From Email:
I’ve added a new sample in the repository with a botium.json file configured without redis. According to the example a local proxy server will be started in the background on 4002 port with /chatlayer endpoint: http://localhost:4002/chatlayer. This url you have to make reachable from outside using e.g. ngrok and have to copy this webhook url on chatlayer surface into the webhook channel settings dialog. My ngrok url for this example looks like this: https://679c3a9ca91a.ngrok.io/chatlayer.
Be careful because we developed a new version of chatlayer connector now and it’s on master branch already. The configuration is a little bit changed: You don’t need the channel id anymore, but you have to copy the whole API URL from the webhook channel settings dialog into the CHATLAYER_URL. Optionally you can configure the CHATLAYER_BOT_ID as well, to get more detailed nlp data. Further more in the CHATLAYER_VERSION capablility you can choose between ‘DRAFT’ and ‘LIVE’ and you can add the language of the bot in the CHATLAYER_LANGUAGE capability as well.
Thank you Szabi! I was able to run it this way 
Another question if I may: is there a way to execute botium via code, using custom parameters received from an api, for example? Instead of using botium.json file and botium configuration in package.json
Sure, if you know how to code in Node.js
Right. I know how to code in Node.js, but I can’t figure it out how I can pass the botium.json and the botium configuration in package.json via code. I’m using botium-bindings. Should I use something else? My goal is to make a simple service that receives the configuration and tests and executes them
hm … there is an old feature called “Botium Grid”, this might help you. It can launch Botium with a Swagger API, and you can inject botium configuration and test case scripts with HTTP/JSON.
https://chatbotsmagazine.com/how-i-used-tosca-to-test-an-alexa-skill-2fefd59f72c1