How to request the Chatbot?

The moment is there! You're going to make the Trengo Chatbot ready for use 🤖🎉*.*

By going to https://app.trengo.eu/admin/chatbots, you can click on the green plus sign (+) to create your first Chatbot.

https://s3.eu-central-1.amazonaws.com/trengo/media/hc_m_YUnuywZK9L6LYN62ojm94xvHpYQm07.png

Start

When you create a Chatbot, you'll first be asked which language (Chatbot language) the Chatbot should understand. This is important, so the Chatbot can give good suggestions in the selected language.

Select channels

In addition, you're asked on which channels the Chatbot will work. Select the relevant channels such as Website chat, Facebook, Twitter and/or Telegram. Once you've requested and/or linked channels in Trengo, you can select the relevant channels here.

You don't have to worry that the Chatbot will be active immediately. You can first set up the Chatbot on your own and publish it at a later time. The Chatbot won't be active on the selected channels yet. In the meantime, you're able to fill it with information and extensively test it.

https://s3.eu-central-1.amazonaws.com/trengo/media/hc_m_zOjCqWu4CElIeDh8xODx9uQmQr9W66.png

Client access via Google

To connect the Chatbot to the Machine learning technique from Google, it's necessary to request the Chatbot client access from Google (Dialogflow). Follow the steps below to create the Chatbot:

Generating token and secret in Dialogflow:

  1. Go to http://dialogflow.cloud.google.com/ and login with a Google account

  2. Give Dialogflow access to the Google account and click on "Allow"

  3. Accept (if not agreed upon) the conditions by clicking "Accept"

  4. Click on "Create new agent" in the drop-down menu (left-top corner)

  5. Come up with a name for the Chatbot, select the desired languages, choose the right time zone and click on "Create" and wait for the Chatbot to be created

  6. Once it's been created, click on the gearwheel icon in the left menu next

  7. Click below "BETA FEATURES" on "Enable beta features and APIs" and click "Save".

  8. Next click on the link of the Project ID (see image).

    https://s3-us-west-2.amazonaws.com/secure.notion-static.com/dfbcd45d-002f-4078-a940-d51e3c708f9f/Screenshot_2020-09-04_at_12.21.18.png

  9. This link navigates to Google's "Cloud console". You may first be asked to agree to the Terms and Conditions to continue.

  10. Click in the console, in the menu on the left, on "IAM & Admin" and then choose "Service accounts".

  11. Create a new Service Account.

  12. Give the Service Account a name and a short description (for example, the name of the Chatbot) and proceed to step 2.

    https://s3.eu-central-1.amazonaws.com/trengo/media/hc_m_gMsNJNHCn3QzSEfC6vIaugnnVacWpL.png

  13. In step 2, provide the Service account with the roles Dialogflow API Admin and Dialogflow API Client (to be searched in the search window) and continue to the next step.

  14. It is not necessary to grant users access to this account. Skip this part and click on "Done".

  15. Click on the "three-dots menu" next to the created service account and select " create key".

  16. Indicate that the Keytype is "JSON" and click on Create.

  17. A file is now downloaded, open this and copy all of the content.

  18. Paste the code that is just copied within Trengo in the Google Cloud Key field.

  19. Click on Create Chatbot.

  20. Go back to the cloud console (click on the Google cloud console to go back to the homepage) and navigate in the menu on the left to "APIs & Services" -> "Dashboard".

  21. Click on "Enable APIs and Services".

  22. In the next screen, search for "Dialogflow API" and make sure it is "enabled".