As far as I understood the manual, if I want to use headless browsers I’m running 3 containers, one per browser and one for the agent. Thats ok, what I do not like so much is that I have to bind it to a defined job by the id of that job.
What I need is a configuration where the agent can be called for every test offering to run it on a browser and not browserstack. Can I use the above mentioned setup too?
Or do I have to install both headless browsers in the container where the agent runs?