How to combine multiple agents in a process (possibly with branching)

[martinwunderlich]

Hello,

I am wondering, if there is a simple way to define a process that consists of multiple agents whereby the output of one agent becomes the input of the next? Also, it would be interesting to have a way to do conditional branching based on the output results and merging of branches in the overall process.
Is this already possible with langchain at the moment? If not, is it considered as a feature to allow more complex use cases?
Thanks!

Cheers,
Martin