Conversation Agent With Express, Dialogflow, and MongoDB
A conversation agent using the MVC Architecture to simulate the working of a conversation agent used by a third party repair company.
Getting Started
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
Prerequisites
A working version of Node and Node Package Manager is a must. Also, a UNIX based OS is ideal. Please refer to this guideline for further details.
Run this command to see what version of npm you have.s
npm --version
The application has been tested to be stable with npm version 7.0.14.
Also, highly reccomended is ngrok. Also, also, we highly reccomend Insomnia or Postman for testing.
Installing
Download the repo like so
git clone https://github.com/anshumankmr/Conversation-Agent-using-Dialogflow-and-NodeJS
Change into the directory using
cd Conversation-Agent-using-Dialogflow-and-NodeJS
Run
npm install
Running the Bot
Run this command
npm start
Run this
ngrok http 3000
Copy the https link generated by ngrok into the Dialoglow Agent's fulfillment section. It should look like this:
https://{{a randomly generated string }}.ngrok.io
Check this to be sure the API is running:
curl --request GET \
--url https://{{the same randomly generated string as before }.ngrok.io/serverHealth \
--header 'Content-Type: application/json' \
--data '{
"model": "XR",
"brand": "iPhone",
"issue": "Battery life has reduced by a lot. I think you guys have created a software update to slow it down."
}'
We can also try this in Insomnia we have here. Either way you should see this:
Server Health Looks Okay. We are good to go.
TO BE SHARED: The agents file
Running the tests
Currently, I am going to add some tests using Mocha and Chai. So that's work in progress.
License
This project is licensed under the MIT License - see the LICENSE.md file for details