Response variations via Node Red...

[Swiftnesses]

Hi,

I've just started to implement Watson into our workflow. I've noticed that if the response type is sernt to sequential, the assistant node always returns the same response (unlike the test chat online), how can we avoid this?

[chughts]

Please provide an example for clarification. Sequential is an option on the Conversation dialog editor, so don't know how it would relate to node-red. What happens if you clear out the context and session on the online test chat?