Google Dialogflow ES (Essentials)
Create Google Cloud Project and API Credentials
See Dialogflow Documentation on how to create the project and the API credentials. In the end, you download a Service Account Credentials JSON File holding the private key.
Botium Box requires API Credentials with Dialogflow API Client role (for dialogflow.sessions.detectIntent-permission) as well as Dialogflow API Reader role (for dialogflow.agents.export/get/list-permissions).
For using the Botium conversation model uploader, additional permissions (dialogflow.agents.restore) are required. See here for an overview of the available roles - the Dialogflow API Administrator role has all permissions and can be used for this purpose.
Upload Private Key to Botium
In Botium Box, select Google Dialogflow ES as Connector/Chatbot Technology, and upload the Service Account Credentials JSON File.
Quotas and Limits
Dialogflow enforces quotas on the number of API calls. You can read more about it here.
Each API call - every step of every conversation from Botium - will count against these quotas. In the Retry & Quotas section of the Chatbot in Botium Box you can set the upper limit of user inputs (= API Calls) that are sent by minute.
When running tests in parallel, the API call counter on Dialogflow side is increased for each individual test session job and the quotas are reached faster. You should either run the tests sequentially or apply another rate limit to have less calls in total when running in parallel.