploy now you're ready to test the easiest way to do that is to test in dialogue flow simulator let's try an example is it still safe to receive packages the response should indicate that you setup your agent correctly if you plan
on enabling the bot on your website go to the integration section on the Left panel and enable dialogue flow messenger copy this code snippet and add it to your website you can also use the try it now button to test it out let's say I select start screaming
the agent starts asking me for specific symptoms and directs me based on my answers if you want to enable this experience we are holophone go to the integrations select dialogue flow phone gateway and a phone number place a call to the number and test it out if you
want to interact with the dialog flow API you will need to create a service account and download the private key here's how you would do that go to the I am an admin on the Left panel and select service accounts create a new service account give your new
account a name the role would be project owner and then click create create key will let you download the JSON key on your computer you will use this file in your environment so save it open the terminal on your local machine or any other interface where you can
run a coal command execute export Google application credentials which would be equal to the location of your key file in quotations so make sure you provide the right location to test we will lead a user request let's say the user asks how can I get tested we put
that in a file called request our JSON and next we will send this request to dialogue flow detect intent API using the coal command with post request option authorization header that gets the token value via G cloud art command and then the data gets passed as request dot
JSON and there you have a response in JSON format all that's left is to parse it out and do what you need to do with it in your custom application curl is just an example you can interact with the dialogue so API using other language integrations like Python
Java etc refer to the documentation linked below for language of your choice and that's it we hope you and your customers will find the rapid response virtual agent helpful during this unprecedented time you
