• MCP3008
• LM35
• Raspberry pi 3B
• LCD Display 16x2

First we need to open a outside route for our raspberry, so we can access it from the internet. Here we used Remote.it to open the path. But you can use ngrok or similar.

We run a flask server to run the script to get the temperature value from the LM35 sensor while doing the proper convertions.

On the DialogFlow dashboard, we create the intent and user phrases so our Google Assistant Bot run the query to our webserver via a webhook

Then we format the response, so DialogFlow can parse it and display to the user, over text and TTS